Tdoc List

2017-01-23 17:03

Agenda Topic TDoc Title Source Type For Avail Treated Decision Wdrn Replaced-by Replaces
1 Opening of the meeting                      
2 Approval of the agenda R4‑1700001 Proposed agenda Chairman agenda Approval Yes
Yes
 		approved No    
3 5G Study items: new radio access technology [FS_NR_newRAT]                      
3.1 WP 5D [FS_NR_newRAT] R4‑1700291 Evening AH minutes for NR WP5D NTT DOCOMO, INC report Approval Yes
No
 		not treated No    
3.1.1 Co-existence [FS_NR_newRAT] R4‑1700205 TP for TR38.803: Coexistence simulation assumptions DOCOMO Communications Lab. pCR   Yes
Yes
 		approved No    
    R4‑1700206 TP for TR38.803: Antenna and beam forming pattern modelling for coexistence simulation assumptions . DOCOMO Communications Lab. pCR   Yes
YesNokia wanted clarification that modification to this tdoc was possible if changes were identified during the meeting. DOCOMO believed that no changes would be identified. NEC was happy with the content. It would be readdressed at the next meeting.
 		noted No    
3.1.1.1 Co-existence simulation results [FS_NR_newRAT] R4‑1700002 Consideration on dense urban BS noise figure for coexistence study at 30 GHz for WP5D on new radio access technology Nokia, Alcatel-Lucent Shanghai Bell other Approval Yes
YesNokia introduced the document. The ACIR 9dB figure was proposed, there being little difference between that and the results for 11dB. ZTE was not yet ready to accept this proposal. Ericsson highlighted that a unique value needed to be sent to ITU-R, so it was necessary to come to some decision. Huawei had found that either 9 or 11 dB would be satisfactory. There was further discussion off line.
 		noted No    
    R4‑1700003 Proposal on dense urban indoor UE ratio for coexistence study at 70 GHz for WP5D on new radio access technology Nokia, Alcatel-Lucent Shanghai Bell other Approval Yes
YesNokia introduced the document, and proposed 10% as a figure. He noted that most other companies had used 30 GHz results only, but Nokia thought a different response to ITU would be appropriate for different frequency ranges. The focus should be on both 30 GHz and 70 GHz, starting with the lower range. Further simulation was needed with 70 GHz. Samsung proposed to not to include 70 GHz. Huawei had agreed to modify the simulation results when they had been discovered to be irrealistic. But they wondered how and when they could provide new simulation results. Samsung proposed that there was already agreement on indoor 70 GHz. ZTE supported Nokia's position. DOCOMO proposed to reply to ITU on the basis of the existing assumptions. Intel agreed with Nokia. The reply to ITU should make it quite clear which scenario(s) were covered. The Chairman suggested that RAN4 could check this situation at the next (February) meeting. Huawei wished to revise the assumptions for the macro case, and wanted to revisit all assumptions. Nokia thought there were two ways forward: either RAN4 should answer straight away, but making it clear to ITU that 5% for 70 GHz was not included; or the reply could be delayed until February and be more inclusive. Ericsson insisted that a single value with no comments should be supplied from this meeting, and this should be the final value. Otherwise, this would spread confusion in ITU. Qualcomm preferred to spend more time during the present meeting analysing the existing data, and that a final response could be made at this meeting. No simulation assumptions needed to be revised.
 		noted No    
    R4‑1700014 NR coexistence results for urban macro scenario China Telecom discussion   Yes
Yes
 		noted No    
    R4‑1700019 NR coexistence study methodology and assumption Intel Corporation other Discussion Yes
Yes
 		noted No    
    R4‑1700020 Simulation results - Indoor Intel Corporation other Discussion Yes
Yes
 		noted No    
    R4‑1700021 Simulation results - Dense Urban Intel Corporation other Discussion Yes
Yes
 		noted No    
    R4‑1700022 Simulation results - Urban Macro Intel Corporation other Discussion Yes
Yes
 		noted No    
    R4‑1700042 Simulation results for NR coexistence study: indoor deployment at 30GHz, 45GHz and 70GHz Qualcomm Incorporated other Discussion Yes
Yes
 		noted No    
    R4‑1700043 Simulation results for NR coexistence study: urban macro deployment at 30GHz Qualcomm Incorporated other Discussion Yes
Yes
 		noted No    
    R4‑1700044 Simulation results for NR coexistence study: dense urban deployment at 30GHz, 45GHz and 70GHz Qualcomm Incorporated other Discussion Yes
Yes
 		noted No    
    R4‑1700085 Simulations Results on for Coexistence Studies in Urban Macro Scenario for NR in mmwave spectrum Ericsson discussion Discussion Yes
YesEricsson had considered the 2% throughput loss situation, and many earlier discussions had centered around this value. The chairman was not enthusiastic to include both 2% and 5%, and all other companies had given 5%. It was difficult to compare values. Qualcomm preferred the 5% figures. Nokia countered that the Ericsson proposal only considered 30 GHz. Tdocs 86 and 87 proposed different values. The Chairman concluded that Ericsson would need to share their 5% results.
 		noted No    
    R4‑1700086 Simulations Results for Coexistence Studies in Dense Urban Micro Scenario for NR in mmwave spectrum Ericsson discussion Discussion Yes
Yes
 		noted No    
    R4‑1700087 Simulations Results for Coexistence Studies in Indoor hotspot Scenario for NR in mmwave spectrum Ericsson discussion Discussion Yes
Yes
 		noted No    
    R4‑1700102 DL simulation results for urban macro scenario in 30 GHz Huawei, Hisilicon discussion Discussion Yes
Yes
 		noted No    
    R4‑1700103 UL simulation results for urban macro scenario in 30 GHz Huawei, Hisilicon discussion Discussion Yes
Yes
 		noted No    
    R4‑1700105 Simulation results for dense urban scenario in 30 GHz Huawei, Hisilicon discussion Discussion Yes
Yes
 		noted No    
    R4‑1700106 Simulation results for dense urban scenario in 70 GHz Huawei, Hisilicon discussion Discussion Yes
Yes
 		noted No    
    R4‑1700107 Simulation results for indoor scenario in 30 GHz Huawei, Hisilicon discussion Discussion Yes
Yes
 		noted No    
    R4‑1700108 Simulation results for indoor scenario in 70 GHz Huawei, Hisilicon discussion Discussion Yes
Yes
 		noted No    
    R4‑1700109 TP for 38.803: simulation results for urban macro scenario Huawei, Hisilicon pCR Approval No
YesThe document was not provided.
 		withdrawn Yes    
    R4‑1700144 Simulation results for NR coexistence study in dense urban scenario ZTE Corporation discussion   Yes
Yes
 		noted No    
    R4‑1700145 Simulation results for NR coexistence study in indoor hotspot scenario ZTE Corporation discussion   Yes
Yes
 		noted No    
    R4‑1700146 Simulation results for NR coexistence study in Urban Macro scenario ZTE Corporation discussion   Yes
Yes
 		noted No    
    R4‑1700156 5G NR coexistence calibration results for all test scenarios LG Electronics Inc. discussion Discussion Yes
Yes
 		revised No R4‑1700245  
    R4‑1700245 Includes simulation results. LG Electronics Inc. discussion Discussion Yes
Yes
 		noted No   R4‑1700156
    R4‑1700157 5G NR coexistence evaluation results for Indoor scenario LG Electronics Inc. discussion Discussion Yes
YesThe Chairman declared that this contribution would be treated as a late submission since the actual date was not reflected.
 		revised No R4‑1700246  
    R4‑1700246 Includes simulation results. LG Electronics Inc. discussion Discussion Yes
Yes
 		noted No   R4‑1700157
    R4‑1700158 5G NR coexistence evaluation results for Dense Urban scenario LG Electronics Inc. discussion Discussion Yes
YesThe Chairman declared that this contribution would be treated as a late submission since the actual date was not reflected.
 		revised No R4‑1700247  
    R4‑1700247 Includes simulation results. LG Electronics Inc. discussion Discussion Yes
Yes
 		noted No   R4‑1700158
    R4‑1700159 5G NR coexistence evaluation results for Urban Macro scenario LG Electronics Inc. discussion Discussion Yes
YesThe Chairman declared that this contribution would be treated as a late submission since the actual date was not reflected.
 		revised No R4‑1700248  
    R4‑1700248 Includes simulation results. LG Electronics Inc. discussion Discussion Yes
Yes
 		noted No   R4‑1700159
    R4‑1700164 Simulation results for NR coexistence study in dense urban scenario Samsung electronics co., LTD discussion Discussion Yes
Yes
 		noted No    
    R4‑1700165 Simulation results for NR coexistence study in indoor scenario Samsung electronics co., LTD discussion Discussion Yes
Yes
 		noted No    
    R4‑1700166 Simulation results for NR coexistence study in urban macro scenario Samsung electronics co., LTD discussion Discussion Yes
YesSamsung introduced the document, and wished to unify the results across scenarios. Qualcomm agreed, and had used full correlation in their simulation results. True collocation would result in a correlation value of 1. If at 200m distance, correlation was 50%, this implied a value of 1 for collocation. He wondered if all companies had used the same simulation assumptions. Nokia had studied this in some detail, but the use of large-scale fading gave different results for indoor path loss because of line of sight considerations. Samsung had considered this, and considered that the path loss was only a function of distance. Ericsson agreed. Huawei believed that the same parameters (loss probability, …) could give good simulation results for the coordinated scenario.
 		noted No    
    R4‑1700171 TP for 38.803: simulation results for group of scenarios HuaWei Technologies Co., Ltd pCR   Yes
YesThe Chairman believed that preliminary discussion was necessary to determine how each result would be captured in the TR. Huawei wondered how to capture the results. The Chairman proposed to hold it over to the next meeting.
 		postponed No    
    R4‑1700183 Co-existence results of Urban macro CATT discussion Discussion Yes
Yes
 		noted No    
    R4‑1700184 Co-existence results of Dense urban CATT discussion Discussion Yes
Yes
 		noted No    
    R4‑1700185 Co-existence results of Indoor hotspot CATT discussion Discussion Yes
Yes
 		noted No    
    R4‑1700202 Downlink simulation results for NR coexistence study NEC discussion Discussion Yes
Yes
 		noted No    
    R4‑1700203 Uplink simulation results for NR coexistence study NEC discussion Discussion Yes
Yes
 		noted No    
    R4‑1700207 Summary on simulation results for WP5D co-existence study DOCOMO Communications Lab. other   Yes
YesDOCOMO had shared these results on the RAN4 reflector. The spreadsheet compared all companies values, to be used as a basis for off line discussions. LG had distributed revised simulation results on the reflector. The Chairman concluded that each company needed to check their own values. DOCOMO wanted to discuss the ACIR average value. There were some situations not yet covered and further discussions were needed on how to split ACIR and ACLR. The Chairman believed that a full consideration was necessary, ACLR and ACIR. The table should be updated off line.
 		noted No    
    R4‑1700280 Updated summary on simulation results for WP5D co-existence study NTT DOCOMO, INC. other discussion Yes
YesNTT DOCOMO had captured the results in a spreadsheet.
 		noted No    
3.1.1.2 Summary and proposed ACIR [FS_NR_newRAT] R4‑1700004 Summary of simulation results for coexistence study for WP5D on new radio access technology Nokia, Alcatel-Lucent Shanghai Bell other Discussion Yes
Yes
 		revised No R4‑1700252  
    R4‑1700252 Summary of simulation results for coexistence study for WP5D on new radio access technology Nokia, Alcatel-Lucent Shanghai Bell other Discussion Yes
Yes
 		noted No   R4‑1700004
    R4‑1700005 Proposed ACIR values for coexistence study for WP5D on new radio access technology Nokia, Alcatel-Lucent Shanghai Bell other Approval Yes
Yes
 		noted No    
    R4‑1700023 Summary and proposed ACIR Intel Corporation other Approval Yes
Yes
 		noted No    
    R4‑1700045 ACIR, ACLR and ACS proposals Qualcomm Incorporated other Approval Yes
YesQualcomm presented the document. RAN4 companies had done considerable work, and the mean throughput results were well aligned. The uplink and downlink numbers were aligned. In NR, equal split uplink/downlink was reasonable. Nokia was concerned about the 30 GHz vs 45 GHz relaxation. How much relaxation was proposed? Qualcomm responded that the concentration would be on 30 GHz, but for the higher frequencies, there was no macro case, so relaxation was appropriate. Huawei could not accept the lower ACLR value because of the different details of implementations. He believed that the Huawei platform produced the worst case values, and neither uplink or downlink values proposed by Qualcomm were realistic. Qualcomm responded that, given the assumptions used, they believed the results were appropirate, but understood that other companies felt other scenarios were better. And for this reason, Qualcomm had built in some margin in arriving at its worst case results, and would concentrate on average values. Huawei had considered these different companies' results, and wondered if using an average value was realistic. There was a very wide spread of results from different companies. Further, equal split between base station and UE was not a good approach: a base station needed to exhibit appreciably better performance. ZTE considered that the new waveform could potentially affect the figures, and at present some overhead for this implementation was too risky. Samsung to examine some details off line with a view to having a smaller range of results. LG wanted to set down some principles for ACLR value. Nokia thought agreement on a single principle would be very difficult. Probably the Qualcomm approach was simpler. The Chairman noted that, at present, there was no agreement on which value of ACS to use. Qualcomm sought to explain the different sets of results, considering different shadowing conditions. They believed that taking the average was indeed a viable approach, and wished to avoid imposing too onerous requirements. Vodafone understood why the results differed from each other, but was not at all sure that averaging them was not a good approach. For example, antennas could not be exactly collocated.
 		noted No    
    R4‑1700104 Layout parameter consideration for urban macro scenario Huawei, Hisilicon other Approval Yes
YesHuawei introduced the document, concluding that the ACIR requirement should be 27 dB for the worst case, and that using 5%-tile user throughput loss statistic to determine DL ACIR was justified in the urban macro scenario with 300m ISD. Urban macro BS had the capability of covering more than a 300m ISD. 200m ISD (less than 70m cell radius) was too small for an urban macro base station. In conclusion it was proposed to adopt an urban macro ACIR of 27 dB DL to DL, and to define the urban macro base station ACLR as 40 dB. xxx believed only around 28 dB ACIR could be achieved. Samsung also had some concerns over the document's findings.
 		noted No    
    R4‑1700110 Summary and analysis on ACIR values for NR coexistence study Huawei, Hisilicon other Approval Yes
Yes
 		noted No    
    R4‑1700152 Summary of proposed ACIR values for NR coexistence study ZTE Corporation discussion   Yes
Yes
 		noted No    
    R4‑1700167 Discussion on simulation results of NR coexistence study Samsung electronics co., LTD discussion Approval Yes
YesSamsung presented the document, highlighting several revisions of previous results. The standard deviation for different companies' results could be as large as 4 dB. It was therefore necessary to restrict the assumptions. Nokia and Ericsson supported proposal 4. Huawei did not think it appropriate to exclude scenarios because the simulation results were too deviant and could not accept the ACIR value for downlink (proposal 1). Samsung clarified that proposal 1 was for both UL and DL. Considering the time limit, it was reasonable to present these results to ITU-R. Nokia recalled that this way forward had already been discussed at the previous meeting. There was a difficulty in separating UL and DL. It had been agreed that 200m was the basic and 300m the optional ISD value. Vodafone questioned this assumption, and also the assumption that the traffic had to be symmetrical. Other scenarios also had to be considered. Qualcomm supported proposals 3 & 4. Their results were very similar to Samsung's.
 		noted No    
    R4‑1700186 Consideration on ACLR_ACS values for WP 5D CATT discussion Discussion Yes
YesThe Chairman proposed that the discussion focus on PA feasibility. The document was presented by CATT. LG and Intel believed that 28 dB was too strict for the ACLR. ZTE also had concerns.
 		noted No    
3.1.2 RF parameters [FS_NR_newRAT] R4‑1700163 Diccussion and proposal for WP 5D related parameters Samsung electronics co., LTD discussion Approval Yes
YesSamsung presented their proposals: Proposal 1: Taking (Compromised value) 10 dB for 30GHz, 12 dB for 45 GHz, 14 dB for 70 GHz as WP5D response. Proposal 2: Determining ACIR based on NR co-existence evaluation results under {11, 13, 15} dB @ { 30, 45, 70 } GHz NF assumptions. Proposal 3: ACLR/ACS values should be aligned with the result of co-existence study. Proposal 4: Spectrum emission mask should meet at least FCC regulation in mmWave bands ? -5 dBm/MHz @ 0 ~ 20 MHz from channel edge ? -13dBm/MHz @ 20 ~ 400MHz from channel edge Proposal 5: No need to respond to WP5D for Blocking response considering ACS and NF already included in response parameters. Telecom Italia supports the lower values in the range (9, 11, 13) for proposal 1. Ericsson favoured propsal 1. Intel believed different values were required for BS and UE. Huawei considered that the noise figure should be based on the architecture. Qualcomm preferred proposal 1. Vodafone recalled the discussions at the previous meeting. On balance, the UE figure should be 10 dB. Qualcomm observed that the numbers with filters were higher. Maybe filters would be needed to achieved the necessary spurious emissions figures, and this would increase the noise figure by one or two dB. Skyworks recalled that their contribution had looked at options where filters might be placed, and depending on the placement in the architecture might NOT influence the NF. Nokia foresaw that the connector and cable loss would eat into the BS's margin. Samsung reminded the meeting that a decision had to be reached at this meeting. They proposed either the same 10 dB value for both sides, or 10 dB for BS and one or two dB less for UE. The Chairman recalled that this had been under discussion for four or five meetings. Qualcomm thought that for the UE it was almost already agreed (at the previous meeting). Huawei hoped consensus could be reached offline during the week. Concerning the spectrum mask, Ericsson had a different view (tdoc 65). They proposed a tighter value in the second adjascent channel. Nokia also had a contribution on blocking. Huawei had a proposal for blocking based on ACS, but the current definition was based on a static interface. Which blocking was under consideration in this document?
 		noted No    
    R4‑1700073 Discussion of IMT parameters for response to ITU-R WP5D Ericsson discussion Discussion Yes
YesEricsson outlined the document, which was a collection of discussions. Some time would have to be spent on the spectrum masks and on the ACLR values. It was proposed not to submit the receiver sensitivity values.
 		noted No    
    R4‑1700074 IMT parameters for final response to ITU-R WP5D Ericsson other Approval Yes
YesThe Chairman proposed that no presentation of the document was necessary, but that the contents of the annex should be checked in relation to tdoc R4-1700170 (Samsung). Ericsson suggested that the important part was in the annex.
 		noted No    
3.1.2.1 ACLR/SEM/ACS [FS_NR_newRAT] R4‑1700076 NR unwanted emissions for BS and UE in ITU-R response Ericsson other Approval Yes
YesThe Chairman considered that even if the idea were accepted, the ACLR values would need to be revised according to the final values. Ericsson presented the document, which recalled agreements already achieved, and offered the following proposals: PROPOSAL 1: For the ITU-R response, the emissions in the out-of-band domain for BS and UE will be specified using a “transmission centric” Spectrum Emissions Mask (SEM), applicable for a 200 MHz channel bandwidth, extending out to 500 MHz from the centre of transmission and with a measurement bandwidth of 1 MHz. PROPOSAL 2: The emissions limits should have the new limits in FCC Title 47, §30.203 as a baseline. PROPOSAL 3: The BS emissions in the out-of-band domain should reflect also the expected ACLR performance by having a relative component corresponding to the ACLR in the two first adjacent channels. PROPOSAL 4: An absolute level should limit the spectrum mask (as well as ACLR) for lower BS output power levels. Proposal 1 and 2 were agreed. On proposal 3, Nokia had a simpler proposal. There was no need to adjust the mask according to BS output power. Huawei supported the Ericsson proposal 3. Ericsson agreed that the Nokia proposal had merits, and in the end there would only be two masks, one for high level, one for low.
 		noted No    
    R4‑1700117 Emission mask for mmWave bands Huawei, Hisilicon other Approval Yes
YesHuawei presented the document and offered the following proposals: Proposal 1 The SEM shall be independent from contiguous allocated spectrum larger or equal 200MHz. Proposal 2 The SEM can be adjusted accordingly for contiguous allocated spectrum less than 200MHz, specific boundary is FFS. Proposal 3 The SEM shall be independent from specific mmWave bands for 24.25~86GHz. Ericsson thought the scope of the document was a little wider than the response to ITU. Proposals 1 and 2 were not relevant under this agenda item. Proposal 3 needed further consideration, because there would be a need for different masks in different bands. Nokia agreed with Ericsson. ZTE believed further discussion was required on this. Huawei responded that the ACM, based on ACLR, should be the same for all bands. Further study was needed. It was clarified that this document only related to 200 MHz bandwidth.
 		noted No    
    R4‑1700118 TP for 38.803: Emission mask for mmWave bands Huawei, Hisilicon pCR Approval Yes
YesThe Chairman stated that this document would be treated only if time allowed. The Chairman proposed to hold this over to the next meeting.
 		noted No    
    R4‑1700101 Introduction of ITU-R WRC19 AI 1.13 IMT candidate frequency band and passive service co-existence scenarios Huawei, Hisilicon other Approval Yes
YesThe Chairman wondered if the intention of the contribution was to include this aspect into an LS to WP 5D, or constituted proposals for future work. Huawei explained the intention behind this document, which was mainly to provide supporting information and the background to the ITU position. But it offered the following proposals: Proposal 1: To consider using at least -20dBm/MHz spectral emission mask for below 24.25GHz bands. Proposal 2: To consider only using -30dBm/MHz spurious emission level for below 24.25GHz bands. It was understood that these two parameters should be condered in this meeting's response to ITU-R. Ericsson recalled that what was needed was a generic report to ITU, to help them in their subsequent studies. It was premature to anticipate what the ITU would eventually decide. Huawei recalled the importance of the two bands in consideration. The Chairman indicated that the way forward was with generic values at this stage.
 		noted No    
    R4‑1700098 Spectral Emission Mask for UE mmWave Qualcomm Incorporated other Approval Yes
YesQualcomm offered one proposal: Proposal : Spectrum Emission Mask for NR mmWave will be defined in the following way: The spectrum emission mask of the NR mm wave UE applies to frequencies (?fOOB) starting from the +/- edge of the assigned NR channel bandwidth.
 		noted No    
    R4‑1700082 On SEM in UE RF Intel Corporation other Discussion Yes
YesOriginal agenda item 3.5.4.
 		noted No    
    R4‑1700065 UE Spectrum Emissions Mask Ericsson other Approval Yes
YesEricsson proposed an absolute mask, noting that the second adjascent channel had lower limits, so they proposed to use a smaller delta of 5 dB. It was proposed that the total conductive power or the TRP measured in 1 MHz in the OOB domain should not exceed the limits given by the general spectrum emission mask in Figure 1 [of the contribution]. Qualcomm believed more discussion on the 5 dB delta was needed. Maybe no delta was needed at all. Ericsson responded that the important aspect was to get appropriate spectrum in the ITU, and it was best for the spectrum mask to reflect actual findings, bearing in mind linearity performance. The Chairman indicated that if no consensus on the delta could be achieved, the fallback was to use the FCC limit as is.
 		noted No    
    R4‑1700077 Spectrum emission mask for NR BS in ITU-R response Ericsson other Approval Yes
YesThe Chairman noted that even if the idea were accepted, the ACLR values would need to be revised. Nokia had a view on SEM in R4-1700006. Ericsson indicated that this contribution was a modified version of a document seen at a previous meeting. It applied to the base station performance. He recalled that the spectrum mask could be reduced by, say, 3 dB, and recalled the siimilar agreement for LTE. The document addressed three different frequency ranges. Nokia noted that the calculation of BS output power might not be easy because of complex antenna arrays. Huawei did not think it necessary to relax the limits and recalled the experience of LTE in this regard. Intel questioned the difference between the UE and BS limits. Ericsson indicated that there were two different approaches. In the BS, the mask was adapted to the power output, but for the UE it was fixed. The situations were not directly comparable. Qualcomm wondered whether the response to ITU would be a single value regardless of Tx power. Ericsson confirmed this: the ACLR would be identical in both high and low power cases. Huawei thought an absolute level of -20 dB needed an rationale. Ericsson responded that indeed there was no rationale explained in the paper, but this method was in line with existing specs. But it was agreed that a precise justification was not easy.
 		noted No    
    R4‑1700160 UE ACLR/ACS requirements for 5G New Radio LG Electronics Inc. discussion Approval Yes
YesLG briefly outlined the document and explained their proposals: Proposal 1: The ACLR/ACS levels […] are proposed for 5G NR UE with channel bandwidth of 200MHz according to the operating frequency ranges. Proposal 2: The proposed ACLR/ACS levels of 5G NR UE are considered as the reference levels at 30GHz/70GHz frequency ranges. To OTA ACLR/ACS requirements, RAN4 should be considered the test tolerance of the OTA test methodologies.
 		noted No    
    R4‑1700024 Proposal on ACLR and ACS Intel Corporation other Approval Yes
YesIntel briefly outlined this contribution.
 		noted No    
    R4‑1700081 On ACLR in mmWave UE RF Intel Corporation other Discussion Yes
YesOriginal agenda item 3.5.4. Intel presented the document and its proposals. Proposal 1: 22dBc ACLR is feasible for 28GHz frequency. Proposal 2: Alternative ACLR could be tighter than 22dBc Proposal 3: Study feasible ACLR levels for 45GHz and 70GHz Proposal 4: To define some key RF parameters including ACLR as feasible and reasonably achievable values, a maximum allowed transmitted power should be decided in EIRP or TRP. One possible starting point is choosing one of LTE power categories and defines it as EIRP or TRP (or both).
 		noted No    
    R4‑1700229 A feasible ACLR metric and operating point for mm-wave UE’s Qualcomm Incorporated discussion Discussion Yes
YesQualcomm presented the document and its analysis to conclude that the ACLR level should be -16 dBc. Skyworks wondered about the basis of this document. What techniques had been taken into consideration? Qualcomm outlined the background. Qualcomm indicated that the device was fabricated in CMOS, for ease of integration and cost reasons. They also outlined the output power and efficiency considerations. Battery capacity of UEs was an important consideration. The Chairman noted that there were now several achievable ACLR values on the table.
 		noted No    
    R4‑1700231 A feasible ACS blocker specification for mm-wave UE’s Qualcomm Incorporated discussion Discussion Yes
YesQualcomm believed a feasible UE ACS level was -23 dBc. Huawei had comments on the table some of the values in the table were a little surprising. It was agreed to discuss this off line.
 		noted No    
    R4‑1700066 TP for 38.803: UE ACLR Ericsson pCR Approval Yes
YesEricsson would wait for the discussion on ACLRs to be resolved before pursuing.
 		postponed No    
    R4‑1700006 Proposed BS ACLR/SEM/ACS values for coexistence study for WP5D on new radio access technology Nokia, Alcatel-Lucent Shanghai Bell other Approval Yes
YesNokia presented the document and offered: Proposal 1: To consider using different BS ACLR/ACS values in different deployment scenarios, or use 23 dB BS ACLR and 19 dB BS ACS if a single BS ACLR/ACS value is preferred for coexistence study for WP5D. Proposal 2: To adopt the FCC limits for the SEM for coexistence study for WP5D for BS with maximum transmit power higher than 23 dBm, and use the FCC limits minus 7 dB for the SEM for coexistence study for WP5D for BS with maximum transmit power lower than or equal to 23 dBm. Ericsson was concerned that this document anticipated what ITU-R might decide. What was a reasonable mask for a macro cell was not necessarily appropriate for the indoor case. Nokia indicated that they had looked into this, and had received indications that ITU-R would prefer a single value. Re-use considerations would mainly apply to the outdoor case. Huawei was concerned that the mask might differ by about 10 dB depending on how it was calculated. Intel indicated that the nonlinearity of the transmitter between high and low power might not justify a single mask.
 		noted No    
    R4‑1700155 On feasible BS ACLR level Sumitomo Elec. Industries, Ltd other   Yes
YesThe document made several observations and proposals: Observation 1 It is not feasible to consider DPD as prerequisite when studying the feasibility of BS ACLR Observation 2 Advanced power amplifier design to improve the PA power efficiency is needed for mmWave NR BS Observation 3 Using the advanced power amplifier design to improve the PA power efficiency degrades the PA linearity. Proposal 1 Linearity degradation due to advanced power amplifier design should be considered when studying the BS ACLR feasiblity Proposal 2 At least 5dB ACLR degradation margin for using advanced power amplifier design should be taken into account when studying the BS ALCR feasilibity Proposal 3 The feasible BS ACLR level for example frequency range of 30GHz is 31dB Proposal 4 The feasible BS ACLR level for example frequency range of 45GHz and 70GHz is 28dB and 25dB, respectively. Nokia wondered what was the maximum achievable ouput power of the BS. The Chairman proposed off line discussion.
 		noted No    
    R4‑1700111 Consideration on BS ACLR for NR Huawei other Approval Yes
YesHuawei presented the document and its conclusion: Proposal 1: ACLR for NR is defined at the lowest/highest carrier transmitted on the assigned channel frequency, which is the same as legacy UTRA/E-UTRA/MSR ACLR definition. Proposal 2: It is proposed to adopt 40 dBc ACLR for 30 GHz NR band. Sumimoto questioned whether the distortion came from beam forming. Ericsson agreed with Sumimoto: the performanced of the tx power amplifier was critical. Nokia feared that the multiple PA scenario was very complicated, and would result in reduced efficiency. Huawei responded.
 		noted No    
    R4‑1700068 UE ACS and blocking for mm-wave frequencies Ericsson discussion Discussion Yes
YesThe contents had been captured in the way forward.
 		noted No    
    R4‑1700223 BS ACS and blocking for mm-wave frequencies Ericsson other Approval Yes
YesEricsson explained that the document was a proposal on how to reply to the ITU-R LS. Blocking was determined by the ACS level. Nokia saw a big difference in the simulation presented here and in the Huawei tdoc 137: 40 dB. It could be a question of the utilization of the 200 MHz channel width. Ericsson explained the difference in the curves based on blocking as a function of ACS. Indeed the channel occupancy had been taken as 100%, but did not think this would lmake a big difference to the result. Further discussion was evidently needed. Huawei stated that the current blocking requirements had not been based on ACS, and the term "blocking" was in this case misleading.
 		noted No    
    R4‑1700274 Way Forward on BS SEM Ericsson other Approval Yes
YesEricsson presented the outcome of the discussions. Nokia said there was no agreement on the 20 dB figure. Ericsson pointed out that this was just an example. Nokia believed that the Plimit would also change as a function of ACLR agreed. Ericsson was willing to revise the document after further discussions. Nokia would prefer to remove the example figures.
 		revised No R4‑1700287  
    R4‑1700287 Way Forward on BS SEM Ericsson, Nokia, Alcatel-Lucent Shanghai Bell other Approval Yes
Yes
 		approved No   R4‑1700274
    R4‑1700276 WF on UE sensitivity blocking response for the ITU-R LS Ericsson other Approval Yes
Yes
 		approved No    
    R4‑1700285 Way Forward on UE ACLR and BS ACS Qualcomm Incorporated other Approval Yes
YesQualcomm had had extensive discussions with other companies to arrive at this docoument. Huawei had reservations on ACS and wished for a little more time.
 		revised No R4‑1700289  
    R4‑1700289 Way Forward on UE ACLR and BS ACS Qualcomm Incorporated other Approval Yes
Yes
 		revised No R4‑1700303 R4‑1700285
    R4‑1700303 Way Forward on UE ACLR and BS ACS Qualcomm Incorporated other Approval Yes
Yes
 		approved No   R4‑1700289
    R4‑1700302 Way forward on ACLR and ACS for WP5D LS Huawei,Nokia, Ericsson, Qualcomm other Approval Yes
YesHuawei indicate that the discussions had been long and tough, but it had been possible to propose a way forward.
 		approved No    
3.1.2.2 Spurious emissions [FS_NR_newRAT] R4‑1700233 NR spurious emissions for BS and UE in ITU-R response Ericsson LM other Approval Yes
YesEricsson explained that the title on the cover was incorrect, but the contents was valid. Two proposals were made: PROPOSAL 1: The spurious emissions for BS and UE should in the ITU-R response be based on Category A limits. A text should be included in the LS response to reflect that additional limits and conditions can be added for specific scenarios that may require a lower limit, as has been done before. PROPOSAL 2: The spurious emissions for BS and UE should in the ITU-R response be defined as TRP. Nokia cautioned against confusing ITU-R by the RAN4 response. Huawei believed that both category A and B were a regional requirement and should be included, at least for the BS. Ericsson clarified that this proposal was for both UE and BS. He noted that the Region 1 requirement was in fact a recommendation, and some European Harmonized Standards did not follow the recommended values. Qualcomm had a supporting contribution. In principle, 3GPP should agree on what was feasible. Qualcomm had a strong preference to retain -13 dB. Nokia wanted this value for both UE and BS. Skyworks agreed in the case of the UE at least. The Chairman noted that the current specs had different values for UE and BS. Ericsson was anxious to avoid inconsistency in response and the implied difficulty of roaming of Ues between regions.
 		noted No    
    R4‑1700119 Further consideration on spurious emissions Huawei, Hisilicon other Approval Yes
YesThe document was noted without presentation.
 		noted No    
    R4‑1700083 On Emissions in UE RF Intel Corporation other Discussion Yes
YesOriginal agenda item 3.5.4. Intel explained that the ITU limits were very challenging and they prefered the FCC limits.
 		noted No    
    R4‑1700011 Potential issue with second harmonic emission level for 28GHz bands Skyworks Solutions Inc. discussion Approval Yes
YesSkyworks believed the -13 dBm/MHz was well accepted. It was concluded that - A -13dBm/MHz spurious emission level seems achievable without a specific harmonic filter - To reach -30dBm/MHz spurious emissions however some form of H2 filter need to be added between the PA and the antenna or some exceptions in a limited set of directions should be allowed for second harmonic. And the following proposals were made: Proposal 1: It is proposed to study if post-PA filtering is required for an NR transmitter architecture using the two spurious emissions limits of -13dBm/MHz and -30dBm/MHz specifically for H2 of NR terminal above 6GHz. Proposal 2: It is proposed to study if post-PA filter is required for NR transmitter architecture above 6GHz assuming a H2 occupied bandwidth with a minimum of 1MHz. It is also proposed to focus the study on NR bands around 28GHz since they may present the larger H2 attenuation needs. Proposal 3: - Assume that H2 is beam-formed with a smaller beam width than fundamental and thus e.i.r.p. spurious emission requirement should focus on the strongest H2 beam. - H2 antenna gain vs fundamental is FSS. DOCOMO wondered about the second harmonic for other regions. Was the -13 dB level possible? Skyworks wsa not sure and it would depend on the architecture of the radio. They encouraged other company to consider the bandwidth issue too, for future work. Qualcomm was unsure on proposal 2 and 3, and did not wish to get distracted from the necessity of responding to ITU-R WP 5D. It was agreed that -13 dm/MHz was the figure to be used.
 		noted No    
    R4‑1700097 Spurious emissions for mmWave Qualcomm Incorporated other Approval Yes
YesThe document was not presented, in the light of the discussion on the previous tdoc.
 		not pursued No    
    R4‑1700120 TP for 38.803: spurious emissions for mmWave Bands Huawei, Hisilicon pCR Approval Yes
YesThe Chairman stated that the document would be treated if time allowed. The question was whether or not to include category B.
 		noted No    
    R4‑1700275 Way Forward on BS Spurious Emissions Ericsson other Approval Yes
YesEricsson presented the result of discussions. Huawei queried where was the Note 2 mentioned by Ericsson. Ericsson replied that it was in tdoc 232 but was not reproduced in the present document, but the text was indeed on slide 3. Huawei was not comfortable with the background slide. Huawei also wanted the text "Limit for all countries" removed. Ericsson sought to justify leaving it, because the Recommendation did cite other values, but this was the only one mandated, and appeared in the radio regulations.
 		revised No R4‑1700288  
    R4‑1700288 Way Forward on BS Spurious Emissions Ericsson other Approval Yes
Yes
 		approved No   R4‑1700275
3.1.2.3 Noise figure [FS_NR_newRAT] R4‑1700093 TP for NR Rel-14 TR 38.803: Noise figure in mm-wave systems as assumed for ITU-R related coexitence simulations Ericsson pCR Approval Yes
YesThe document was noted without presentation.
 		noted No    
    R4‑1700115 NF for mmWave bands Huawei, Hisilicon other Approval Yes
YesThe document was noted without presentation.
 		noted No    
    R4‑1700116 TP for 38.803: NF for mmWave bands Huawei, Hisilicon pCR Approval Yes
YesThe document was noted without presentation.
 		noted No    
3.1.2.4 Others [FS_NR_newRAT] R4‑1700170 Text improvement of throughput vs SINR mapping SAMSUNG Electronics Co., Ltd. pCR Approval Yes
YesOriginal agenda item 3.8. Samsung introduced the document. Nokia noted the presence of SNR and SINR; this needed to be harmonized. They also wished to retain similar wording to that used for LTE, otherwise there would be confusion amongst users. Huawei believed the SINR/SNR confusion was historical in origin. They also believed that it was not appropriate to substitute "MCS level" for "code rate". Also, in the first paragraph, SNR was the correct term, not SINR.
 		revised No R4‑1700254  
    R4‑1700254 Text improvement of throughput vs SINR mapping SAMSUNG Electronics Co., Ltd. pCR Approval Yes
Yes
 		approved No   R4‑1700170
    R4‑1700067 UE OTA sensitivity for mm wave frequencies Ericsson discussion Discussion Yes
YesOriginal agenda item 3.1.2.1. No sensitivity values would be reported to ITU.
 		noted No    
    R4‑1700224 BS OTA sensitivity for mm wave frequencies Ericsson other Approval Yes
YesNo sensitivity values would be reported to ITU.
 		noted No    
    R4‑1700137 BS In-band blocking for urban macro scenario in 30 GHz Huawei, Hisilicon other Approval Yes
YesOriginal agenda item 3.1.1.1. Huawei explained that this contribution was based on simulation. Ericsson noted that some of the difference in results between Huawei and Ericsson might be explained by the 99% vs 99.9% point considered. Nokia recalled a similar problem when defining LTE.
 		noted No    
    R4‑1700138 TP for 38.803: BS In-band blocking for urban macro scenario in 30 GHz Huawei, Hisilicon pCR Approval Yes
YesOriginal agenda item 3.1.1.1. Nokia noted that this proposal was not aligned with those of other companies and therefore should not be approved. The authors accepted this and asked other companies to contribute. Nokia believed there were multiple interpretations leading to widely different values, so the simulation assumptions were FFS. Ericsson had raised several points for further study, and it was premature to draft a way forward. It was confirmed that no values were needed for the response to ITU-5D, just the general simulation approach.
 		noted No    
    R4‑1700228 On BS sensitivity and blocking response Nokia, Alcatel-Lucent Shanghai Bell other Approval Yes
YesNokia introduced the document which presented a different approach to that taken by Ericsson and Huawei. Ericsson agreed that this approach was viable. Adjusting for the bandwidth correction might give similar results to theirs. Huawei understood that the inband blocking requirement was based on the second adjascent channel, and this would give a substantially different result from using the first adjascent channel. They believed there was some confusion over the terms "blocking" and "ACS". Ericsson believed there was no confusion and that a blocking level based on ACS could be offered to ITU-R. Huawei believed RAN4 used different terminology. To avoid misunderstanding it was necessary to align terms and methods. Nokia believed their document was clear in its conclusion and its definition of ACS. Off line discussion was needed.
 		noted No    
    R4‑1700268 WF on BS sensitivity and blocking for WP5 response Nokia, Alcatel-Lucent Shanghai Bell, Ericsson other discussion Yes
YesThe document was presented by Nokia.
 		approved No    
3.1.3 LS to WP 5D [FS_NR_newRAT] R4‑1700112 Reply LS to WP5D on co-existence parameters Huawei, Hisilicon LS out Approval Yes
Yes
 		noted No    
    R4‑1700232 [DRAFT] LS on Characteristics of terrestrial IMT systems for frequency sharing/interference analysis in the frequency range between 24.25 GHz and 86 GHz Ericsson LM LS out Approval Yes
YesIt was noted that this document had been drafted as a reply to an earlier incoming LS from ITU-R WP5D, R4-1609014 at RAN4#80 (Gothenburg, August 2016) and not as a reply to the follow up LS from 5D seen in R4-1609014 at RAN4#81 (Reno, November 2016). This would be corrected in a revision of the present tdoc.
 		revised No R4‑1700253  
    R4‑1700253 [DRAFT] LS on Characteristics of terrestrial IMT systems for frequency sharing/interference analysis in the frequency range between 24.25 GHz and 86 GHz Ericsson LM LS out Approval Yes
YesEricsson detected an error in the BS spectrum mask.
 		revised No R4‑1700305 R4‑1700232
    R4‑1700305 LS on Characteristics of terrestrial IMT systems for frequency sharing/interference analysis in the frequency range between 24.25 GHz and 86 GHz Ericsson LM LS out Approval Yes
Yes
 		approved No   R4‑1700253
3.2 General [FS_NR_newRAT] R4‑1700150 Updated TR for Study on New Radio Access Technology: RF and co-existence aspects NTT DOCOMO INC. draft TR Approval No
YesDocument not provided.
 		withdrawn Yes    
3.2.1 Wide Channel bandwidth/Transmission bandwidth configuration [FS_NR_newRAT] R4‑1700238 UE RF Bandwidth Adaptation in NR RAN1 LS in Action Yes
YesOriginal agenda item 3.5. The document was presented by MediaTek.
 		noted No    
    R4‑1700196 How to handle wider channel band for NR DOCOMO Communications Lab. other   Yes
Yes
 		revised No R4‑1700250  
    R4‑1700250 Editorial correction. DOCOMO Communications Lab. other - Yes
YesDOCOMO presented the document and its proposal that ? When the utilized channel bandwidth is equal or narrower than the feasible maximum bandwidth, single carrier approach is applied. ? When the utilized channel bandwidth is wider than the feasible maximum bandwidth, carrier aggregation is applied. Intel wondered how to define a "feasible" maximum bandwidth. ZTE commented on the second bullet that if the bandwidth was wider than feasible, there would be a negative impact on the spectral efficiency. CA would not be needed, and efficiency could be increased. DOCOMO explained why CA might be needed. Vodafone concluded that RAN4 would ask RAN1 for both options, but it would have been nice if RAN4 could have been more precise. But with both options, the market would decide. ZTE understood that this was implicitly refering to contibuous bandwidth.
 		noted No   R4‑1700196
    R4‑1700191 Wider Single Carrier Channel for NR Qualcomm Incorporated discussion Discussion Yes
YesQualcomm presented the document describing a wide channel approach, contrasting with LTE carrier aggregation, and concluding that careful design would ensure flexibility of implementation, and an operational equivalent between a wide single carrier channel and LTE-style carrier aggregation.
 		noted No    
    R4‑1700114 On wide channel bandwidth for NR Huawei, Hisilicon other Approval Yes
YesHuawei compared wide flexible channels with carrier aggregation and analysed the difference between the two. Four proposals were offered: Proposal 1: In one carrier, OFDM based baseband signal is generated by one single FFT for a given numerology from the implementation point of view. Proposal 2: Multi-carrier approach (e.g. CA/DC) should be used to support 1 GHz contiguous spectrum in NR. Proposal 3: The largest channel bandwidth should be separately defined for bands below 6 GHz and above 6 GHz. Proposal 4: It is proposed that the largest channel bandwidth for bands below 6GHz is 40MHz and the largest channel bandwidth for bands above 6GHz is 80MHz. T-Mobile thought the proposal for above 6 GHz was not appropriate because the frequencies came in 100 MHz blocks. Sprint was worried about proposal 4. RAN was discussing much wider bands than this tdoc was considering.
 		noted No    
    R4‑1700032 Bandwidth adaptation for NR Ericsson discussion Discussion Yes
YesEricsson presented the document, recalling an LS from RAN1 asking RAN4 to study UE RF bandwidth adaptation for single and multiple carrier operation. The conclusion was several observations: Observation 1 : For cases where the same cell is being received on the same band, switching with X in the range of 50-100uS should be technically feasible. Observation 2 : For multicarrier cases, especially if the band is different and a new RF chain needs to be started, the operation is similar to CA activation/deactivation and much larger X could be expected. Observation 3 : Detailed analysis on the power savings possible with UE RF BW adaptation is a complicated issue which needs further input and discussion in RAN4. Observation 4 : Power consumption seems to be the primary potential benefit of the proposed BW adaptation scheme, since other benefits could be obtained by frequency selective scheduling if power consumption was of no concern. Observation 5 : Neighbour cell syncronisation and RRM aspects need to be discussed in RAN4#82 before replying to RAN1 Intel was concerned that there would be spurious emissions during the settling time. Nokia wondered about the parameters in the tables, and the actual values of the delays. Ericsson thought this needed further analysis.
 		noted No    
    R4‑1700195 Study on specification impact of new CBW concept from view point of UE RF requirements DOCOMO Communications Lab. discussion   No
YesThe document was not provided.
 		withdrawn Yes    
    R4‑1700211 UE-specific RF Bandwidth Adaptation for Single Component Carrier Operation MediaTek Inc. other Discussion Yes
YesMediaTek presented the document, showing the potential for significant power saving using bandwidth adaptation. Several observations were made: Observation #1: Around 64% of daily UE power consumption is occupied by PDCCH-only and low data rate services. Observation #2: There is significant power saving for digital base-band power consumption if UE RF bandwidth adaptation for DL/UL control/data and DL measurements is allowed in single component carrier operation. - For RF bandwidth adaptation range of 5MHz~200MHz, the power saving for digital base-band by bandwidth adaptation is 35~92%, 29~90% and 31~84% when 10Mbps, 50Mbps and 100Mbps data rate are assumed. - For RF bandwidth adaptation range of 20MHz~1GHz, the power saving for digital base-band by bandwidth adaptation is 37~94%, 36~93% and 34~93% when 10Mbps, 50Mbps and 100Mbps data rate are assumed. Observation #3: The transition time of DL UE RF bandwidth adaptation is 1 ~ 4 slots (e.g. 1 slot = 0.125ms or 0.5ms). - The transition time is 1 slot if layer-1 signalling is considered for RF bandwidth adaptation signalling - The transition time is 4 slots if MAC-layer signalling is considered for RF bandwidth adaptation signalling Observation #4: The transition time of UL UE RF bandwidth adaptation is 1 ~ 4 slots (e.g. 1 slot = 0.125ms or 0.5ms). Observation #5: AGC settling may not require reference signals. - AGC settling doesn’t require reference signals if Tx PSD within a component carrier is uniform - AGC settling may require reference signals if Tx PSD within a component carrier is not uniform Nokia believed it was necessary to study demodulation techniques before replying to RAN1 Intel noted that DRX might serve the same power saving as this approach. MediaTek had done some studies on this, and had concluded that even with DRX there was scope for considerable power saving. This could be demonstrated at the next meeting.
 		noted No    
    R4‑1700212 UE RF performance for bandwidth adaptation consideration in NR MediaTek Inc. other Discussion Yes
YesMediaTek presented the document which analysed settling time for various parameters, all of which might be changed during a bandwidth reconfiguration. In some cases, power saving could be up to nine times for Rx and six times for Tx. Skyworks wondered whether the 50 us was sufficient allowance. Huawei shared the concerns, and wondered whether 200 us might not be more realistic in some cases. MediaTek had considered settling to 1 ppm but maybe further analysis was necessary before replying to RAN1. MediaTek said that the figures were based on their existing LTE technology.
 		noted No    
    R4‑1700214 Flexible channel bandwidth consideration in NR MediaTek Inc. other Approval Yes
YesMediaTek aired their view on flexible channel bandwidth, overcoming the need for carrier aggregation as in LTE. Proposal 1: NR to support flexible channel bandwidth with bandwidth resolution down to a single PRB in both DL and UL. Proposal 2: From network operation and scheduling perspective, UE only needs to report its maximum total bandwidth handling capability. Proposal 3: For sub-6GHz radio, the maximum channel bandwidth up to 200 MHz with the following bandwidth set for UE RF specifications can be considered in Rel-15. [5MHz, 10MHz, 20MHz, 40MHz, 80MHz, 200MHz] Proposal 4: For above-6GHz radio, the maximum channel bandwidth up to 1 GHz with the following bandwidth set for UE RF specifications can be considered in Rel-15. [100MHz, 200MHz, 400MHz, 800MHz, 1GHz] Proposal 5: The maximum channel bandwidth shall be allowed for further extension for future releases beyond Rel-15. Huawei wondered if it was intended to define fixed channel bandwidths. Skyworks asked if the UE would have to support the same bandwidth set for UL and DL. MediaTek responded that their proposals applied to both UL and DL, BS and UE. In real applications, it was possible to apply any bandwidth. Vodafone wondered if this implied some frequency-dependent scaling. MediaTek said that a finite set of fixed bandwidths would be defined. Contiguous CA was effectively a single carrier. Vodafone wondered how all this could be configured and managed. And these requirements would be band-independent. MediaTek emphasised the benefits of their approach when it came to establishing testing regimes. T-Mobile wondered whether the maximum capability would be limited to the values indicated. MediaTek responded that a finite set of maximum bandwidths could be defined. Sprint observed that the present bandwidth arrangements in LTE were somewhat constraining, and they hoped that much finer granularity would be possible in NR. Qualcomm wondered if a UE which only supported 40 MHz could operate in an 80 MHz channel. Also, the more options there were, the more testing was required. Testing time might be running to months to complete coverage. Sprint noted that even now, RAN5 did not specify that every bandwidth should be tested. Vodafone sought more details of precisely what was meant. RAN4 would have to write highly detailed specs, and RAN5 could choose an appropriate subset to test. T-Mobile pointed out the benefit of this system was the redundancy of traditional carrier aggregations. Skyworks understood that if a UE was 80 MHz capable, it could handle any narrower bandwidth. The Chairman gave his understanding of how the mechanism worked, and an exchange of clarifications ensued. Intel stated that there were three important parameters: the maximum bandwidth, the subcarrier spacing, and the largest FFT size. MediaTek pointed out that an FFT could be divided into three sub-FFTs. It was only the maximum capability which needed to be defined. If non-contiguous frequency ranges could be avoided, life would be much simpler.
 		noted No    
    R4‑1700062 Discussion on bandwidth adaptation in NR Nokia, Alcatel-Lucent Shanghai Bell discussion Discussion Yes
YesNokia's document dealt further with bandwidth adaptation and gave some observations: Observation 1: We expect that RF BW adaptation latency does not depend on whether the BW is increased or decreased. Observation 2: We expect that RF BW adaptation latency would be impacted if the adaptation includes change of center frequency. Observation 3: We expect that a bandwidth adaptation including band change would have longer latency than bandwidth adaptation which does not include band change. Observation 4: We expect that there will be some power saving opportunities from enabling UE bandwidth adaptation. RRM issues would need to be discussed further at the next meeting.
 		noted No    
    R4‑1700234 Tx BW configuration application to UE Tx MOP requirements VODAFONE Group Plc discussion Discussion No
YesLATE DOCUMENT Document not provided.
 		withdrawn Yes    
    R4‑1700127 Discussion on UE RF Bandwidth Adaptation in NR Huawei, Hisilicon discussion Discussion Yes
YesHuawei presented the document and its observations: Observation 1: It is expected that RF bandwidth adaptation without centre frequency change can be settled down within 20us. Observation 2: If maintaining the same conclusion of LTE, 100us~200us is a typical implementation for the PLL retuning. Vodafone considered that it might not be valid simply to take LTE values: there might be more scope for savings in NR.
 		noted No    
    R4‑1700128 DRAFT reply LS on UE RF Bandwidth Adaptation in NR Huawei, Hisilicon LS out Approval Yes
YesHuawei presented the draft outgoing LS. After some discussion, it was concluded that it was premature to reply at this stage,but that RAN4 should have some off line discussion.
 		noted No    
    R4‑1700243 LS on wider bandwidth operation for NR RAN1 LS in discussion Yes
YesDOCOMO presented the LS from RAN1 meeting concurrently with the RAN4 meeting.
 		noted No    
    R4‑1700272 Way Forward on flexible channel bandwidth consideration for NR Mediatek other discussion Yes
YesMediaTek presented this way-forward document. It incorporated remarks made off line. Companies were encouraged to suply more information on the feasibility of flexible bandwidth. The first Way Forward slide listed the major points to be considered. Onlly contiiguous bands needed to be considered. Vodafone did not really understand the implementation complexity in terms of the RF. The (LTE) UE can already transmit flexibly today. Was less flexibility being advocated. MediaTek did not think there was a real problem. Vodafone though the document mixed the question of flexible bandwidth with maximum bandwidth. These aspect should be separated to avoid confusion. Some clean up was needed. MediaTek stated that the implication was that flexibility was possible up to the maximum bandwidth. They were not entirely separate.
 		revised No R4‑1700295  
    R4‑1700295 Way Forward on flexible channel bandwidth consideration for NR [title to be checked] Mediatek other discussion Yes
Yes(invalid document, not treated)
 		revised No R4‑1700301 R4‑1700272
    R4‑1700301 Way Forward on flexible channel bandwidth consideration for NR [title to be checked] Mediatek other discussion Yes
Yes
 		approved No   R4‑1700295
    R4‑1700278 Way Forward on UE RF bandwidth adaptation in NR MediaTek other Approval Yes
YesMediaTek presented this way-forward document. Qualcomm could not agree to the numbers for transmission time, which had been taken from the Ericsson paper. It should be FFS.
 		revised No R4‑1700293  
    R4‑1700293 Way Forward on UE RF bandwidth adaptation in NR MediaTek other Approval Yes
YesAfter checking with Qualcomm, a new text was provided.
 		approved No   R4‑1700278
3.2.2 Others [FS_NR_newRAT] R4‑1700239 Subcarrier spacing and carrier frequencies RAN1 LS in Action Yes
YesOriginal agenda item 3.2.1. The document was noted without presentation.
 		noted No    
    R4‑1700181 Consideration on parameters in physical layer to be assumed for Rel-15 RAN4 NR specifications DOCOMO Communications Lab. other Approval Yes
YesOriginal agenda item 3.4.
 		revised No R4‑1700251  
    R4‑1700251 Editorial correction. DOCOMO Communications Lab. other Approval Yes
YesDOCOMO presented the document. In order to rationalize the work being performed in Rel-15, concluding with: Observation 1: It is quite important to clarify which parameters of each feature in PHY layer should be taken into account for Rel-15 RAN4 NR specifications. Observation 2: RAN4 workload significantly increases if we aim to specify NR RF requirements taking account of all possible combinations of modulation order and subcarrier spacing. For example if we assume three parts of frequency range, RAN4 has to discuss total 96 and 216 patterns for downlink and uplink respectively. Proposal 1: Clarify modulation order and subcarrier spacing to be assumed for Rel-15 NR RF requirements in each spectrum frequency range. Observation 3: Parameters which obviously have less feasibility at the time of introduction of Rel-15 NR could be excluded for Rel-15 NR RF requirements. Proposal 2: Analyse modulation order and subcarrier spacing to be assumed for Rel-15 NR RF requirements from view point of both feasibility and necessity. ? Analyse these parameters at least for sub 6GHz range and above [24GHz] range, respectively ? Analyse whether sub 6GHz range should be divided into several parts as well as above [24GHz] range ? Analyse these parameters for eMBB and URLLC, respectively Proposal 3: Clarify whether more than 15kHz subcarrier spacing need to be assumed for eMBB downlink requirements in sub 6GHz range.
 		noted No   R4‑1700181
    R4‑1700129 Consideration on subcarrier spacing for NR Huawei, Hisilicon other Approval Yes
YesHuawei discussed subcarrier spacing for NR, considering doppler spread and phase noise both below and above 6 GHz. The conclusion was: Proposal 1: 15kHz, 30kHz and 60kHz are proposed to be supported for below 6GHz. Proposal 2: 60kHz and 120kHz are proposed to be supported for NR bands up to 52.6GHz. Ericsson believed more discussion in RAN4 was required before formulating a response to RAN1. Concerning the proposals, it was not clear why three subcarrier spacing options might be necessary. And were the proposals appropriate? ZTE thought that aspects other than Doppler spread and phase noise needed to be discussed, as well as single carrier cases. Very wide channels might require much wider subcarrier spacing. Intel wondered why CMOS could not be used for the scenario above.
 		noted No    
    R4‑1700130 LS on subcarrier spacing and carrier frequencies Huawei, Hisilicon LS out Approval Yes
Yes
 		noted No    
    R4‑1700226 Aspects to consider for settling NR subcarrier spacing Ericsson other Approval Yes
YesOriginal agenda item 3.2.1. Ericsson presented the document, concluding with the proposal: Outline to RAN1 the RAN4 dependencies on subcarrier spacing and propose that the decision on applicable spacing for each band is taken in RAN4. Ericsson believed there had been some discussion in RAN1 on phase noise, but so far not in RAN4. Huawei considered that the subcarrier spacing needed to be decided during the work item phase. Ericsson believed RAN1 had already specified this. Huawei still believed that further discussion in RAN4 was needed.
 		noted No    
    R4‑1700060 Supported numerologies per spectrum range for NR Phase I Nokia, Alcatel-Lucent Shanghai Bell discussion Approval Yes
YesOriginal agenda item 3.2.1. Nokia gave the background to the questions asked by RAN1 and revealed their findings on appropriate subcarrier spacing for the various frequency ranges. Qualcomm wondered if there was a connection between the subchannel spacing the nominal bandwidth. Nokia replied that the figures were based on RAN1 findings, based on LTE compatibility. AT&T wondered if multiple carrier spacing per band could be supported. Nokia clarified that these were frequency ranges, not bands. But yes, multiple subcarrier spacings were supported.
 		noted No    
    R4‑1700061 Response LS on subcarrier spacing and carrier frequencies Nokia, Alcatel-Lucent Shanghai Bell LS out Approval Yes
YesOriginal agenda item 3.2.1.
 		noted No    
    R4‑1700041 LS on UE capability aspects for LTE/NR tight interworking RAN2 LS in Action Yes
Yes
 		noted No    
    R4‑1700240 UE capability aspects for LTE/NR tight interworking RAN2 LS in Action Yes
Yes
 		withdrawn Yes    
    R4‑1700063 UE capability aspects for LTE/NR tight interworking Nokia, Alcatel-Lucent Shanghai Bell discussion Discussion Yes
YesNokia recapped the deployment scenarios and reiterated the questions which needed to be addressed, effectively proposing an answer to the incoming LS in tdoc 41. Samsung wanted more detail on inter-band capabilities in the answers to questions 8 & 9. For question 5, other candidate solutions might exist. DOCOMO believed questions 6 to 9 did not need answeing at the present meeting. RAN4 was not yet ready to reply to question 1. Nokia agreed that there were a lot of FFSs in this document. They clarified that questions 8 & 9 related to > 6 GHz. In conclusion, perhaps it would be better to delay the response to the next meeting.
 		noted No    
    R4‑1700064 Response LS on UE capability aspects for LTE/NR tight interworking Nokia, Alcatel-Lucent Shanghai Bell LS out Approval Yes
YesThe foregoing discussion revealed that RAN4 was not yet ready to reply.
 		noted No    
    R4‑1700187 Synchronization signal frequency raster considerations Qualcomm Incorporated discussion Discussion Yes
YesQualcomm presented this document, dealing with the sync signal raster. This contribution presented synchronization raster design for both below 6GHz and above 6GHz. More specifically, it had the following proposals: Proposal 1: The frequency of synchronization signals should be aligned with the subcarriers of data channel. More specifically, the following design considerations should be taken into account when designing the synchronization signal frequency raster. - The synchronization signal frequency raster is an integer multiple of data subcarrier spacing. - The maximum data subcarrier spacing is 120kHz for below 6GHz and 240kHz for above 6GHz. Proposal 2: The channel raster is either 120kHz or 300kHz for below 6GHz, and [960 kHz] for above 6GHz. Proposal 3: The synchronization signal frequency raster is 36MHz for band category C. In addition, the following options for the synchronization signal frequency raster can be considered for band category A with Option A2 and band category B, - SR option 1: If the channel raster is 120kHz (CR option 1) for below 6GHz, the synchronization signal frequency raster is 2.28MHz for band category A with Option A2 and 4.68MHz for band category B. - SR option 2: If the channel raster is 300kHz (CR option 2) for below 6GHz, the synchronization signal frequency raster is 2.1MHz for band category A with Option A2 and 4.5MHz for band category B. Furthermore, if band category A with Option A1 is employed, a set of synchronization signal frequency locations are specified in the specification and no synchronization signal frequency raster is defined for band category A. Proposal 4: A larger synchronization frequency raster than the synchronization frequency raster for band category B can be considered for the below 6GHz frequency bands with wider bandwidth. Ericsson believed these proposals were interesting and warranted deeper investigation. There should be some flexibility to handle different frequency allocations. Concerning proposal 2, ZTE wondered was this applicable to the legacy bands? Qualcomm indicated that this was not aligned to the LTE raster, but this was of no consequence. Doing a brute force for sync signals would be far too time (and battery) consuming. Vodafone asked whether any margin around the centre point of the raster for putting the sync channel had been considered, as had been done in RAN1. Further off line discussions between interested parties should take place.
 		noted No    
    R4‑1700035 Discussion on the channel raster for NR ZTE Corporation discussion Discussion Yes
YesOriginal agenda item 3.6.1. ZTE presented the paper briefly. Off line discussions were needed.
 		noted No    
    R4‑1700198 Coordinate system for NR NEC other Approval Yes
YesNEC expressed their view that a common coordinate system was needed and offered two options: Option 1: Coordinate system for MIMO-OTA is adopted for both NR UE and NR BS Option 2: Coordinate system for MIMO OTA is adopted for NR UE, coordinate system for AAS is adopted for NR BS Huawei agreed it was important to do this, the AAS and the NR base stations should use the same system (option 2). Ericsson agreed, but considered a third option, with a minor change to the orientation of the Z axis. This could be captured and a decision taken next meeting. NEC clarified that the proposal was to use the AAS system for requirements, not for simulation. Off line discussion was needed.
 		noted No    
    R4‑1700282 Way forward on subcarrier spacing for NR Huawei, Hisilicon other Approval Yes
YesHuawei presented the way forward. Qualcomm asked if the subcarrier was for data transmission, or did it cover other uses such as synchronization signalling. They wished also to include 240 kHz. Huawei provided a lengthy but inaudible response.
 		revised No R4‑1700292  
    R4‑1700292 Way forward on subcarrier spacing for NR Huawei, Hisilicon other Approval Yes
Yes
 		approved No   R4‑1700282
    R4‑1700284 WF on coordinate system for NR NEC other Approval Yes
YesNEC presented this way forward. NTT DOCOMO wondered what was the reason to change coordinate system. NEC replied that this had been proposed by some companies. Huawei believed the point was that the AAS and NR BS should both use the same coordinate system.
 		approved No    
    R4‑1700286 Response LS on UE capability aspects for LTE/NR tight interworking RAN1 LS in Information Yes
Yes
 		noted No    
3.3 Spectrum [FS_NR_newRAT] R4‑1700013 Frequency Bands for 5G-RAN SPRINT Corporation other Approval Yes
YesT-Mobile preferred to take a holistic view to review all the bands rather than dealing with each band individually. Sprint pointed out a typo which needed coirrection.
 		revised No R4‑1700255  
    R4‑1700255 Frequency Bands for 5G-RAN SPRINT Corporation other Approval Yes
Yes
 		noted No   R4‑1700013
    R4‑1700015 Possible frequency ranges for NR below 6GHz China Telecom discussion   Yes
YesDOCOMO suggested modifying the fequency range of the bands to take account of regional variations. Care was needed not to confuse frequency ranges with actual bands. Huawei presented the document on behalf of China Telecom.
 		revised No R4‑1700256  
    R4‑1700256 Possible frequency ranges for NR below 6GHz China Telecom discussion - Yes
Yes
 		noted No   R4‑1700015
    R4‑1700193 12 GHz band for NR frequency prioritization in 6-24GHz Dish Network discussion Approval Yes
YesDOCOMO asked for clarification on 12 GHz: was it a candidate for the earlier or later stage of exploitation. Dish clarified that it was not for early use.
 		noted No    
    R4‑1700034 Frequency bands for the NR Rel-15 Work Item AT&T GNS Belgium SPRL discussion Approval Yes
YesDOCOMO wanted to consider the feasibility of a single band 37 to 40 GHz. T-Mobile supported AT&T's proposal. The FCC rules applied to the two frequency bands in question were different. ZTE recalled tdoc 249 which was related to this topic.
 		noted No    
    R4‑1700194 Frequency bands for New Radio ORANGE other Approval Yes
YesQualcomm noted that a lot of LTE bands mentioned in this document. Perhaps this was somewhat premature since no refarming of those LTE bands was yet contemplated as far as was known. Orange believed that it was in fact appropriate to consider these bands, though some reduction was possible for Rel-15. Sprint said that it was a question of how much it would be possible to reuse the existing specifications. Intel was nevertheless concerned at the extra work implied. Mediatec believed that the wider bands were better candidates for NR use rather than reuse of LTE bands. Intel remained concerned about the workload if LTE bands were considered, and it might not be possible to satisfy the wishes of all operators. Operators should be encouraged to prioritize their band requests. Vodafone remarked that the LTE bands needed to be considered for dual connectivity.
 		noted No    
    R4‑1700210 Frequency Ranges Proposal for NR in Rel-15 WI KT Corporation discussion Information Yes
YesChina Unicom was happy with this proposal, and band 42 should be considered first. China Unicom had a contribution in tdoc 241.
 		noted No    
    R4‑1700162 Dicussion on the proposal on harmonized TDD band plan in the 3.5GHz frequency range for 5G NR CMCC, Ericsson discussion Discussion Yes
YesCMCC presented the document, proposing bands for early consideration. Qualcomm was puzzled that the two proposals were incompatible. CMCC responded tht it was a question of timing and referred to the DOCOMO proposal. DOCOMO did not believe that a study item for a Rel-15 work item was needed for 3,5 GHz.
 		noted No    
    R4‑1700189 Draft SID: Harmonised TDD band plan in the 3.5GHz frequency range for 5G NR CMCC, Ericsson discussion   Yes
Yes
 		noted No    
    R4‑1700124 On NR bands Huawei, Hisilicon other Approval Yes
YesThe Chairman requested that the presentation should focus on the migration part. Huawei presented the relevant part of the document. There was no need to migrate all LTE bands to 5G, and the two RATs would co-exist in some parts of some bands. Method 2 was considered difficult to achieve in reality and its usefulness was limited. Samsung noted the need to make progress for Rel-15 and some example bands needed to be chosen. However LTE band 32 was not a good choice, since it was not a full-functionality band. The Chairman recalled similar discussions at Rel-10. ZTE was concerned on proposal 2 concerning co-existance. There was need for this to be covered in a study item, and may require a subsequent later work item. Verizon was concerned about the timings of proposal 1. Many ITU regions were impacted, and the study might be prolonged. They proposed other bands as better candidates. Qualcomm wondered why in proposal 3 the particular bands were proposed. It would be difficult to refarm only part of a band with a mix of uplink and downlink. This would result in chunks of unused spectrum. Huawei responded that band 32 had been proposed in consideration of WRC decisions. Qualcomm wondered what was the use case, and what UL/DL pairing wsa proposed. Ericsson noted that multiband support above 6 GHz was difficult, but it would be good to find some harmonized ranges. Vodafone noted that this was under discussion in RAN1. Vodafone questioned the term "TDD" and preferred the term "paired" or "unpaired" bands. The Chairman wished to define carefully the terminology of band re-use; what exactly was meant? Nokia agreed that re-use was tricky, and that studies on the impact of re-use needed to be done. Vodafone had understood that "re-use" meant examining the existing specs and seeing what changes might be needed for NR. It was not necessarily appropriate to simply retain the parameters inherited from LTE (some of which had in fact been inherited from WCDMA). Intel warned against delaying the time to market because of too complex a re-use policy when defining bands.
 		noted No    
    R4‑1700139 Potential harmonized bands for NR NTT DOCOMO, INC. other Approval Yes
YesNTT DOCOMO presented the document, outlining options for NR band arrangements. MediaTek was not in favour of the first option, and was concerned about the signal level of LTE in band 28. Therefore a single band might be the best approach. Qualcomm agreed. CMCC supported proposal 1. Huawei thought that specifying a single wide band was promising but there was some risk because the range 4.2 to 4.4 GHz had some military rader operating in it. Samsung thought that the option of a single band from 24.25 to 29.5 GHz was not technically feasible. Skyworks agreed that the full mm-band was not achievable in a single band. Nokia wondered whether contiguous spectrum would be available in these bands. The Chairman observed the use of an allocated band was up to the operator in question.
 		noted No    
    R4‑1700136 Overview on how to organize the NR band related work Huawei, Hisilicon other Approval Yes
YesHuaway presented the document, concluding that the Rel-15 NR band related work should include: • Specify the NR bands and the corresponding requirements • Identify and define the new bands in the frequency range higher than or equal to 3.5GHz • Identify and migrate LTE bands to NR • Investigate and specify a new set of RF requirements referring to the existing LTE RF requirements • Specify the necessary RF requirements for the new NR features based on the identified NR bands • Identify and specify the LTE-NR DC band combinations and the corresponding requirements • Identify and specify NR-NR CA band combinations and the corresponding requirements, if needed Vodafone thought that LTE-NR DC was definitely required. Lower bands would be needed early. The Chairman thought this would require further discussion at the next meeting.
 		noted No    
    R4‑1700125 Consideration of flexible duplex distance Huawei, Hisilicon other Approval Yes
YesHuawei presented the document concluding that flexible duplex distance should be supported for NRbecause: - Pairing a low frequency UL carrier with a high frequency DL carrier is a solution to solve UL coverage problem in high frequency and make best use of the UL spectrum resource in low frequency. For example, a UL below 6GHz (e.g. 700MHz) pairing with a DL above 6GHz (e.g. 28GHz). - Flexible duplex distance can provide a good application scenario for LTE/NR co-existence or LTE/NR dual connectivity in the stage of LTE migration to NR and it cannot be realized by neither LTE CA nor NR CA. For example, 1.4GHz is a SDL band in Europe which may be allocated for 5G NR and it can be paired with another UL band, e.g. Band 8 UL (880MHz - 915MHz), then the UL carrier is shared with LTE and NR with the DL carrier dedicated for LTE and NR. - If this flexible duplex distance can be supported in 3GPP, it would help the spectrum regulatory bodies to allocate more fragmented spectrum to IMT. T-Mobile wondered if both spectra would be owned by the same operator. Huawei replied that this was indeed the intention. Ericsson had difficulty to distinguish carrier aggregation from flexible duplex. For some DL only bands, they could be paired with different UL bands. Figure 1 was confusing. The red block was likely to be owned by a single operator. Nokia could not find any justification for this flexible duplex. Carrier aggregation was sufficient. If there were coverage problems with the high frequencies, there could be CA with lower bands. Vodafone wondered whether the RF architecture did not cater for UL and DL as it had originally been defined for LTE. This separation of UL and DL might just cause more difficulty than it was worth. He believed that flexible duplex probably differed from CA by the protocols used. Intel was inclined to agree with Ericsson. The case needed to be consiered where CA interband and intraband needed to be defined, considered from both the UL and the DL point of view. But what were the benefits? Huawei responded that this catered for asynnetric traffic, and where LTE and NR could be combined for uplink and downlink.
 		noted No    
    R4‑1700153 Discussion on NR Bands Definition ZTE, ZTE Microelectronics discussion Decision Yes
Yes
 		revised No R4‑1700249  
    R4‑1700249 Discussion on NR Bands Definition ZTE, ZTE Microelectronics discussion Decision Yes
YesZTE considered the need for three band groups, much along the lines proposed by China Telecom.
 		noted No   R4‑1700153
    R4‑1700230 Considerations on multi-band support in NR ZTE Microelectronics Tech. Co. discussion Approval Yes
YesZTE presented the document. Observation 1: UE still needs to support multi-band due to the lack of unique band for the purpose of initial access and global roaming. Observation 2: Multiple RF chain and reconfigurable RF front-end may offer a promising solution to reduce UE’s complexity and costs to support flexibly multi-band and multi-standard. Proposal 1: Standardization may not reduce the freedom for a UE vendor to define supported band portfolio from specifications according to its own global market strategy and technical considerations. Proposal 2: UEs’ complexity and costs reduction is the primary target for standardization on multi-band support to facilitate initial access and global roaming.
 		noted No    
    R4‑1700235 Consideration on specifying frequency bands VODAFONE Group Plc discussion Discussion No
YesLATE CONTRIBUTION Document not provided.
 		withdrawn Yes    
    R4‑1700237 Feasibility of wide operating band for mmWaves Qualcomm Incorporated discussion Discussion No
YesLATE CONTRIBUTION Document not povided.
 		withdrawn Yes    
    R4‑1700241 Consideration on possible frequency ranges for NR China Unicom discussion Discussion Yes
YesLATE CONTRIBUTION
 		noted No    
    R4‑1700242 Frequency 28GHz band for the NR Rel-15 Work Item Verizon other Approval Yes
YesLATE CONTRIBUTION
 		revised No R4‑1700244  
    R4‑1700244 Adds source T-Mobile Verizon, T-Mobile other Approval Yes
Yes
 		noted No   R4‑1700242
    R4‑1700260 Candidate frequency bands for 5G New Radio Etisalat other Approval Yes
Yes
 		revised No R4‑1700267  
    R4‑1700267 Adds additional source. ETISALAT, Huawei, HiSilicon other Approval Yes
Yes
 		noted No   R4‑1700260
    R4‑1700257 WF on NR spectra related work NTT DOCOMO other discussion No
Yes(Document not available; revised without presentation)
 		revised No R4‑1700304  
    R4‑1700304 WF on NR spectra related work NTT DOCOMO other discussion Yes
YesThe way forward was presented by NTT DOCOMO. T-Mobile and Verizon supported the AT&T contribution of tdoc 34. Sprint thought that slide 4 was redundant. Slide 6: Intel wondered what was the objective of this way forward. How long could the discussion continue: until the end of the SI or to continue during the subsequent WI. Sprint observed that the frequency ranges listed in slide 3 was incomplete, and could be expanded in the next meeting. NTT DOCOMO believed that Sprint's question would be resolved in the next meeting. The intention of slide 5 was to encourage the suggestion of further bands. Vodafone believed that some of the bands of slide 6 were already the subject of existing work items. Overlap should be avoided. ETISALAT warned agaiinst mixing the LTE and NR situations.
 		noted No   R4‑1700257
    R4‑1700281 WF on flexible duplex distance Huawei, HiSilicon, China Unicom, China Telecom, Orange other Approval Yes
YesHuawei presented the way-forward document. Ericsson understood that this was an LTE system cooxisting with an independent NR system. But from a band arrangement, how would this differ from inter-band CA? This topic could be deprioritized. Qualcomm agreed with Ericsson. Intel commented that this was new work. In addition the document did not make it clear whether UL or DL was being considered; much more specific wording was needed. Huawei stated that this topic had been discussed in RAN1, and would be included as a Rel-15 WI. RAN4 should embrace this work. It should be clarified that the UL was under consideration.
 		revised No R4‑1700298  
    R4‑1700298 WF on LTE-NR coexistence Huawei, HiSilicon, China Unicom ,China Telecom, Orange, Vodafone, Deutsche Telekom other Approval Yes
Yes
 		approved No   R4‑1700281
3.4 Common RF issues for UE and BS [FS_NR_newRAT]                      
3.4.1 Spectrum utilization [FS_NR_newRAT] R4‑1700134 Consideration on NR waveform and spectrum utilization Huawei, Hisilicon other Approval Yes
YesHuawei presented the document which discussed window size etc for optimum spectral efficiency for various technologies and configurations. The following proposals were offered: Proposal 1: NR spectrum utilization should be defined as a range, taking into different numerology, carrier bandwidth, and applied waveform with different spectrum confinement performance. Proposal 2: For above 10MHz carrier bandwidth, the spectrum utilization range can be [90%, 99%] for all applied numerologies. For below 10MHz carrier bandwidth (including 10MHz), the spectrum utilization range can be [90%, 97%] for all applied numerologies. Proposal 3: Ues’ capability in terms of the supported data transmission bandwidth for both downlink receiving and uplink transmitting should be reported to network for accurate scheduling. Proposal 4: gNBs’ capability in terms of the supported data transmission bandwidth for both downlink transmitting and uplink receiving can be notified to Ues for possible resource coordination. Nokia believed that the figures of §2.1 had not been defined by RAN4. He believed a similar approach should be taken in NR as for LTE. In §2.2, it was stated that the flexibility implied a too high degree of complexity. Deutche Telekom was not clear on proposal 4. Ericsson thought that a more interesting metric was a measure pf end user service and throughput. Also the concentration had been on downlink emissions, but it was also necessary to examine the uplink, eg for narrowband blocking. Was it the intention that there were two values of utilization (97 or 99%) or was it possible to have intermediate valaues? ZTE believed proposal 1 was reasonable. But for proposal 2, 90% was reasonable, but only a single numerology was offered. However it was likely that a mixed numerology would pertaiin in practice; were the figures still valid? It was also necessary to measure spectral efficiency, not just utilization. Other studies had shown that with stringent filtering, the throughput might decrease, but a lower upper limit might also be acceptable. For proposals 3 & 4, more study was needed to see what really needed to be reported; there was a danger of over complex signalling. DOCOMO wondered why the capability needed to be reported in proposal 4. Qualcomm thought the capabilities of proposals 3 & 4 were important to consider. There would be different blocking requirements, and if there were a different operator in a differenct cell, the situation would be more complicated. Huawei responded to these comments. There was as yet no RAN4 agreement on the figures, but RAN4 had seen input from many companies. On how to define the utilization, he believed that the focus of the contribution was to see what level of utilization could be achieved, but this was pre-normative work. On proposal 4, it was appreciated that there could be differrent treatment of signalling of BS and UE. It was possible for neighbouring base stations to cooperate in handover, and other scenarios could be envisaged. The benefits of spurious emission complexity was a given. For the uplink, the limiting case of narrowband blocking, there was no immediate answer. Views of other companies were welcome. For mixed numerology, the focus of the study should be how to enable cooexistance. On the topic of fractional PRB, there were other papers at this meeting which should be addressed. Vodafone thought this type of proposal where the system was allowed to operate in higher utilization was valuable, and needed to be evaluated further.
 		noted No    
    R4‑1700204 PA and TRX impairment impact on guard band and in band emissions performance with multiple numerologies in UL Skyworks Solutions Inc. discussion Agreement Yes
YesSkyworks briefly presented this discussion document. The proposed way forward was: in order to achieve a realistic design of the spectrum utilization, in-band emissions and guard bands it is proposed that for sub-6GHz NR: • TRX impairment budget is equivalent to the agreed LTE UL 256QAM numbers (-34dBc image rejection); • Simulations takes into account realistic PA out-of-band emissions with NR waveforms; • Introduction of simple PAPR reduction techniques is studied to improve PA contribution and/or efficiency; and • Applicability of the above for above 6GHz NR is FFS. In response to a question from Intel, Skyworks clarified that base statiions would be able to suppress leakage, but this might need further study. In response to Qualcomm, Skyworks proposed not tostart with 25 dBc as a starting point for NR, but rather start with the known best for LTE. Ericsson wondered whether other aspects such as clipping had been considered. Further discussion was needed.
 		noted No    
    R4‑1700099 Spectral Utilization Analysis for NR Qualcomm Incorporated discussion Discussion Yes
YesQualcomm analysed the techniques for spectral utilization. The document compared windowing and filtering and showed the trade-offs. It was also necessary to examine the complexity of implementation of the different approaches. Filtering was much more complex and therefore costly than windowing. ZTE agreed that both windonwing and filtering had their merits. The paper had concluded that both options should be "enabled" but was puzzled by exactly what was intended: were two sets of requirements implied? Qualcomm responded that one set of conformance tests had been established, the implementation should allow either implementation technique. Nokia agreed that it was important to study the single PRB and the guard bands. Ericsson agreed that the spec should not prevent any implementation method, but should cater for less than optimum techniques. Huawei was concerned that table 1 indicated poorer performance than current LTE. But both windowing and filtering had their merits and their disadvantages.
 		noted No    
    R4‑1700009 Considerations on NR spectrum utilization and guard band using fractional PRB ZTE Wistron Telecom AB discussion Discussion Yes
YesZTE presented the document, which sought to address some pointe raised at the previoius meeting. An extensive list of use cases was analysed. The paper concluded with the following observations and proposals: Proposal 1: In the existing bands assigned to the legacy RATs below 6 GHz, there are use cases e.g. URLLC and URLLC/eMBB multiplexed in the same band for the combinations of narrow channel BW and larger SCS such as 5MHz@60kHz. RAN4 should continue to study the methods to increase spectrum utilization Y over 90% for these combinations such as 5MHz@60kHz. Proposal 2: In order to make NR efficiently using refarmed narrow bands from the existing bands assigned to the legacy RATs such as LTE, UMTS, etc. below 6 GHz, for NR RAN4 could consider to support LTE channel BWs 1.4/3/5/10/15/20 MHz for these existing bands assigned to the legacy RATs. Proposal 3: Fractional PRB method could be considered as one of the methods for channel bandwidth configuration at the band edges to improve spectrum utilization Y over 90% for the combinations of narrow channel bandwidths and larger SCSs such as 5MHz@60kHz SCS. Proposal 4: In NR at basestation TX side it’s proposed to have guard band between two different numerologies to support high order modulations such as 256 QAM for DL when the radio condition is good for using the high order modulations. Proposal 5: In NR it’s proposed to use 15 kHz as the granularity to define guard band size between subbands with different numerologies, and the guard band size could be defined as 15 kHz * n (n is positive integer). Proposal 6: Using the same principle as NB-IoT with subcarrier granularity resource allocation, in NR it’s proposed that several subcarriers next to the guard band at band edges could be studied further for data transmission to increase spectrum utilization. Proposal 7: Using the same principle as NB-IoT with subcarrier granularity resource allocation, in NR it’s proposed that the rest of subcarriers in the RB where part of the RB is used as subband guard band between two different numerologies could be studied further for data transmission to increase spectrum utilization. Proposal 8: RAN4 could consider to use this equation [not reproduced in these notes] later on to calculate the spectrum utilization Y. Observation 1: With the fractional PRB method at the band edges a spectrum utilization gain of more than 5% could be achieved than only integer number of PRBs used for channel bandwidth configuration for the combination of 5MHz@60kHz SCS. Observation 2: No matter integer number of PRBs or fractional PRB (at band edges) is used for channel bandwidth configuration, the guard band size at both band edges is fractional PRB (less than 12 SCs) for 5MHz@60kHz SCS. Observation 3: If the subband guard band size could be reduced to what’s really needed e.g. some subcarriers, this could help to increase intra-band frequency resource utilization especially when several different numerologies such as different SCSs are multiplexed and large PRB sizes (i.e. large SCSs such as 240 kHz) are used. Observation 4: Several 15kHz-subcarriers guard band are needed to mitigate interference from adjacent interfering sub-bands. Observation 5: RB is not suitable as a basic unit to define guard band size since RB frequency sizes are different for different SCSs. Observation 6: NB-IoT uses 1 subcarrier instead of 1 RB as the minimum scheduling resource. The same principle could be used in NR. Nokia did not see the need to support lower frequency bands in proposal 1, and did not accept proposal 2. For proposals 6 & 7, they were not in favour of the proposed technique. Qualcomm did not agree with the first three proposals. They did not indicate whether these were practical use cases. Testing with fractional PRBs would be much more complex (factor of 12?) uising this approach. ZTE would have liked to have heard the views of operators on these matters. The Chairman noted that there had been no objections to proposals 4 & 5. But DOCOMO asked whether it was intended to specify the guard band requirements. ZTE replied that this was indeed the intention. Ericsson did not think it was necessary to specify a guard band: there were other techniques available.
 		noted No    
    R4‑1700188 Different Numerologies for NR Qualcomm Incorporated discussion Discussion Yes
YesOriginal agenda item 3.4.3. Qualcomm recalled discussions at previous meetings, observing that introduction of partial RB allocation to support large subcarrier spacing in narrow channels wasnot justified. An analysis on feasible/useful numerologies (subcarrier spacing) for different frequency bands was needed to limit the number of options allowed Samsung did not agree and sought operator input. ZTE was less dismissive of the low but count scenarios. Those message sizes might be useful in some circumstances. Vodafone (and ZTE) wondered whether this sub-PRB approach was just for use at the edge of the channel, or over the whole bandwidth. ZTE believed that it was a function of the target spectral utilization. Qualcomm would like to see what sort of messages could be send using these very small number of bits, including upper layer aspects.
 		noted No    
    R4‑1700057 NR DL spectrum confinement techniques at BS TX Nokia, Alcatel-Lucent Shanghai Bell discussion Discussion Yes
YesNokia had studied the different numberology cases including the impact of some spectrum confinement techniques (tx only). Performance was always rather poor without guard bands. In the 50 PRB case the different waveform techniques did not impact the performance at BS Tx. Basic CP-OFDM was sufficient without additional processing on top of CP-OFDM. Ericsson believed the conclusion did not take account of other techniques and strategies. Perhaps these should be left as implementation options, and not specified. Huawei referred to fig. 3, and pointed out that the results depended on the scenario in question. Other, less demanding coding rates, would reduce the difference in performance. Nokia agreed that this study had been specific for 64 QAM.
 		noted No    
    R4‑1700131 OOBE, EVM and BLER performance evaluation for NR waveforms Huawei, Hisilicon discussion Discussion Yes
YesHuawei introduced this lengthy contribution, concluding: Observation 1: Without any BLER performance degradation, the achievable maximum spectrum utilization by f-OFDM approaches theoretical upper limit. Observation 2: For W-OFDM, desirable demodulation performance in fading channel and spectrum utilization can’t be achieved simultaneously. To be specific, - The demodulation performance loss for 64QAM is 2dB, with 90% spectrum utilization - The spectrum utilization is lower than 90% if not sacrificing demodulation performance. Observation 3: The band edge PRB distortion can be easily solved by network scheduling, and the additional band edge PRB reaped by f-OFDM will contribute to system throughput improvement. Nokia noted that the paper had not evalluated 256 QAM. They believed that high efficiency was more important than high utilization. Some comparisons were not valid due to the different techniques used. Ericsson thought that the first observation in the document had not really been demonstrated. A better spectral efficiency did not necessarily come from the highest spectral utilization. They were not convinced that observation 3 was true in all cases. ZTE recalled that this had already been discussed in the previoius meeting, and they shared Ericsson's concerns. Huawei responded to Nokia's point: he believed extension to 256 QAM had been an option, and its use needed very favourable channel conditions. Network scheduling would help in this case. Concerning the unfair comparrison, figure 1 showed theoretical performance, and there was no difference: thus there was no unfairness in the comparison.
 		noted No    
    R4‑1700132 Evaluation on the complexity of NR waveforms Huawei, Hisilicon discussion Discussion Yes
YesHuawei briefly introduced the document, first looking at band edge block and mid-band block filtering, and evaluation of the performance. Observations 1: The block-wise filtering scheme can reap the benefits of both W-OFDM and f-OFDM, and the filter design in this scheme can be fixed independent of signal bandwidth. Observation 2: The complexity of block-wise filtering scheme is comparable to the existing LTE Tx shaping filter and also the Rx filter. Observation 3: The link performance degradation in terms of OOBE and BLER for the block-wise filtering scheme is negligible. Qualcomm wondered about the case of different numberologies on the rx side. ZTE thought that fig. 1 mid-bandblocks would need a steep roll-off, and if the filter were fixed for the edge block, worse performance would be expected. What was the result with a small number of allocated PRBs? Ericsson was concerned over the comparison with the LTE situation, and noted the complexity of the need for block trimming. Huwei was concerned that ZTE seemed to be working on a way forward even before the discussion documents had all been trreate. In response to the Qualcomm comment, it would be possible to greatly simplify the filtering. For the different numerology, the band edge effects could minimize the filtering complexity. In response to ZTE, the foregoing was relvant. In response to Ericsson, LTE could deliver greater performance at reasonable cost (complexity). Concerning uplink blocking, offline discussions were needed. Intel wondered if the intention of the proposal was to standardize the windowing techniques. Huawei was trying to achieve appropriate specification, reflecting the current state of the art technology. Referring to figures 4 & 5, there was no real difference, leading to observation 3. Qualcomm believed that table 1 did not show a fair comparison: multiplication did not make sense: shifters and adders were used. Huawei was trying to compare simplified filterling with LTE fillterning.
 		noted No    
    R4‑1700133 Evaluation on the delay overhead of NR waveforms Huawei, Hisilicon discussion Discussion Yes
YesWith time domain truncation, f-OFDM could achieve the same delay overhead as W-OFDM (e.g. 5% OFDM symbol duration without CP), but the link performance degradation in terms of PSD and BLER was negligible. ZTE wondered whether the above observation implied truncation. Huawei believed that this was not an issue. Qualcom wondered why 5% had been chosen. Huawei responded that this was a reasonable figure for reasonalble comparisons.
 		noted No    
    R4‑1700168 Evaluation results for maximum spectum utilization Samsung electronics co., LTD discussion Discussion Yes
YesSamsung recalled discussions at the previous meeting on the topic and analysed spectral utilization. The document reached the following conclusions: Observation 1: For maximum spectral utilization Y(%), it is observed that 93.6%~97.2% is feasible when we consider candidate combinations of system BW and SCSs. Observation 2: For ACLR, it is observed that 52~54dB ACLR is feasible when we consider candidate combinations of system BW and SCSs. Observation 3: For EVM, it is observed that EVM values of all combinations satisfy the requirements up to 256QAM. Observation 4: For spectral efficiency, it is observed that fraction PRB has better spectral efficiency compared to integer PRB. For example, 60kHz 6.5PRB has about 8% gains in terms of spectral efficiency compared to 6 PRB case. Qualcomm noted the assumptions on the rx filter, and wondered if different filters were required for different UL configurations. Samsunt replied that this was for DL only, but agreed that maybe different filters were needed at the rx side. Qualcomm was concerned about the added compliexity this implied.
 		noted No    
    R4‑1700169 Evaluation results of Guard band in case of mixed numerology Samsung electronics co., LTD discussion Discussion Yes
YesThe document was noted without presentation.
 		noted No    
    R4‑1700216 On spectrum utilization at the transmitter Ericsson other Discussion Yes
YesEricsson indicated that this paper was idealistic in its approach (other Ericsson documents used more realistic approaches). The maximum theoretical gain from increasing spectral utilization depended on the SINR operating point. The bandwidth utilization did not seem to make a large difference at low SINR. The gain of increased bandwidth utilization at high SINR depended on the underlying EVM of the system (and the supported MCS set). Huawei noted the idealistic evaluation. In a real system, even higher performance gain might be achieved with increasing PRB usage. Ericsson noted that performance gain was not proportional to PRB usage. They noted the extra signalling overhead involved in increasing the number of PRBs. Intel wondered why the gain was a function of the SINR, and also asked about the effect of impairments of real conditions. In response, Ericsson thought that low SNIRs would be similar, but higher SINR situations needed further evaluation. Vodafone thought that with current technology, there was little to be gained with these techniques, but why was it necessary to limit the capabilitiies of future technologies? In some deployment scenarios, blocking might not be an issue. What was important was to maximize the use of expensive spectrum. The specs should not limit the exploitation of evolving technology. Ericsson responded: what were the situations where blocking was not an issue? The Chairman warned about extending these discussions indefinitely. It was desirable to capture all factors in the way forward.
 		noted No    
    R4‑1700220 Spectrum Utilization link level simulations Ericsson other Discussion Yes
YesThe document was noted without presentation.
 		noted No    
    R4‑1700236 Guard band need due to OOBE caused by PA nonlinearity ZTE Microelectronics Tech. Co. discussion Discussion Yes
YesLATE DOCUMENT
 		noted No    
    R4‑1700258 WF on NR spectrum utilization and guard band ZTE, ZTE Microelectronics, Nokia, Alcatel-Lucent Shanghai Bell, Samsung, Ericsson other discussion Yes
YesZTE presented this way-forward document, concentrating on the most important elements. DOCOMO was concerned with the implications of slide Way Forward 1. If a range were introduced, how many PRBs would be specified … On the fourth bullet, the need for a guard band would be decided by the network so the fourth bullet was not needed. Vodafone had suggested changes off line. Some wording improvement could be considered. Intel was also concerned about the same point. It should be clarified that this was just an example. Concerning Way Forward slide 3, some further improvements were needed on the third main bullet. The word "should" was perhaps too strong. Qualcomm believed that the use case was useful (Way Forward slide 2). Slide WF 3 was too strong, RAN4 had not agreed on recommendations. Ericsson was concerned with the list of source organizations.
 		revised No R4‑1700297  
    R4‑1700297 WF on NR spectrum utilization and guard band ZTE, ZTE Microelectronics, Nokia, Alcatel-Lucent Shanghai Bell, Samsung, Ericsson other discussion Yes
YesIntel was not content.
 		noted No   R4‑1700258
3.4.2 In-band requirements [FS_NR_newRAT] R4‑1700052 Assumptions for NR in-band emission, EVM and in-band selectivity requirements with different numerologies Nokia, Alcatel-Lucent Shanghai Bell discussion Approval Yes
YesNokia's earlier contribution had proposed that narrow bandwidth mMTC and wider bandwidth eMBB sub-blocks with different numerologies would be used as basis for developing DL and UL in-band requirements for supporting frequency domain multiplexing of different numerologies on the same NR carrier. This proposal was not accepted but instead different service multiplexing options were considered to be possible. Proposal 1: When developing NR DL and UL in-band emission, EVM and in-band selectivity requirements for BS and UE with mixed numerologies within one NR carrier, good spectrum usage is considered. Proposal 2: Measure EVM over [1-4] PRBs in addition to EVM measurement over the whole sub-block in order to ensure good EVM performance on edge subcarriers and over the whole sub-block. Proposal 3: Define two different sub-blocks within one NR carrier for developing the first phase DL and UL in-band requirements for supporting frequency domain multiplexing of mixed numerologies Proposal 4: Use guard band between sub-blocks with different numerologies in the RAN4 requirements Observation 1: Both BW of sub-blocks and allocated PRBs per UE need to be decided for developing the mixed numerology in-band requirements. On proposal 2 Huawei suggested the use of 4 PRBs. On proposal 4, this was not acceptable, it was not necessay to specify it (it was an implementation option). ZTE supported proposal 4. Ericsson believed the guard band was implementation dependent, there were benefits to both wider and narrower guard bands. DOCOMO has the same concerns over proposal 4 as Huawei. Nokia agreed that the guard band issue was an implementation option, but it was important that RAN4 specified minimum performance requirements. Concerning the DOCOMO comment, further investigations were necessary. On proposal 4, Qualcomm noted the UE requirement for tx and rx, so the guard band provision needed to be compatible with these. The Chairman concluded that proposals 1 & 3 were agreed.
 		noted No    
    R4‑1700053 NR UL in-band emissions and EVM requirements at UE TX Nokia, Alcatel-Lucent Shanghai Bell discussion Approval Yes
YesNokia recalled the discussion at the previous meeting. This document continued the discussion. The document concluded that, in order to enable better UL coverage for CP-OFDM waveform, RAN4 should study a possibility to define two sets of in-band emissions and EVM requirements for NR UE Tx; more and less stringent requirements. Huawei supported the intention of the document, and agreed to study requirements in RAN4. Qualcomm was not against the study, and the way forward at this stage would include two sets of requirements. The Chairman noted that there was support for the document.
 		approved No    
    R4‑1700054 NR UL selectivity requirements at BS Rx Nokia, Alcatel-Lucent Shanghai Bell discussion Approval Yes
YesNokia recalled the previous discussions on the topic. The differences from the LTE case were identified. Proposal 1: Draft requirement definition of NR BS in-channel selectivity requirements with mixed numerologies: In-channel selectivity (ICS) with mixed different numerologies is a measure of the receiver ability to receive a wanted signal at its assigned resource block locations in the presence of an interfering signal using different numerology received at a larger power spectral density. In this condition a throughput requirement shall be met for a specified reference measurement channel. The wanted signal and interfering signal shall be as specified in Annex TBD. The interfering signal shall be time aligned with the wanted signal” Proposal 2: The minimum requirement for NR BS in-channel selectivity requirements with mixed numerologies: For NR, the throughput shall be = TBD % of the maximum throughput of the reference measurement channel as specified in Annex TBD with parameters specified in Table TBD for NR TBD type BS. Huawai supported proposal 1, but believed that in proposal 2, the selectivity requirements should not be over specified. Ericsson believed that the trade off between guard band width and filtering should be left as an implementation matter. The Chairman noted that proposal 1 was agreed.
 		noted No    
    R4‑1700055 NR DL in-band emission and EVM requirements at BS TX Nokia, Alcatel-Lucent Shanghai Bell discussion Approval Yes
YesNokia recalled that the topic had been under discussion for several meetings. The document offered four proposals: Proposal 1: Start developing both 5G NR BS Tx EVM and in-band emission core requirements for the mixed numerology case. Proposal 2: Investigate whether the number of test cases and testing time could be reduced by utilizing only EVM requirements for BS Tx in-band requirements in the mixed numerology case. Proposal 3: In order to ensure acceptable DL 256 QAM performance, the guard band between sub-blocks is needed in the RAN4 requirements. Proposal 4: Agree common RAN4 simulation assumptions and scenario[s?] to identify a suitable guard band for in-band requirements with mixed numerology. Qualcomm wondered if any rx filter had been used in the simulations. Huawei agreed with proposal 1. But for proposal 2 it was too early to decide. For proposal 3, this should be left as a scheduling decision. Nokia cited the different methods to be analysed. Qualcomm supposed that the requirements were only on the tx side, but again asked about rx filtering. Nokia stated that no filtering had been used in the simulation. The Chairman concluded that proposal 1 was agreed.
 		noted No    
    R4‑1700056 NR DL selectivity requirements at UE RX Nokia, Alcatel-Lucent Shanghai Bell discussion Approval No
No
 		withdrawn Yes    
    R4‑1700135 Further consideration on in-band requirements for NR Huawei, Hisilicon other Approval Yes
YesHuawey introduced the document, which made the following observations and proposals: Observation 1: Spectrum efficiency with less than 1 PRB guard band is similar, and obviously higher than the 1PRB guard band case. Observation 2: It is feasible to adopt a unified in-band emission requirements framework for both LTE UE in-band emission and NR UL in-band emission for mixed numerologies. Proposal 1: RAN4 should not specify fixed guard band between sub-bands with different numerologies, but leave it as a network scheduling decision. Proposal 2: The guard band should have a granularity of PRB, and the size of guard band is up to scheduling decision. Proposal 3: Reuse LTE UE in-band emission requirement definition methodology for NR uplink in-band emission requirements for mixed numerologies. Proposal 4: NR in-band selectivity for mixed numerologies adopts the similar format as LTE in-channel selectivity (ICS) requirements, taking additional constraints into consideration, including 1) Target bandwidth, 2) MCS level on target bandwidth, 3) Power imbalance between sub-bands. ZTE believed that it was too early to agree to proposals 1 and 2 and 3 was probably also acceptable.. Proposal 4 needed more study. Ericsson noted the potential to use wider guard bands (for example) to improve the situation, and sought clarification on the emissions situation. On proposal 4, MediaTek wanted it clarified that it related to the rx sensitivity, and Huawei clarified it was for both UL and DL. MediaTek wondered if different numerologies could come from the same BS, and if so how power imbalance could be achieved. Huawei responded that this was not a concern where a single BS was involved. Ericsson could potentially agree proposals 3 & 4 but wanted further off line discussions.
 		noted No    
    R4‑1700217 On multiplexing of numerologies Ericsson other Discussion Yes
YesEricsson claimed that there were different strategies possible, each having different trade offs. Thus it was not clear which numerology should be prioritized. Any implementation should be allowed. Huawei referred to fig. 4, suggesting that that the preconditions influenced the results. BS side filtering should be an implementation matter. The general principles were acceptable.
 		noted No    
    R4‑1700266 Way forward on in-band requirements for NR Nokia other Approval Yes
YesNokia had captured the agreements of discussions, as mentioned in the listed tdocs. Huawei did not see anything new compared to the last meeting. However, they were not happy with the second bullet of the "issues for the next meeting": this had not been agreed. Nokia pointed out the study would be "with / without" guard bands. ZTE was happy, but would like to remove the reference to the guard band between diferent numerologies to BS scheduling. Huawei wished for more extensive excisions, but Ericsson felt it was important to retain the "with/without" aspect. On slde 4, Qualcomm were not entirely happy with this wording, further study on this would be necessary. Huawei understood the intent of slide 5, but felt that the sentiments should be extended to other UE requirements, e.g. tx power. How was it possible to draw conclusions. Nokia recalled that this slide had been based on tdoc 204 (Skyworks). Huawei believed that a separate place holder should be provided for this. There were further discussions, with arguments from Ericsson and Skyworks: what it was critical to capture was that there would be an effect of multiple numerology on many system aspects. The Chairman concluded that this aspect was wider than in-band emission, and the further discussions would take place under an appropiate agenda item at the next meeting. On slide 6, Qualcomm presumed that the ICS related to the BS, which was a little confusing, given the slide's title. Nokia indicated that the last bullet point explained the situation. On slide 7, Huawei did not understand the second bullet. Ericsson responded that the requirements needed to be desired so that the BS could select the guard band and choose an appropriate filter. Huawei believed this was already covered in the assumptions, so there was no need to single out this aspect on this slide. Ericsson insisted that the aspect be implement agnostic, and did not want to lose that meaning in any rewordiing. Huawei was still concerned with the proposed wording, and further reviision was made, including to the agreements/assumptions slide. Similar arguments applied to slide 8. Qualcom remarked that UE receiver impacts should also be considered. This led to further discussion and explanations.
 		revised No R4‑1700290  
    R4‑1700290 Way forward on in-band requirements for NR Nokia other Approval Yes
YesNokia described the changes with respect to the original document. Qualcomm wanted to retain the bullet concerning the impact on the UE receiver. This was an important issue. Huawei needed more time to study this issue. He proposed that Qualcomm bring a contribution at the next meeting. Further discussion on the point between these companies ensued. But no progress was possible.
 		noted No   R4‑1700266
3.4.3 Others [FS_NR_newRAT] R4‑1700121 Consideration on ECC requirements in 3400-3800MHz Huawei, Hisilicon discussion Discussion Yes
YesHuawei stated that this contribution brought further proposals for requirements.
 		noted No    
    R4‑1700122 TP for 38.803: ECC requirements in 3400-3800MHz Huawei, Hisilicon pCR Approval Yes
Yes
 		approved No    
    R4‑1700123 WF on evaluation of ECC requirements in 3400-3800MHz Huawei, Hisilicon other Approval Yes
YesThe document was presented by Huawei, based on the discussions at the presenet meeting. Qualcomm noted that this had not been discussed off line and requested more time to review it. Ericsson wondered what would be the follow up action on this document. Huawei responded that RAN4 needed to evaluate the requirements. Nokia wanted to know the justification for the values appearing against the various parameters. Huawei had taken appropriate existing vallues from 37.840 & 37.842, noting that these values were relative, not absolute. Nokia did not want to show specific numbers. After off line checking, the document was agreeable.
 		approved No    
    R4‑1700209 Guard period for NR Ericsson discussion Discussion Yes
YesOriginal agenda item 3.6.1. Ericssoon introduced the document describing the RDD timing budget and corresponding formulae, concluding that transient times for BS • ˜ 6 µs subcarrier spacing 15 to 30 kHz. • ˜ 3 µs for subcarrier spacing 60 kHz • ˜ 1 µs for subcarrier spacing 120 kHz and an eNB-eNB synchronization of • TSync of 3 µs = ±1.5 µs. provided a balanced set of requirements used as a starting point for further discussion. Nokia thought smaller transient times below 6 GHz with larger cell sizes may need more investitation Huawei had contributions on this topic, and had some concerns over Ericsson's figures. ZTE had also analysed the BS transient times and had a contribution on the subject. They too arrived at different figures. CATT agreed with Huawei and ZTE. Such short transient times were not feasible due to the RF components. Ericsson replied to Nokia that it was correct that the spectrum dimension had not been clearly described. One technical reason to move to wider carrier spacing (not discussed in this spacing) was the idea of moving to wider subcarrier spacing to deal with phase noise from Los and for other reasons. To Huawei's remarks, yes there had been several document not presented, and it was noted that the Ericsson and Huawei formulae were identical, but there were some different assumptions over cell sizes. The Ericsson approach had been to keep overhead constant. To ZTE, Ericsson thought that ultimately this requirement was hardware dependent, and the BS had to harmonize with what the UE could handle. To CATT, Ericsson noted their position. The Chairman urged UE vendors also to perform feasibility studies on the topic.
 		noted No    
    R4‑1700208 TP for NR Rel-14 TR 38.803: TDD timing budget Ericsson pCR Approval Yes
YesThe Chairman stated that this document would not be treated because it was not specific to NR. It was too general.
 		revised No R4‑1700269  
    R4‑1700269 TP for NR Rel-14 TR 38.803: TDD timing budget Ericsson, ZTE pCR Approval Yes
YesThe pCR was presented by Ericsson. Huawei did not think there was sufficient time to check the contents at the present meeting. Ericsson said that the changes since the last version were negligible, and it was feasible to approve it.
 		noted No   R4‑1700208
    R4‑1700038 Observation on 5G NR TDD ON-OFF Switching Time Budget ZTE Corporation discussion Discussion Yes
Yes
 		noted No    
    R4‑1700126 TP for 38.803: NR timing budget Huawei, Hisilicon pCR Approval Yes
Yes
 		noted No    
    R4‑1700200 Derivation of Reference Sensitivity in NR above 6GHz NEC discussion Discussion Yes
YesOriginal agenda item 3.4.2. NEC presented the document, which examined many parameters (thermal noise, phase noise, non-linear distortion, internal distortion, industrial margin) affecting the reference sensitivity level. Intel asked whether the equation included a factor on beam forming. NEC replied that the figures were for conducted only and therefore did not cover beam forming. MediaTek asked why the numbers for industrial margin differed for each frequency band. NEC stated that these had been provided by one of the referenced documents. Ericsson stated that the reference sensitivity had to be differenciated from minimum sensitivity.
 		noted No    
    R4‑1700218 Considerations for deriving a blocking requirement for NR Ericsson other Discussion Yes
YesEricsson introduced the document and recalled how the blocking level was derived. NR might make some difference compared to LTE. Blocking experienced after beam forming would be very high but very improbable, and without beam forming, blocking would be relatively low but relatively probable. Different beam shapes might influence multi-path and fading effects, with corresponding effects on blocking. Huawei mentioned that they had a relevant contribution elsewhere in the agenda. But concerning the present document, they warned that any deviation from today's method might be dangerous. Techniques above 6 GHz would differ from the eAAS techniques below that frequency. Intel asked why a value of 6 dB had been chosen. Ericsson replied that this was a value for illustration.
 		noted No    
    R4‑1700075 TP for TR 38.803: Re-use of requirements below 6 GHz Ericsson pCR Approval Yes
YesThe Chairman asked Ericsson to explain the delta compared to the previous meeting. Ericsson indicated that it was the same, but had not received sufficient attention last time. But the tables had been updated with other companies' views. DOCOMO thought the information in the tables went a little too far. Nokia warned against stating all requirement, some were identical but some had changed, and some were new.
 		noted No    
    R4‑1700222 TP for 38.803: Editorial correction to PA models Ericsson pCR Approval Yes
YesOriginal agenda item 3.2.2.
 		approved No    
    R4‑1700088 TP for NR Rel-14 TR 38.803: LO generation and phase noise aspects for mm-wave technologies Ericsson pCR Approval Yes
YesEricsson had had offline discussions with other companies.
 		noted No    
    R4‑1700089 TP for NR Rel-14 TR 38.803: PA considerations for mm-wave technologies Ericsson pCR Approval Yes
Yes
 		noted No    
    R4‑1700090 TP for NR Rel-14 TR 38.803: Noise figure for mm-wave technologies Ericsson pCR Approval Yes
Yes
 		noted No    
    R4‑1700091 TP for NR Rel-14 TR 38.803: Carrier frequency and mm-wave technology aspects Ericsson pCR Approval Yes
Yes
 		noted No    
    R4‑1700092 TP for NR Rel-14 TR 38.803: Filtering aspects in mm-wave technology Ericsson pCR Approval Yes
Yes
 		noted No    
    R4‑1700279 Way forward on UE and BS NF for mm-waves Ericsson other Approval Yes
YesEriicsson presented the results of the disussions and agreement arrived at off line. Telecom Italia highlighted that further discussion would be needed on these values, but there was a wide consensus on lower values. Nevertheless, these values were acceptable for the reply to ITU-R. Verizon wished for the notes to be captured in the meeting report: These NF values shall be used only for WP5D response. Further study on the actual noise figure to be used to define RF requirements for UE and BS shall be performed in the WI phase.
 		approved No    
3.5 UE RF [FS_NR_newRAT] R4‑1700182 SI status of UE RF aspects NTT DOCOMO, INC. other Approval Yes
YesNTT DOCOMO presented the document. Observation 1: It needs to be clarified which one(s) of each fundamental factor specified in physical layer (e.g., modulation, sub-carrier spacing) is(are) assumed for the initial NR specs in advance in order to complete the Rel-15 WI [in a timely manner]. Observation 2: On top of observation 1, parameters (e.g., NF, SNDR) to specify each RF requirement with consideration of the feasibility should be discussed in parallel. Further discussion was necessary based on the table in the documents. Huawei agreed with the first observation. The discussion lacked important assumptions. For carrier leakage, they were not convinced the wafeform was correct because the physical layer was different below 6 GHz. How was performance to be tested? For the occupied bandwidth, ERP might be a better measure. Ericsson agreed that some further study was needed, but no further investigation was needed for other parameters. The Chairman asked what was the next move from NTT DOCOMO. NTT DOCOMO replied that they would modify the table and then capture it in the TR at the next meeting. The Chairman indicated that this table was necessary to make known the result of the RAN study.
 		revised No R4‑1700259  
    R4‑1700259 SI status of UE RF aspects NTT DOCOMO, INC. other Approval No
YesDocument not available.
 		withdrawn Yes   R4‑1700182
    R4‑1700264 WF on UE RF Qualcomm other discussion Yes
YesThe way-forward was presented by Ericsson. Slide 4: DOCOMO thought some starting conditions were needed, but were still under discussion. Ericsson proposed a minor wording change. Slide 6: Huawei asked for clarification on the -40 dB figure. ZTE believed that the -50 dBm power off level needed to be reconsidered.
 		revised No R4‑1700300  
    R4‑1700300 WF on UE RF Qualcomm other discussion Yes
Yes
 		approved No   R4‑1700264
3.5.1 Reference architecture [FS_NR_newRAT] R4‑1700007 UE reference architecture for NR Sony Mobile Communications discussion Discussion Yes
YesSony presented the document, which covered antenna arrangements, beam forming, and filters. Qualcomm observed that the insertion loss of about 5 dB coming from two filters, and what was the effect on noise? Sony agreed, and the noise figure of 10 dB had been used, though this was something of a guess. Qualcomm noted that the physical size of those filters was also significant. Huawei wanted other companies' views on the need for filters. Intel asked how to have so many filter arrays in the UE, considering physical space. Sony respondid that this was as yet unknown. Concerning the question of how the calculations had been made, it had been assumed that each element had 3 dB gain. And the UE EIRP value of 27 dBm took account of conducted signal plus beam forming. MediaTek wonderd if the UE conducted power assumed an integrated PA or an external PA. Sony believed this was an integrated one.
 		noted No    
    R4‑1700010 Discussion on reference architecture and performance for NR UE below 6GHz Skyworks Solutions Inc., Sprint discussion Approval Yes
YesSkyworks extended the discussion to a consideration of baseline performance. The available bands were listed (tables 1, 2). They believed that NR was only useful if it was better than LTE. Observation 1: • Most regions have a 5G 3.5GHz spectrum consistent with or slightly extending the B42/B43/B48 LTE cluster. • All bands are bounded by LTE bands 7, 41 bellow and 46 above • 3.5GHz bands have sufficient separation from 2.7GHz bands (>600MHz, 20% frequency) • "JP4400" and "CH4800" have small separation to band 46 (<250MHz, 5% frequency) Proposal 1: NR UE below 6GHz reference spectrum range • 1a: NR and LTE bands between 2.3GHz and 6GHz shall be the focus for NR <6GHz high data rate UE solution • 1b: The 3.4GHz to 3.8GHz LTE band cluster grouping band 42, 43 and 48 can be used as reference for NR architecture and performance for high data rate services <6GHz Proposal 2: LTE Band and CA, NR anchor band support for NR UE in the 2.3GHz to 6GHz range • 2a: NR NSA added UE functionality between 2.3GHz to 6GHz shall allow any LTE anchor band <2.7GHz • 2b: NR NSA added UE functionality between 2.3GHz to 6GHz shall allow support of legacy LTE bands in this frequency range including their associated UL/DL carrier aggregation cases. Proposal 3: Antennas for NR UE in the 2.3GHz to 6GHz range • 4 antennas shall be assumed for NR UE in the 2.3GHz to 6GHz range and at least 4x4 MIMO supported. • Same 4 antennas will be shared between NR and LTE bands supporting 4x4 DL MIMO. Proposal 4: Receiver baseline performance for NR UE in the 2.3GHz to 6GHz range • 4x4 DL MIMO • DL 256QAM capable o Potential reference sensitivity benefits is FFS • 80MHz minimum aggregated DL bandwidth • -20dBm range 3 out of band blocker Proposal 5: Transmitter baseline performance for NR UE in the 2.3GHz to 6GHz range • At least two active UL antennas enabling antenna selection, MIMO and beamforming operation • 4 antenna UL is FFS • UL 256QAM capable o Potential In-band and out of band emission benefits is FFS • 40MHz minimum aggregated UL bandwidth o Minimum carrier allocation and related 0dB MPR is FFS • HPUE 26dBm class for <50% TDD duty cycle o Duty cycle dependent MPR or PCmax is FFS • Removed UTRA ACLRs in the 2.3GHz to 6GHz range Huawei asked Skyworks to comment on how to improve NR rx sensitivity for the UE. On proposal 3, MediaTek asked why four antennas had been assumed. Were all values based on this configuration? DOCOMO asked for further study on the proposals. In response, Skyworks observed that today's tx performance was much better than the minimum requirements. To arrive at a target performance, it was appropraite to start from the current LTE values. The intention was not to allocate particular bands to NR, but just to identify bands which were good candidates in terms of available bandwidth. Ericsson agreed with the general gist of the document, but warned agaiinst copying too many of the LTE requirements, since some of them were not optimal (dating back to Rel-8). Skyworks noted that each generation of digital radio had broought improvements, and this need should influence the starting point for NR.
 		noted No    
    R4‑1700069 TP for 38.803: UE beamforming and number of UE transmitter antennas Ericsson, Sony pCR Approval Yes
YesEricsson explained that this was an update of a document seen at the Reno meeting. Huawei considered that the disposition of antennas on the UE might not be optimal (Figure 6.2.1.X.2-1: antenna arrangement on the UE with groups of mmW antennas). Ericsson emphassed that this was only a typical example.
 		approved No    
3.5.2 EIRP/EIS directional requirements [FS_NR_newRAT] R4‑1700080 On EIS EIRP direction requirement in mmWave UE RF Intel Corporation other Discussion No
YesLATE DOCUMENT Document not available.
 		withdrawn Yes    
    R4‑1700095 Discussion of mmWave UE EIRP and EIS test Huawei, Hisilicon other Approval Yes
YesHuawei explained that this was an update of that seen at the last meeting. Qualcomm thought this was an interesting paper, but believed there was another important discussion on beam reciprocity. The question on measurement points was very important, and Qualcomm also had a paper on this. The UE coordinates discussion was similar to Qualcomm's but had some doubts about the UV system. Ericsson were not quite sure about the reciprocity aspect.
 		noted No    
    R4‑1700100 Using CDF to define EIRP and EIS Qualcomm Incorporated other Approval Yes
YesQualcomm's paper described how to acquire CDF from points defined on a spherical coordinate system, the number and placement of those points being chosen to ensure appropriate test results. Proposal 1: Testpoints for EIRP and EIS should be spaced with equal density around sphere surface Proposal 2: Device EIRP and EIS requirement for different directions will be defined with CDF. Details what percentile achieves what value is FFS Proposal 3: Number of testpoints for UE EIRP and EIS should not exceed 47 for full sphere. Skyworks supported the need to minimise the number of tests. But did it mean tht the requirement would be based on a fairly small array? Intel believed a bit of surface integration should be used to establish the points, but the geometry needed to be adpted to the approach taken. The maths for forming the CDF needed to be appropriate. Huawei was concerned that the DCF curve should be constant for the different approaches. They understood that the number of test points was a function of the antenna configuration. If the beam was very narrow, more test points might be needed. But the number should be minimised. Qualcomm responded on the number of test points. He agreed it was a function of the number of beams. The nuimber of test points should indeed be minimised, and how to choose those points was contained in this document, although the (quite simple) maths had been omitted. Qualcomm asked why the number of points would impact the accuracy of the IERP measurement, as had been suggested by Intel. Intel believed the number of points would affect the uncertainty of the measurements. DOCOMO noted some side conditions.
 		noted No    
    R4‑1700197 Further consideration on EIRP/EIS directional requirements DOCOMO Communications Lab. other   Yes
YesDOCOMO's contribution provided further consideration of the definition of the EIRP/EIS requirement. And then discussed how to develop the EIRP/EIS requirement in WI phase. Observation 1: CDF method with side conditions in Proposal 1 and 2 can accommodate requirements for wide variety of UE types. Proposal 1: Following two side conditions X and Y for EIRP/EIS directional requirement. ? X: Special coverage in which EIRP/EIS requirement should be covered. ? Y: {Y1, Y2, ... , Yy}, Yn is one part of X when X is divided into y equal parts. (n = 1, 2, …, y) Proposal 2: With the side conditions in proposal 1, each of Yn should have at least one direction in which measured EIRP/EIS performance is higher than the required value. Observation 2: Since specific special coverage is a part of full sphere, EIRP/EIS requirement for full sphere can be baseline for that for specific special coverage. Proposal 3: Discuss and specify EIRP/EIS requirement for full sphere as priority in WI phase. Proposal 4: After EIRP/EIS requirement for full sphere is understood, then that for other UE types should be discussed if needed. Huawei supported having one or several UE types to define the Rel-15 requirements. The objective should be to minimise the test time (and cost). Qualcomm supported proposals 3 & 4. But for proposals 1 & 2, these were very generic; how could this be defined as a three dimensional problem?
 		noted No    
    R4‑1700154 On UE types with different spatial coverage Sumitomo Elec. Industries, Ltd other   Yes
YesSumitomo presented the document concerning the definition of different UE types. Proposal 1 The following UE types should be taken into account when defining mmWave NR UE categories. • Smart phone (which has been agreed as the baseline) • Laptop mounted equipment (such as plug-in devices like USB dongles) • Laptop embedded equipment • Tablet embedded equipment • Virtual Reality Glasses • Wearable devices like smart watch • Vehicular mounted device • Fixed Wireless Access (FWA) terminal Proposal 2 Multiple UE categories should be defined with different spatial coverage requirement Proposal 3 Categorize the UE types based on whether a fixed portion of the UE’s radiation sphere blocked when transmitting/receiving Huawei wanted a prioritization of which types to examine in Rel-15. ZTE agreed to have different categories. But the differences were often a function of antenna placement (eg vehicle mounted UE might have characteristics which varied from vehicle to vehicle. Sumimoto understood the comments.
 		noted No    
3.5.3 NSA device IDC study [FS_NR_newRAT] R4‑1700096 Preliminary IDC study for NSA device Huawei, Hisilicon discussion Discussion Yes
YesHuawei stated that this contribution had been discussed off line. It would be premature to reach any conclusions at this stage. DOCOMO noted deviation amoongst the results of different companies. Further discussion was needed. Skyworks explained the question was how much attenuation was experienced at a given frequency, and this should be addressed in the way forward.
 		noted No    
    R4‑1700177 Necessity of power sharing between NR and LTE in NSA operation NTT DOCOMO, INC. other Approval Yes
YesNTT DOCOMO presented the document, explaining that this topic had not been discussed before. Proposal 1: It should be studied if there is any justification to assume power sharing between NR and LTE in NSA operation from regulatory and/or SAR point of view in each region/country (other aspects are not precluded). If any justification is identified, dependent power control (namely, Option 1) should be adopted and how to calculate and test the total power should be studied. If no justification is identified, independent power control (namely, Option 2) should be adopted. Proposal 2: An LS to inform RAN1 the outcome of the RAN4 study should be sent in future meeting.
 		noted No    
    R4‑1700012 Further discussion on coexistence between mmW NR and sub-6GHz NR and LTE Skyworks Solutions Inc. discussion Approval Yes
YesOriginal agenda item 3.2.2. Skyworkds introduced the document. It was important to be aware that 2nd and 3rd harmonic would also be beam-formed. Harmonics above the 5th could be ignored. There was no critical problem over cooexistance. Proposal 1: • Since no direct harmonic emissions nor spurious harmonic responses are foreseen with the currently known 5G spectrum, it is proposed that general harmonic emissions requirement is used; and • TX and RX harmonic related coexistence issues above 6GHz is FFS depending on potential issues related to new spectrum allocation. Proposal 2: Since the sub-6GHz RF front-end provides important 28GHz selectivity, extra 28GHz selectivity can be added after the LNA and accounting for path losses, it is proposed that no Reference sensitivity degradation of the sub-6GHz bands need to be studied in relation to spurious harmonic responses to mm-wave signals in the following coexistence cases: • In-device concurrent sub-6GHz LTE or NR and above 24GHz NR operation • Above 24GHz NR UE to sub-6GHz LTE or NR UE coexistence • Above 24GHz NR BS to sub-6GHz LTE or NR UE coexistence Proposal 3: Given the very low level of interfering harmonic and out-of-band noise observed, it is proposed that no Reference sensitivity degradation of the above 24GHz NR bands need to be studied in relation to spurious emissions from sub-6GHz LTE or NR radios in the following coexistence cases: • In-device concurrent sub-6GHz LTE or NR and above 24GHz NR operation • Sub-6GHz NR or LTE UE to above 24GHz NR UE coexistence • Sub-6GHz NR or LTE UE to above 24GHz NR BS coexistence DOCOMO supported proposals 1 & 2. The Chairman believed this was very important information, if no sensitivity degradation was observed. The proposal needed to mention the agreed -13 dB figure.
 		noted No    
    R4‑1700180 Further studies between sub-6GHz and mmWave bands NTT DOCOMO, INC. other Approval Yes
YesDOCOMO indicated that the document had been based on Skyworks contribution and aimed to clarify the scope for the work item phase. Proposal 1: In-device and UE-to-UE coexistence between licensed sub-6GHz and mmWave bands should be studied first. Proposal 2: Once reference sensitivity in mmWave bands for NR SA is derived, it will also apply for ones for NR NSA without MSD. Proposal 3: -50 dBm/MHz will be used for UE-to-UE coexistence from sub-6GHz to mmWave as a default value (and vice versa). Proposal 4: Existing reference sensitivity in licensed sub-6GHz bands can be reused without MSD for NR NSA cases. (Note: Band 43 is FFS as below) Proposal 5: In case a band plan proposed in [4] is agreed, only 7th harmonic spurious in Band 43 should be investigated and other licensed band cases don’t have to be studied (i.e., Existing reference sensitivity in licensed sub-6GHz bands can be reused without MSD) in the first NR WI phase. Huawei asked which frequency was used to arrive at the -50 dBm figure. Skyworks stated that the only distinction was between below and above 6 GHz, at least for the moment.
 		noted No    
    R4‑1700270 WF on NSA operation between sub-6GHz and mmWave bands NTT DOCOMO, INC., Skyworks, Qorvo other Approval Yes
YesThe way-forward was presented by NTT DOCOMO. Qualcomm wanted more time to check. Sprint wished for more explanation on slide 3. Band 41 was a candidate for NR. The Chairman indicated that the scope was for sub-6 GHz range. Skyworks observed that the idea had been to study in a generic way the sub-6 GHz range and the above (but < 24 GHz). Sprint was worried that this scenario might artificially restrict the studies. This work needed to be prioritized over other work. Qualcomm thought sub-6 GHz was business as usual, but in mmWave there would probably be new aspects to investigate. ETISALAT was concerned with bands between 6 & 24 GHz, and the concentration would be on licensed bands (not unlicensed).
 		revised No R4‑1700296  
    R4‑1700296 WF on NSA operation between sub-6GHz and mmWave bands NTT DOCOMO, INC., Skyworks, Qorvo other Approval Yes
YesNTT DOCOMO described the changes made for this version.
 		approved No   R4‑1700270
3.5.4 Transmitter characteristics [FS_NR_newRAT] R4‑1700227 More on ACLR with beamforming MediaTek Inc. other Discussion Yes
YesOriginal agenda item 3.4.2. MediaTek intended to do further measurements on ACLR with beamforming. The maths showed that the IM3 would be beamformed as for the main frequency. For higher order IM products, these too would be beamformed in the same direction. The CIM3 would be phase (time) shifted with respect to the main tone. Table 1 summarized the findings. Qualcomm said the CIM3 was beamformed but not in the same direction as the main tone. Skyworks understood that baseband CIM3 appears after phase shift applied at baseband. Qualcomm believed that the issue was insignificant at 100 MHz. MediaTek responded that the CIM3 would be formed but not in the same direction, and its magnitude would be less than the IM3. The phase shift applied in the time domain relative to the frequency shift would be multiplied by three, and would therefore not be in the same directon. MediaTek agreed that the squinting effect would not be apparent at 100 MHz. The intention was not to define a more stringent ACLR.
 		noted No    
    R4‑1700070 On TRP testing for verification of UE unwanted emissions Ericsson discussion Discussion Yes
YesEricsson presented the document, showing a practical method of measuring TRP. The methodology was to measure the reactive near field. The technique of cardinal cuts could considerably reduce the number of test points. The document also discussed the frequency grid and arrived at an appropriate compromise. A different grid might be needed for different measurements. It was noted that it was not always the harmonics which were the worst spurious emissions.
 		noted No    
    R4‑1700178 UE Tx/Rx spurious emissions in NR OTA NTT DOCOMO, INC. other Approval Yes
YesDOCOMO presented the document. Proposal 1: Both Tx and Rx spurious emissions should be specified in TRP for NR OTA. Proposal 2: Upper frequency limits for Tx/Rx general spurious emissions should be specified as 2nd harmonics of the upper edge of the UL/DL operating band respectively. Proposal 3: OTA testability (e.g., power level, testing time) of Tx/Rx spurious emissions should be studied for UE NR conformance spec. Proposal 4: If it is identified that -30 dBm/MHz cannot be satisfied, trade-off between relaxation of the requirement and having additional effort (e.g., MPR, filter implementation) should be evaluated quantitatively. Proposal 5: The existing (conductive) additional spurious limits should be maintained in UE NR core spec. ZTE sought clarification: should the EIRP also be considered, and how about 6 GHz? DOCOMO responded that EIRP should indeed be included. And the upper limit would be 12 GHz.
 		noted No    
    R4‑1700143 Transmit ON/OFF power for NR UE ZTE Corporation discussion   Yes
YesThe document was noted without presentation.
 		noted No    
    R4‑1700179 UE ON/OFF time mask in NR NTT DOCOMO, INC. other Approval Yes
YesThe document was introduced by NTT DOCOMO. Proposal 1: ON/OFF time mask should be specified in TRP for NR OTA. Proposal 2: Achievable transient period in mmWave devices (e.g., 28 GHz) should be investigated. Proposal 1 was agreed. But later, Huawei proposed to have futher discussions. The Chairman asked how proposal 2 would be tackled.
 		noted No    
    R4‑1700190 Beam Correspondence for NR Qualcomm Incorporated discussion Discussion Yes
YesQualcomm introduced the document, indicating that beamforming capability signalling for Ues would probaably be needed. Huawei agreed that reciprocity should be considered, but UE testing might be difficult. There were also NodeB considerations. Qualcomm agreed there would probably need to be a similar test for the base station, but this had not been considered in the present document. It would be necessary to establish requirements prior to defining negotiation signalling.
 		noted No    
    R4‑1700039 On Tx consideration in mmWave UE RF Intel Corporation other Discussion No
YesLATE DOCUMENT The document was not provided.
 		withdrawn Yes    
3.5.5 Receiver characteristics [FS_NR_newRAT] R4‑1700017 RF filters for mm-wave OOB blocking Intel Corporation SID new Discussion Yes
YesOriginal agenda item 3.5.4.
 		revised No R4‑1700261  
    R4‑1700261 RF filters for mm-wave OOB blocking Intel Corporation SID new Discussion Yes
YesIntel presented the document. A major problem was the size of the filter at lower bands, and acoustic filters might not be sutable for higher bands. Proposal 1: Feasibility to produce UE RF filters for NR frequency bands (in millimetre wave region) must be studied further. Possibility to integrate the RF filters within UE must be prioritized. Proposal 2: In scenarios where appropriate UE RF filters are not feasible, out of band blocking for UE receiver characteristics should be defined appropriately. Skyworks wondered whether this filtering was needed in the UL, in the DL? There seemed to be no evidence requiring this. And at 28 GHz, there was no real "ground" as claimed in the figure. But narrow band filtering was not feasible. Huawei wanted clarification on the meaning of scenario 2. Intel stated that the author of the document was not present at the meeting, so only a partial response could be made.
 		noted No   R4‑1700017
    R4‑1700018 On side lobes consideration in antenna array for ACS and blocking requirement Intel Corporation other Discussion Yes
YesIntel addressed side lobes in antenna arrays of various configurations. Observation 1: Side lobe rejection is increasing according to the number of antenna array and saturated around 13dB. Observation 2: The first side lobe angle is decreasing according to the number of antenna array. Proposal 1: A blocker coming to the first side lobe direction needs to be took account in blocking and ACS requirements. Proposal 2: A small amount array antenna steering error could face a blocking situation. Proposal 3: Study incident angle for various array geometries and decide an incident angle scenario. Qualcomm thought the scenario was not a feasible array for practical use. The first side lobe coould be controlled, and so the problem seemed ill defined. DOCOMO observed that proposal 1 was less a proposal, more an observation. Intel responded to DOCOMO, the document considered only the lobe from one direction.
 		noted No    
    R4‑1700051 Discussion of mmWave blocking requirements Huawei, Hisilicon discussion   Yes
YesHuawei presented the document, examining a single antenna module with a given pattern. In 95% of the cases, the blocking direction would be that of the wanted signal. Skyworks understood that it was proposed only to measure blocking only of the wanted frequency. But some more probablistic approach was needed. Huawei responded that proposal 1 was a viable scenario. Qualcomm found the beam pattern rather confusing. Huawei replied that the beam was tilted 90 degrees, but performace was not perfect. Qualcomm thought that if this were the case, there could be greater gain in other directions. Off line discussion was needed.
 		noted No    
    R4‑1700071 Further elaboration on UE ACS behaviour Ericsson discussion Discussion Yes
YesThe document was presented by Ericsson and was a follow up to a previous document on UE ACS behaviour. For the ITU-R coexistence studies, the ACS could be assumed to be spatially flat, but there might be additional cases that had to be considered for the 3GPP specifications; the ACS needed for supporting the maximum SNIR of the MCS should be considered. A TRS metric could perhaps be used for ACS verification of a “spatially flat ACS”.
 		noted No    
    R4‑1700040 On Rx consideration in mmWave UE RF Intel Corporation other Discussion No
YesLATE CONTRIBUTION The document was not provided.
 		withdrawn Yes    
    R4‑1700008 Out-of-band blockers in the mm-Wave spectrum Korea Testing Laboratory discussion Information Yes
YesLATE CONTRIBUTION
 		noted No    
    R4‑1700084 On Rx consideration in UE RF Intel Corporation other Discussion No
YesLATE CONTRIBUTION The document was not provided.
 		withdrawn Yes    
3.5.6 Testability [FS_NR_newRAT] R4‑1700016 On UE OTA testability for NR above 6 GHz CATR other Approval Yes
YesCATR presented this short document. Proposal: OTA measurement is the baseline testing methodology for UE RF at high frequency (f > 6 GHz).
 		approved No    
    R4‑1700025 TP on general aspects of NR UE test Intel Corporation, CATR pCR Approval Yes
Yes
 		revised No R4‑1700262  
    R4‑1700026 TP on NR UE test of transmitter characteristic Intel Corporation, CATR pCR Approval Yes
YesComments had been received, and a revised draft was being prepared. Qualcomm wondered if the baseline had yet appeared in any contribution. Intel stated that most of this had indeed been seen in earlier documents.
 		merged No    
    R4‑1700027 TP on NR UE test of receiver characteristic Intel Corporation, CATR pCR Approval Yes
Yes
 		merged No    
    R4‑1700028 TP on NR UE test interface aspects Intel Corporation, CATR pCR Approval Yes
Yes
 		merged No    
    R4‑1700262 TP on general aspects of NR UE test [title to be verified] Intel Corporation, CATR pCR Approval Yes
YesIntel presented the document. Drafts had been shared on the reflector.
 		noted No   R4‑1700025
    R4‑1700029 On NR UE testability for below 6 GHz Intel Corporation, CATR discussion Discussion Yes
YesIntel introduced the contribution. Methodologies for conducted and radiated values already existed in TSs. Intel responded to a question from DOCOMO. It was important to discuss the radiated test cases. There were two cases, control above 6 GHz, test below 6GHz or both below 6 GHz.
 		noted No    
    R4‑1700030 NR UE testability terminology Intel Corporation, CATR discussion Discussion Yes
YesIntel presented the contribution, concluding that with the reference coordinate system and test conditions defined, terminology for the baseline testing setups could be developed. The figures of merit (FoM) for each requirement could be defined as a specific measured quantity given the above parameters. However, further progress on the definition of the RF requirements was needed. Rhode und Schwartz considered the document to be a very good baseline.
 		noted No    
    R4‑1700031 Way Forward on NR UE Testability Intel Corporation, CATR other Approval Yes
YesIntel presented the way-forward document. On the NSA testability slide, what did the bullets in the way forward really mean. The NSA test would be OTA and high enough SNR could be set to ensure intermod would not be a problem. NTT DOCOMO thought the testing method was still not clear. But Qaualcomm considered it was already clear that below 6 GHz would use the legacy method. But the real purpose of the test had to be reconsidered. Intel found the discussion useful, and perhaps a refocussing of the question was needed.
 		revised No R4‑1700294  
    R4‑1700294 Way Forward on NR UE Testability Intel Corporation, CATR other Approval Yes
YesOnly slide 5 (NSA testability) had been changed. Qualcomm asked for clarification of the first subbullet of the second bullet. Intel sought to clarify. Qualcomm was concerned over the control channel part. DOCOMO, as the author of this text, was asked to explain the control channel. Huawei thought a minor wording change would make it clear that the control channel was not the important aspect, but the fact that it was < 6 GHz. Further clarifications were crafted.
 		revised No R4‑1700299 R4‑1700031
    R4‑1700299 Way Forward on NR UE Testability Intel Corporation, CATR other Approval Yes
Yes
 		approved No   R4‑1700294
    R4‑1700072 Proposed near field test setup for EIRP measurement MVG Industries other Approval Yes
YesMVG introduced the document proposing a near field test setup for measuring EIRP, and outlined the advantages of the techniques proposed. At mm-wave frequencies, testing with phantoms would be needed as well as free-air tests. Proposal: Multi-probe Near Field Spherical test range can be used for EIRP and TRP measurements of User Equipment at mmWaves. We propose to consider it as a possible solution for NR UE testability. Keysight believed this was an interesting document, but wondered whether the uncertainty aspect of non-ideal antennas needed to be examined. Qualcomm noted that table 1 implied downconversion and it would be good to see how this had been achieved. For sperical measurements, the probe had to scan at a constant distance from the antenna, but was this feasible where the antenna was confined, for example, in the corner of the UE? MVG responded to Keysight, and agreed it would be good to investigate this. Respsonding to Qualcomm, it was hoped to bring more detail at the February meeting. The question on spherical testing was not clear, so Qualcomm illustrated the point. The near field could not be measured at a constant distance from the antenna because it was not known precisely where the antenna was situated in the UE. MVG believed that after correct calibration, this would not be a problem.
 		noted No    
    R4‑1700078 NR UE Test Interface Considerations ROHDE & SCHWARZ discussion Discussion Yes
YesRhode und Schwartz presented the contribution, proposing the need for a test interface to support UE RF testing, especially during development, when non-signalling tests were more suitable. A standardized test method would be highly desirable. Intel supported the proposals, and the division into three functions was useful. They sought some additional information. Qualcomm was not really convinced of the need for such an interface. With this approach, the UE under test would be aware and could take "Volkswagen type" strategy to cheat the tests. R&S accepted the Qualcomm fears on complexity, but a standardized interface simplified the situation.
 		noted No    
    R4‑1700079 On Baseline Testing Setups for UE RF Requirements ROHDE & SCHWARZ discussion Discussion Yes
YesRhode und Schwartz had provided this document as the way forward requested at the previoius meeting. The document concerned only far-field considerations. A number of possible configurations were presented. Qualcomm appreciated the good explanations of the various set-ups. The six-axis system required complex arms: was it possible to shield them effectively to avoid destablising the results? MVG shared this concern. Concerning the two-axis positioners, it was known that these did have a considerable impact on the RF measurements. MVG was sceptical about the idea of combining the methods because of the difficulty of isolating the equipement. R&S believed it would be possible to shield the important elements of the arms, but appreciated that this was a potential problem. The two-axis devices were a problem, and this sort of implementation should really be avoided if possible. A way of increasing the isolition between tx and rx was to use opposite polarisation.
 		noted No    
3.6 BS RF [FS_NR_newRAT]                      
3.6.1 General [FS_NR_newRAT] R4‑1700172 SI status of BS RF aspects NTT DOCOMO, INC. other Approval Yes
YesThis was a similar contribution to 182 for the UE. An update was envisaged at the present meeting in the light of discussions. Ericsson noted that they would present some relevant papers shortly. ACS and blocking would need spacial consideration.
 		revised No R4‑1700263  
    R4‑1700263 SI status of BS RF aspects NTT DOCOMO, INC. other Approval Yes
Yes
 		approved No   R4‑1700172
    R4‑1700113 Consideration on requirements below 6 GHz Huawei, Hisilicon other Approval No
YesLATE DOCUMENT Document not available.
 		withdrawn Yes    
    R4‑1700050 Prioritisation of RF requirements for NR Huawei, Hisilicon discussion Discussion Yes
YesHuawei presented the document. The ttesting was part of the conformance rather than core specification, so were de facto deprioritized. It was not useful to spend too much time on the topic. Ericsson had proposed an approach, and some tests might not be approprate in mm-wave bands.
 		noted No    
    R4‑1700225 Further elaboration on NR BS requirement priority for mm-wave frequencies Ericsson other Approval Yes
YesNoted without presentation.
 		noted No    
3.6.2 BS class [FS_NR_newRAT] R4‑1700036 NR BS Classes ZTE Corporation other Approval Yes
YesZTE presented the document examining BS classes below 6 GHz. Observation. In the case of other simulation parameters are kept for sub-6GHz NR BS, MCL has limited impact on the coexistence simulation results. Proposal. The conduct RF requirements of below 6GHz should be based on the BS classes which defined with set of parameters Huawei was not quite sure what was being proposed. What were the parameters. ZTE responded that the minimum distance would be included in a packet of parameters.
 		noted No    
    R4‑1700176 BS classification NTT DOCOMO, INC. other Approval Yes
YesNTT DOCOMO presented the document, which concluded: Proposal 1: The intention of BS classification should be studied and clarified before deciding criteria of definition. After clarifying the intention of classification, RAN4 should discuss appropriate criteria of definition. Proposal 2: It should be clarified and confirmed whether existing BS class is defined per BS equipment or per BS antenna connector. Proposal 3: RAN4 should study and decide the definition unit for NR BS classification (per BS equipment or other) by taking into account the existing definition unit. Ericsson considered that BS classification was to provide sets of requirements appropriate to difference deployments. The class should apply to the whole base station rather than parts of it. ZTE wondered what was the reference point in proposal 3. NTT DOCOMO replied that it was not the intention to use minimum coupling losss in proposal 3. Nokia and Huawei believed the upper power output requirement depended on class: it was important not to confuse these two aspects. Proposal 2 was referring to power output.
 		noted No    
    R4‑1700219 BS classes Ericsson other Discussion Yes
YesBase station classification might vary with frequency, and the MCL became less important once beam-forming was employed. Observation 1: The MCL parameter is not anything measurable either for a real BS or for any real deployment. Observation 2: The MCL is only a minor part of the description of the deployment scenarios for which the BS class related requirements were derived. Observation 3: Changing the description of the BS class related deployment scenarios from MCL for NR does not imply any need to re-define existing requirements. Huawei agreed that the scenarios in the relevant TRs was appropriate, and the use of minimum distance was useful. Nokia supported the minimum distance concept. Ericsson therefore proposed that classification would be based on minimum distance.
 		noted No    
    R4‑1700277 Way Forward on BS class description Ericsson other Approval Yes
YesEricsson introduced this way-forward document.
 		approved No    
3.6.3 Transmitter characteristics [FS_NR_newRAT] R4‑1700175 Upper frequency limit for TX spurious emission NTT DOCOMO, INC. other Approval Yes
YesNTT DOCOMO presented the document and reached the following conclusions: Observation 1: To define the spurious emission requirements, it is required to consider both ITU-R recommendation and OTA testability. Observation 2: Core requirement and conformance requirement should be discussed separately since testability aspects affect only for conformance. Proposal 1: 2nd harmonic should be an upper limit of spurious emission for NR bands above 13 GHz for core requirement. Proposal 2: It should be studied the measurable power level at high frequencies to decide reasonable upper frequency limit for conformance requirement. Huawei believed that having an untestable core requirement should provide some other means to demonstrate compliance. Ericsson was not sure that the document was aligned with FCC and ITU requirements. This needed checking. Nokia prefered to limit the core requirement rather than limit the test spec. Ericsson thought that consideration shold be given to measuring TRP closer than far-field. The feasibility of measurement of the second harmonic needed investigation.
 		noted No    
    R4‑1700037 NR BS Tx spurious emission frequency limits ZTE Corporation other Approval Yes
YesZTE presented the document. Proposal 1: 30 MHz as the lower frequency limit for below 6GHz Proposal 2: 5th harmonic as upper limit of frequency range for below 6GHz On proposal 1, Huawei thought the existing NSR and the derived harmonized specifications ended up in regulations outside the scope of 3GPP. For below 6 GHz, it was proposed to follow the existing LTE specs. Ericsson agreed that alignement with AAS was appropriate. Huawei agreed, but there if there were non-AAS specs, then the existing specs should be followed.
 		noted No    
    R4‑1700058 Spurious emissions for NR BS Nokia, Alcatel-Lucent Shanghai Bell discussion Approval Yes
YesNokia introduced the document Observation 1: RAN4 should study further the following options - Option 1: Limit the upper frequency below 100 GHz in the core specification. Value is FFS. - Option 2: If the core requirement need to go up to 100 GHz, limit the upper frequency below 100 GHz in the test specification. Value is FFS. Observation 2: RAN4 should study further the realistic requirement levels. Reasonable test system complexity and uncertainties need to be considered while setting the requirements. The Chairmen urged other equipment vendors to share their views on this topic.
 		noted No    
    R4‑1700059 OTA unwanted emissions for NR base stations Nokia, Alcatel-Lucent Shanghai Bell discussion Discussion Yes
YesNokia introduced the document, concluding: Observation 1: If there are corresponding TAB connectors specified, then NR BS operating at frequency less than 6 GHz can adopt the same methodology as in the AAS BS conducted case. Observation 2: TAB connectors do not exist in OTA environments. The TAB connector should be replaced with an appropriate term for OTA in current AAS base station definitions. Observation 3: For operating frequency < 6 GHz, unwanted emission limit scaling is possible provided the manufacturers declare the number of active transmitter units in the NR BS. Observation 4: For operating frequency < 6 GHz, if the OTA emissions are treated as the sum power radiated by all active transmitter units in the NR base station, then emission limit scaling by the number of TAB connectors is allowed and the black box concept is preserved. Huawei agreed that it was wise to follow AAS, but noted that how to count for OTA had not yet been decided.
 		noted No    
    R4‑1700174 Discussion on how to derive EIRP accuracy value NTT DOCOMO, INC. other Approval Yes
YesNTT DOCOMO presented the document, arriving at Observation 1: Firstly, RAN4 needs to study whether the same EIRP accuracy equation with AAS can be reused for NR or not. Observation 2: As soon as possible RAN4 needs to have agreement on EIRP accuracy equation for NR, since there is risk to not complete this work if RAN4 cannot reuse the same EIRP accuracy equation with AAS. Proposal 1: In NR SI phase, following should be studied to conclude EIRP accuracy modelling; ? Whether RAN4 can reuse the same EIRP accuracy equation with AAS ? If not, how to model EIRP accuracy for NR Huawei thought that in AAS it was one of the methodologies used to get information, and the question of accurace was avoided. It was probably not a good approach. Above 6 GHz, the OTA method was to be applied. Perhaps the starting point should be to examine the requirments of the network. Ericsson recalled that below 6 GHz it was already agreed to use legacy methods. NTT DOCOMO indicated that they had separated the above and below cases; it was questioned whether OTA could be specified also below 6 GHz. Ericsson countered that OTA would not be used below 6 GHz: existing LTE methods would be used. Huawei clarified that their comments had related to above 6 GHz.
 		noted No    
    R4‑1700048 Full power condition and beam directions for TRP requirements Huawei, Hisilicon discussion Discussion Yes
YesHuawei presented the document, investigating reasons why total output power might not be as assumed by simple addition of full power of all transmitters. Ericsson could envisage problems which might arise in practical base station design. Huawei had concluded that it was probably not useful to try to find solutions to scenarios which would not exist if the proposal could be agreed for the below 6 GHz case.
 		noted No    
    R4‑1700094 TP for NR Rel-14 TR 38.803: BS ACLR for mm-waves Ericsson pCR Approval Yes
YesThe document would be addressed at the next meeting.
 		noted No    
    R4‑1700141 Discussion on the transmit intermodulation requirement for NR BS ZTE Corporation discussion   Yes
YesZTE presented the document, concluding: Proposal 1: MCL of NR BS should be further studied taking antenna pattern of NR BS and operating frequency into account. Huawei noted that till now concentration had dealt with collocated base stations. But now the scenario had been extended to co-existance of base stations near to each other but not co-sited. Ericsson wondered if this requirement were actually necessary.
 		noted No    
    R4‑1700142 Transmit ON/OFF power for NR BS ZTE Corporation discussion   Yes
YesZTE presented the document, concluding Proposal 1: when specifying the Transmit OFF power for NR BS, we need to take the following factors into account: 1) noise figure of NR BS; 2) MCL between the aggressive and victim BS; 3) degradation level of noise floor due to interference from aggressive BS transmit OFF power; Proposal 2: it’s suggested that transmit ON power as maximum output power during the transmit ON period. Huawei thought the OTA work overlapped the below 6 GHz work. It was a question of coupling between the BS and equipment located nearby. With the OTA approach, these became proximity requirements, and it would be better to let OTA work develop further before discussing the above 6 GHz case.
 		noted No    
    R4‑1700265 WF on BS output power accuracy value for NR NTT DOCOMO other discussion Yes
YesNTT DOCOMO presented the way-forward document.
 		approved No    
    R4‑1700271 WF on NR BS Tx spurious emission frequency for below 6GHz NR ZTE, ZTE Microelectronics, Huawei other discussion Yes
YesZTE presented this way forward document.
 		approved No    
    R4‑1700283 WF on beam directions for TRP requirements Huawei other discussion Yes
YesHuawei presented the way-forward document. Qualcomm assumed this was for BS, and questioned the precise meaning of the text relating to TRP. Huawei clarified that beam-steering did not alter the total power, just the phasing and direction. Nokia felt that beam direction could alter ACIR slightly, so it was necessary to specify a given direction to ensure standard results.
 		noted No    
3.6.4 Receiver characteristics [FS_NR_newRAT] R4‑1700201 BS OTA Sensitivity for NR NEC other Approval Yes
YesOriginal agenda item 3.4.2. NEC presented the document. DOCOMO wondered how to define sensitiviety value, and the BS NF value already included the loss factor. If it were added again in this equation, it would be duplicated. Huawei wondered if this was above or below 6 GHz. They believed that above, it was necessary to have beamforming gain, which lead to a figure for effective antenna gain. So above 6 GHz, none of this calculation was required, since a single value (possibly frequency dependent) would suffice. Ericsson agreed, but below 6 GHz, the different modulation coding chain meant that legacy methods could not be used. NEC has already presented a proposal in its contribution 200.
 		noted No    
    R4‑1700046 Discussion on BS receiver requirements >6GHz Huawei, Hisilicon discussion Discussion Yes
YesHuawei explained how the receiver sensitivity was handled in NR, and most of the contents of this document had been discussed during the previoius document. A single OTA EIS vallue could be derived. Ericsson liked the idea of directly setting an OTA level. Huawei responded with the suggestion that it was still necessary to declare the range (at the 3 dB points).
 		noted No    
    R4‑1700047 Discussion on BS blocking requirements >6GHz Huawei, Hisilicon discussion Discussion Yes
YesIn this contribution, Huawei analysed how the blocking simulations on the conducted requirements above 6GHz could be purely OTA. Ericsson questioned that in NR this was the right approach, and care had to be taken of the probabilities of the direction and power of the signals. The shadow fading corrupted the spacial information. Nokia noted that the conclusion did not show the same figures as in the body (3 dB difference). Huawei proposed off line discussion. It was probably related to the off-peak margin issue.
 		noted No    
    R4‑1700140 Discussion on receiver intermodulation requirement of NR BS ZTE Corporation discussion   Yes
YesZTE presented the document, contrasting the E-UTRAN intermod situation with NR. Proposal 1: for NR BS, the interfering signal mean power of general intermodulation should also be derived according to the system level simulation results. Proposal 2: For 5G NR BS operating at the below 6GHz, this narrowband intermodulation requirement is also necessary; For 5G NR BS operating at the above 6GHz, the narrowband intermodulation requirement is not needed. On proposal 1, Ericsson understood that the simulations gave a lot of different results, and more thought needed to be given to the aim of NR in this respect. They agreed with proposal 2. Huawei noted the 37.104 had relevent requirements. ZTE was not sure whether these figures were just for GSM band reuse.
 		noted No    
    R4‑1700147 Discussion on ACS and Narrowband Blocking requirement of NR BS ZTE Corporation discussion   Yes
YesZTE proposed to define the narrowband blocking below 6 GHz, but above that figire it was not necessary to consider it. Proposal 1: For the 5G NR BS operating at below 6GHz, it’s proposed to define the NBB requirement accordingly with the similar approach as used for UTRA or E-UTRA system; for the 5G NR BS operating at above 6GHz, it’s proposed not to define NBB requirement if there is no narrowband system deployed in the adjacent channel; Proposal 2: For the power level of interfering signal of ACS requirement of 5G NR BS, it’s proposed to follow the similar approach as defined for UTRA system based on of REFSENS plus degradation level and ACS requirement of NR BS. Proposal 3: Interfering signal mean power of NBB requirement for NR BS could be derived according to the system level simulation results. Huawei agreed, but the derivation of ACS levels in the same was as for UTRA was questionable. The existing methodology had been based on conducted reference points, and these might not be applicable in OTA.
 		noted No    
    R4‑1700148 Discussion on blocking requirement for NR BS ZTE Corporation discussion   Yes
YesZTE presented the document, reaching two proposals: Proposal 1: to investigate the power level of the in-band blocking measured at per receiver connector at the transceiver array boundary for NR BS with conductive measurement point and measured over the air for NR BS without conductive measurement point. Proposal 2: to reuse the existing simulation assumption for NR coexistence study for the evaluation of in-band blocking signal for NR BS above 6GHz. On proposal 1, Ericsson proposed AAS below 6 GHz, and OTA above. For proposal 2, investigation of power control and shadow fading might be needed.
 		noted No    
    R4‑1700149 Discussion on receiver dynamic range of NR BS ZTE Corporation discussion   Yes
YesZTE introduced this generic discussion.
 		noted No    
    R4‑1700151 Discussion on receiver ICS requirement of NR BS ZTE Corporation discussion   Yes
YesZTE arrived at the proposal: Proposal 1: noise floor rise of interfering signal of ICS requirement should be preliminarily studied in SI or WI phase. Huawei considered that the ICS was an LTE requirement, and perhaps a similar requirement would be needed for NR, but this was not yet clear.
 		noted No    
    R4‑1700215 Spatial considerations for BS selectivity requirements Ericsson other Discussion Yes
YesEricsson introduced the OTA requirements and proposed that all types of combining should be catered for. Beamforming would tend to mitigate aggressive Ues, and would cause an averaging effect which needed to be taken into account when determining the requirement. There was also an averaging effect arising from different numerology, analagous to the ACLR requirement. For testing there were several possible approaches, of greater or lesser complexities. ACS and Rx blocking needed to be treated differently. Nokia believed that Ericsson's conclusion on ACS was also applicable for blocking. If the rx were well designed, there would be no saturation. They had a preference for the "same direction" testing scenario. Skyworks assumed that the UE tests were quite different from the BS tests, since the BS might have multiple beams. Ericsson noted that total rejection (digital and IF) had been considered. The document had no proposals relating to the UE, and it was simpler not to consider multiple beams.
 		noted No    
3.6.5 NR specific new requirements [FS_NR_newRAT] R4‑1700049 Discussion on spatial requirements for NR Huawei, Hisilicon discussion Discussion Yes
YesHuawei had captured previous meetings' observations, particularly with reference to side lobe level. Due to the spacial sensitivity, it woudl be difficult to arrive at a useful specification for side lobes. Handling complex beam patterns was very difficult. However, it might be useful to study the front to back ratio, and side lobe emissions outside the main sector. Several conclusions were arrived at: • Wanted beam performance is currently covered by the AAS BS requirement for EIRP accuracy. • Existing AAS BS steering directions declarations may need refining for NR BS. • Beam side lobe level is much more sensitive to errors than EIRP accuracy – however is not a useful parameter to consider as a minimum RF requirement. • Beam forming is rarely in a reflection-less environment, for complex beam forming weights side lobe level requirements are not so meaningful. • Element (or sector) side lobe attenuation and front to back ratio (in azimuth) are more important spatial parameters but are not directly related to beam forming. Ericsson noted the focus was on interference to other sectors, but Ericsson understood the concentration should be on side lobes within one sector. It was not clear what was the effect on end-user experience. CMCC agreed that the formulae for side lobes were difficult to handle. It was also necessary to investigate broadcast channel patterns and the effect on the SINR. They would provide some simulation results. Huawei responded to these remarks by saying that the self-interference in the same sector, the baseband demand for the beam should be achieved in practice. This was dependent on RF phase errors between different transceivers. Rather than settiing spacial requirements, consideration shouild be given to ensuring the RF could actually create the demanded beam shapes. Ericsson asked about the scope of this discussion: was this new requirements for mmWave and grid beams. These requirements would be easier to derive, and would be applicable to any sort of beamforming. Concerning the RF, Ericsson would prefer to avoid any requirements on the RF, and the requirements should not be too complex nor too restrictive.
 		noted No    
    R4‑1700161 Discussion on NR BS specific new OTA requirements CMCC discussion Approval Yes
YesCMCC looked at beam forming requirements and provided some parameter definitions. Observation 1: SLSR and FBR are necessary for 5G NR BS, and the above formulas could be considered when defining the requirements. Proposal 1: SLSR and FBR should be included in the scope of RAN4 specifications for 5G NR BS. Observation 2: The achievable performance of SLSR and FBR for 5G NR BS can be evaluated by simulation. Observation 3: The achievable performance of SLSR and FBR may change according to the different steering angle. Proposal2: The achievable performance of SLSR and FBR for 5G NR BS can be considered as reference to define the requirements. Proposal 3: The requirements of SLSR and FBR need to meet the demand of system deployment. Nokia considered it desirable to decide upon the element spacing of the antenna. Huawei had examined the sensitivity of side lobe level to errors in the transceivers. The requirements should not be too stringent, bearing in mind the capabilities of real implementations. Ericsson wonderedif coordination of the main beams could be coordinated so that the main beams did not interfer, then the side lobes were the most troublesome source of interference, but in the uncoordinated scenarion, the main beams themselves could be the majour source. It was not clear that the investigation was intended to cover generic deployments, or was specific to some common situations. CMCC agreed with Ericsson's remarks, and were also considering these factors.
 		noted No    
    R4‑1700173 NR specific new requirements NTT DOCOMO, INC. other Approval Yes
YesNTT DOCOMO contrasted current requirements and made proposals for new ones. Proposal 1: Fluctuation impact should be evaluated in existing TX/RX requirements to confirm the calibration function performance. Proposal 2: Required minimum EIRP gap level between peak and valley of beams should be specified. Proposal 3: Beam steering speed requirement for TX and RX respectively should be specified. Ericsson recommended that the testability of these aspects, including temperature sentitivity, should be taken into accoount. Also to be considered, the speed of beam tracking. On proposal 3, Nokia considered this to be a sort of end to end functionality, including area and speed. This should be defined on a per scenario basis. DOCOMO agreed with Ericsson's remarks. Proposal 1 might be difficult to test for temperature fluctuation stability. Nokia's remarks on beam tracking were welcome, and such topics were still FFS.
 		noted No    
    R4‑1700199 Discussion on NR BS specific beamforming new requirements NEC discussion Discussion Yes
YesThe NEC document considered potential requirements related to beamforming, noting that testing would present a number of problems, and hence it was difficult to specify minimum requirements for each scenario. Perhaps a vendor declaration approach would be appropriate.
 		noted No    
    R4‑1700221 Beam Quality requirements for NR Ericsson other Discussion Yes
YesEricsson noted the many different ways of achieving beam forming. There was a potentially long list of relevant parameters, but their precise definition was not easy. Further the interaction amongst these was even more difficult to set requirements for, and the importance was the end user experience. Intelligent schedulers might be aware of the potential for beam interference. Where there were large numbers of beams, testing could be complex and lengthy. Huawei agreed with many of these points. For multi-user MIMO, then these conslusions were too simple, and perhaps single user MIMO should be well understood first. The grid of beams approach was attractive, but the specs should be implementation agnostic. Nokia would like to divide the requrements as a function of the maximum number of beams supported. Ericsson wished to take a step back to see what was the aim of the requirements. How much energy was delivered to a single user, then the rest of the beam shape was irrelevant, but in multi-user MIMO, there were other important considerations. Ericsson was sceptical of the possibility of arriving at genericr requirements, and it was likely to result only in a declaration approach instead.
 		noted No    
    R4‑1700273 WF on NR BS specific new requirements NTT DOCOMO, CMCC other discussion Yes
YesNTT DOCOMO presented the document, noting that four potential requirements had been removed compared to the original draft.
 		approved No    
3.6.6 Testability [FS_NR_newRAT] R4‑1700033 Necessity of conducted requirements for below 6GHz Huawei discussion Discussion Yes
YesHuawei explained that this contribution was a response to discussions at the last meeting. It justified the needed for conducted requirements below 6 GHz. Proposal 1: during the NR BS discussion on the conducted vs. radiated testing below 6GHz, it is proposed to consider and adopt the eAAS agreements on the hybrid requirements. Proposal 2: Consider the agreements in the NR TR on testability and strive for the RAN4 alignment among discussions on NR testability in RF, RRM and demodulation requirements. Ericsson did not quite understand proposal 2. Huawei responded that this was not important at this stage.
 		noted No    
3.7 Testability(general such as IF etc) [FS_NR_newRAT] R4‑1700192 Testing Time Considerations Qualcomm Incorporated discussion Discussion Yes
YesQualcomm was concerned about the long testing times needed. The problems would be exacerbated when the upper frequency was around 100 GHz, with very many CA combinations. Testing time would probably extend to several weeks if the same "old" techniques were used. It was therefore necessary to investigate simplifed testing techniques, bearing in mind that typical devices would also support LTE, WCDMA, GSM, … Keysight agreed. Some of the tests would need to be changed - for example, SNR rather than throughput. Keysight were actively looking at ways to substantially cut test times.
 		noted No    
    R4‑1700213 Measurement considerations for large array or multi-array devices Keysight Technologies UK Ltd other Discussion Yes
YesKeysight presented the document, which dealt with devices with multiple antenna arrays. The distance between arrays was an important factor, and implied a considerable distance increase between EUT and test horn. Qualcomm wondered what would be the error if this distance (fig 2) were set at around 1 m rather than the 11 m ideal. Intel believed the next meeting should have the goal of defining the criterion for far field conditions. Keysight indicated that this would be per requirement.
 		noted No    
3.8 Others [FS_NR_newRAT]                      
4 Any other business                      
5 Close of the meeting