CN-122003904-A - Model signaling method for multiple connections

Abstract

A method of pre-configuring AI/ML operation when a multi-connection link is enabled in a wireless mobile communication system that includes a base station (e.g., gNB) and a mobile station (e.g., UE). When applying AI/ML models to a radio access network, model performance such as reasoning and/or training is dependent on different model execution environments with different configuration parameters. Thus, by reconfiguring any model in operation with multiple connection links, the signaling overhead due to various LCM model operations can be reduced and model performance enhanced.

Inventors

• Jin Huchen
• R. SHAH

Assignees

• 欧摩威德国有限公司

Dates

Publication Date

20260508

Application Date

20240912

Priority Date

20230927

Claims (19)

1. 1. A method of applying distributed multi-model operation to AI/ML LCM procedures in a network comprising at least one primary node (MN), at least one Secondary Node (SN), and User Equipment (UE), the method comprising the steps of: the candidate SN list is configured based on SN selection criteria, After receiving the trigger signal, AI/ML configuration and LCM-based model information are sent to the candidate SN, Receiving an SN indication message from at least one candidate SN, the SN indication message including an SN establishment requirement and/or an SN establishment acknowledgment of the AI/ML LCM process, Activating at least one SN link based on the received SN indication message and also based on the condition, and Sending a reconfiguration message to the UE via L1/L2/L3 signaling, reconfiguring the UE to establish multi-model operation of the AI/ML LCM procedure between the UE and the at least one candidate SN, and.
2. 2. The method of claim 1, wherein the trigger signal corresponds to a preconfigured indicator or threshold of a traffic load level and/or additional model operation requirements of a current cell condition.
3. 3. The method of claim 1 or 2, wherein activating the at least one SN link includes performing aggregation, replication, or switching of configured multi-mode operations based on different use cases and LCM phases.
4. 4. The method of any of the preceding claims, wherein the further step comprises sending a UE indication message to the UE via L1/L2/L3 signaling, the UE indication message indicating activation or deactivation of a configured multi-mode operation between the UE and the at least one candidate SN.
5. 5. The method of any of the preceding claims, wherein a group of UEs connected to the MN are connected to the at least one candidate SN for multimodal operation of the AI/ML LCM procedure.
6. 6. The method of any of the preceding claims, wherein it is verified whether an inactive AI/ML model is supporting the multi-model operation, the inactive AI/ML model being deployed in a link between the UE and the at least one candidate SN, wherein any of the inactive AI/ML models can be activated to replace or support a current model of the multi-model operation.
7. 7. A method as claimed in any preceding claim, wherein a decision is made by the network as to whether to perform the distributed multimodal operation, and the list of candidate SNs is provided by the at least one MN.
8. 8. The method of any of the preceding claims, wherein the SN selection criteria is at least one of link quality, AI/ML applicable condition status, and/or model execution environment.
9. 9. The method of any of the preceding claims, wherein the same model is applied for the at least one MN and the at least one candidate SN.
10. 10. The method according to any of the preceding claims, wherein different models are applied for the at least one MN and the at least one candidate SN.
11. 11. The method of any of the preceding claims, wherein a different LCM phase is activated for each of the at least one SN link.
12. 12. The method of claim 11, wherein the LCM phase is one of model training in a MN-UE link and/or model training in a SN-UE link.
13. 13. The method of any of the preceding claims, wherein a UE-side model communicates with a different number of network-side models through a plurality of links with the MN and with the at least one candidate SN, comprising: can deactivate the distributed model operation based on an indication message sent by the UE, or The network can decide to deactivate the distributed model operation for the at least one MN and the at least one candidate SN.
14. 14. A wireless device comprising at least one memory and at least one processor configured to implement the method of any of the preceding claims.
15. 15. A User Equipment (UE) comprising the wireless device of claim 14.
16. 16. A Base Station (BS) comprising at least one memory and at least one processor configured to implement the method of any of claims 1 to 13.
17. 17. A wireless communication system comprising at least one base station according to claim 16 and at least one user equipment according to claim 15.
18. 18. A computer program product comprising instructions which, when executed by at least one processor, configure the at least one processor to implement the method of any one of claims 1 to 13.
19. 19. A computer-readable storage medium comprising instructions that, when executed by at least one processor, configure the at least one processor to implement the method of any one of claims 1 to 13.

Description

Model signaling method for multiple connections Technical Field The present disclosure relates to Artificial Intelligence (AI)/Machine Learning (ML) operation pre-configuration, wherein techniques for reconfiguring and signaling specific information to improve efficient signaling over multiple connection links are presented. Background In 3GPP (3 rd Generation partnership project), one of the selected research projects as an approved version 18 package is AI/ML (Artificial Intelligence/machine learning), as described in the related document (RP-213599) submitted in 3GPP TSG RAN (technical Specification group radio Access network) meeting #94 e. The formal name of the AI/ML Study item is "Study on AI/ML for NR AIR INTERFACE (Study on AI/ML of NR air interface)", and the current RANs WG1 and WG2 are actively formulating specifications. The goal of the research project is to identify a generic AI/ML framework and the area of interest to gain using AI/ML-based techniques and use cases. According to 3GPP, the main goal of this research project is to exploit the target use case to study the AI/ML framework for the air interface by taking into account performance, complexity and potential specification effects. In particular, the AI/ML model, terms and descriptions used to establish the common and specific characteristics of the framework will be one of the key working areas. With respect to the AI/ML framework, aspects are being considered for investigation, and one of the key items is lifecycle management (LCM) with respect to AI/ML models, wherein multiple phases are forcibly included for model training, model deployment, model reasoning, model monitoring, model updating, etc. Earlier, in 3GPP TR 37.817, version 17, entitled "Study on enhancement for Data Collection for NR and EN-DC (research on NR and EN-DC data collection enhancement)", UE (user Equipment) mobility was also considered one of the AI/ML use cases. One scenario for model training/reasoning is that both functions are located within the RAN node. Subsequently, a new work item "ARTIFICIAL INTELLIGENCE (AI)/MACHINE LEARNING (ML) for NG-RAN (artificial intelligence (AI)/Machine Learning (ML) for NG-RAN))" is initiated in release 18 to specify data collection enhancements and signaling support within existing NG-RAN interfaces and architectures, with mobility optimization included as one of the target use cases. For the ongoing standardization work described above, UE mobility to support the RAN-based AI/ML model may be considered to be important for both the gNB (base station or radio access point) and the UE to meet any desired model operations (e.g., model training, reasoning, selection, handover, updating, monitoring, etc.) as the UE moves around. Currently, no specification is defined for the signaling method or gNB-UE behavior for Multiple Connections (MC) as the RAN-based AI/ML model operation continues. Therefore, it is necessary to study any canonical effect by considering the model operation during MC connection. There is also a need to address any mechanisms of additional signaling methods and/or gNB-UE behavior to support MC-based model operation between the gNB and UE such that any potential impact of UE mobility on model operation in the RAN should be minimized under service continuity. On the other hand, in 3GPP, the terms in the working list contain a high-level description set on AI/ML model training, reasoning, verification, testing, UE-side model, network-side model, single-side model, double-side model, etc. The UE-side model and the network-side model indicate that the AI/ML model is located in the UE-side and the network-side, respectively, for operation. In a similar context, the one-sided model and the two-sided model indicate that the AI/ML model is located on one side and on both sides of the network, respectively. All signaling aspects supporting the above items have not been specified so far, as the definition of terms is still under discussion for further modification. Any potential standard impact under the new or enhanced mechanisms supporting the AI/ML model with the above-described work list items is one of the key areas to investigate in the AI/ML study. US 2023,145,079 A1 discloses a wireless communication method implemented by a User Equipment (UE) comprising establishing a Secondary Cell Group (SCG) using a second Radio Access Technology (RAT) different from a first RAT associated with a primary cell group (MCG). The method also includes wirelessly communicating via the secondary cell group and the primary cell group. The method also includes predicting a Radio Link Failure (RLF) for the secondary cell group based on a plurality of inputs to the machine learning model. The method further includes routing data transmissions from the secondary cell group to the primary cell group after the SCG RLF is predicted. WO 2022 144 582 A1 discloses a system, method, and non-transitory computer-readable medium for facilitating