By WG RAN3 Leadership: Yin Gao, Angelo Centonza, Gen Cao

First published June 2024, in Highlights Issue 08  

RAN3 has completed all Release 18 (Rel-18) Work items during the fourth quarter of 2023, continuing the maintenance work in Q1 of 2024. The release is now frozen in terms of functionality, with ASN.1 (Abstract Syntax Notation One – describing the interface between any two entities) scheduled to be frozen by June.

During technical discussions for the normative phase of Rel-18, RAN3 took a leading role for some Work items, including; AI/ML for NG-RAN, SON/MDT, QoE and mobile IAB. This article provides an overview of the RAN3-led features that were included in the Rel-18 package, mainly from - but not restricted to - the perspective of RAN3.

SON/MDT enhancements

The Self-Organising Networks and Minimization of Drive Tests (SON/MDT) feature was first introduced in LTE for the support of system deployment and performance optimization, in Rel-16 and further enhanced in Rel-17 for NR. The Rel-18 SON/MDT enhancements continued the work on features that were not fulfilled in the earlier releases.

The main objective of this Work Item is to specify the data collection mechanism in NR for the purpose of performing SON/MDT, including enhancements for inter-RAT Successful Handover Report (inter-RAT SHR) and Successful PScell Change Report (SPR), MRO for NR-DC Conditional PSCell Change (CPC) and Conditional PSCell Addition (CPA), Fast MCG recovery, Inter-system handover voice fallback, RACH Enhancements, SON/MDT enhancements for Non-Public Networks (NPN), SON for NR-U.

The Rel-18 SON/MDT enhancements help further optimization of service continuity and robustness during mobility and contribute to the construction of wireless network automation capabilities through enhanced data collection mechanisms, which can help operators to gradually improve network performance and maintenance efficiency, leading to the evolution of network automation.

The Rel-19 SON/MDT Work item is expected to tackle some further improvements based on Rel-18 mobility mechanisms, as well as some new features or use cases left from previous releases.

NR QoE enhancement

The support for Quality of Experience (QoE) measurement collection was first introduced in both UMTS and LTE. The NR QoE measurement collection mechanism was standardized in Rel-17, while in Rel18 some new features were introduced and the remaining issues from Rel-17 addressed. For Rel-17, the QoE measurement collection is only supported in RRC_CONNECTED state and standalone architecture. In Rel-18, QoE measurement collection for application sessions delivered via multicast communication service for UEs in RRC_CONNECTED is supported. QoE measurement collection for application sessions delivered via broadcast communication service is supported in RRC_CONNECTED, RRC_INACTIVE, and RRC_IDLE states. For the NR-DC scenario, the QoE configuration and measurement reporting via MN/SN and QMC (QoE Measurement Collection) continuity during mobility are supported, with the necessary coordination between MN and SN specified over XnAP. Some other left-over issues from Rel-17 (e.g., the F1AP enhancement for RAN visible QoE, intra-system inter-RAT Handover) are also solved in Rel-18.

With higher requirements from various new applications, the way to evaluate the quality of different services becomes more important to the users as well as to the service providers.

Mobile IAB for NR

Mobile IAB (Integrated Access and Backhaul) for NR is introduced to support the scenario of IAB-nodes onboard vehicles, to provide 5G coverage/capacity to onboard or surrounding UEs. The Rel-18 framework builds upon the architecture and procedures from Rel-16 and Rel-17 IAB. Rel-18 supports a dedicated network integration procedure for the mobile IAB-node and introduces procedures for mobile IAB-node mobility, including mobile IAB-MT migration and inter-donor mobile IAB-DU migration. The support for NCGI reconfiguration for mobile IAB-DU cells is introduced to solve the issue of NCGI collision during movement of the mobile IAB-nodes.

In addition, enhancements are introduced to improve the performance of mobile IAB-node mobility together with its served UEs, e.g., the support for RACH-less UE handover between source and target logical mobile IAB-DUs, and frequency prioritization for mobile IAB cell reselection can be performed by UEs based on the mobile IAB cell indicator broadcast by the mobile IAB cell and the assistance information about neighboring mobile IAB cells in system information.

The mobile IAB functionality, as a new topology deployment solution for 5G wireless communication, can guarantee the service of onboard UEs such as passengers in public transportation, and it allows a quick deployment of new cells to provide/improve connectivity in areas with high demand. 

AI/ML for NG-RAN

The support of Artificial Intelligence (AI) and Machine Learning (ML) for NG-RAN is based on the outcome of the corresponding study item (TR37.817) in Rel-18. The data collection enhancements and signaling support over existing network interfaces are specified in the Rel-18 AI/ML for NG-RAN Work item, specifically for three use cases, i.e., network energy saving, load balancing and mobility optimization. In order to support the use cases by means of AI/ML functionalities, input information from neighbor NG-RAN nodes is required. Detailed configuration related to the AI/ML data collection and reporting are transferred over the Xn interface via the Data Collection Reporting Initiation procedure, and the reporting mechanism is performed via the Data Collection Reporting procedure. As AI/ML techniques develop and with the deployment of AI/ML in the 5G mobile network architecture, it is believed that AI/ML processing will assist the operators to improve network management and user experience in a remarkable and proactive way.

After the completion of Rel-18 AI/ML Work item, a new study item is launched in Rel-19 to investigate AI/ML support for new use cases, namely Network Slicing and Coverage and Capacity Optimization (CCO), and also to support the use cases in scope for the case of split RAN architecture, which was not fulfilled in Rel-18. The further study of AI/ML in Rel-19 contributes to a better technical support for the deployment of AI/ML in NG-RAN.

NR Timing Resiliency and URLLC

The NR Timing Resiliency and URLLC Work item is aimed at providing RAN support for the corresponding WG SA2 normative work on several aspects of this topic. The three main aspects with RAN impacts are 5GS timing synchronization and reporting, interworking with TSN network deployed in the transport network, and adapting downstream and upstream scheduling based on RAN feedback for low latency communication. Enhancements on NGAP, XnAP and F1AP are introduced, in support to the time resiliency and low latency requirements discussed in WG SA2. For instance, when it comes to RAN TSS reporting, timing synchronization status reporting procedures are defined over NGAP and F1AP, allowing for node level and cell level reporting of RAN TSS, as well as for updating of clock quality information to UEs. During normative phase, tight coordination between RAN, SA and CT Working groups was maintained, which guaranteed an efficient coordination of standard impacts for the established requirements among different WGs jointly contributing to the completion of this Rel-18 feature.

Since robust synchronization and low latency are always the main goal of service providers, positive expectation is held for this feature with respect to further applications, especially with the fast uptake of more diverse services requiring ultra-low latency in future.

Network Slicing

The work on Rel-18 RAN slicing enhancements is introduced with the aim to standardize the RAN impacts of further RAN slicing enhancement based on the discussion and progress in WG SA2. A study item prior to the normative phase of the Rel-18 Work item is carried out in Rel-17 (TR38.832). With the specification of network slicing enhancements on the RAN side, the network slice service continuity is supported, allowing the replacement of an S-NSSAI by an alternative S-NSSAI. The NG-RAN node may also receive the signaling from the AMF which carries the information of partial allowed NSSAI, which can be used to make decisions related to mobility at RRC_CONNECTED mode. When the service area of a network slices does not match the set of cells for a-deployed tracking area, the cells out of the service area could be configured with zero resources for the concerned slices. Information about zero resources configured for a slice in one or more cells can be exchanged between neighbor NG-RAN nodes via the Xn interface.

The enhancement of Network slicing allows a more dynamic allocation and management of radio resources across different slices, and it guarantees the performance of various services.

Non-public Networks Phase 2

The Rel-18 NPN is aimed at supporting RAN functionalities for the corresponding WG SA2 enhancements on non-public networks. According to the Rel-18 enhancements, the UE shall maintain an equivalent SNPN list for equivalent SNPN (re)selection, to support idle mode mobility between SNPNs. In the case of handover, the selected SNPN ID is indicated to the target NG-RAN node for both Xn and NG based handover, so that equivalent SNPNs is supported in connected mode. NR-DC across equivalent SNPNs is also supported. Non-3GPP access for SNPNs is also supported in Rel-18 with the enhancement on NGAP signaling.

The completion of the Rel-18 NPN Work item provides support for the establishment and enhancements of 5G private networks, as well as bringing benefits to industry deployments.

Conclusion

As wireless communication networks will be evolved continuously, network management and optimization are an ever relevant theme in the development of 5G Advanced. These techniques e.g., AL/ML, SON/MDT, URLLC, Slicing, NPN will empower wireless networks to operate autonomously, predictively, on-demand, and to offer flexible and efficient resource allocation further allowing the ‘vertical’ industries to flourish.

For more on WG RAN3: www.3gpp.org/3gpp-groups