Search

CN-122029755-A - Method for beam failure recovery based on network machine learning

CN122029755ACN 122029755 ACN122029755 ACN 122029755ACN-122029755-A

Abstract

The present disclosure provides systems, apparatus, devices, and methods, including computer programs encoded on a storage medium, for beam failure recovery based on network side machine learning. The UE (102) sends (412) a beam failure recovery request BFRQ to the network entity (104) after beam failure detection BFD, the BFRQ indicating information for candidate beam prediction CBP. The UE (102) receives (416) a beam failure recovery response BFRR from the network entity (104), the BFRR indicating a candidate beam for recovering from the beam failure associated with BFD. The candidate beams are based on information for CBP.

Inventors

  • ZHANG YUSHU

Assignees

  • 谷歌有限责任公司

Dates

Publication Date
20260512
Application Date
20230929

Claims (17)

  1. 1.A method of wireless communication at a user equipment, UE, (102), comprising: Transmitting (412) a beam failure recovery request BFRQ to the network entity (104) after beam failure detection BFD, said BFRQ indicating information for candidate beam prediction CBP, and A beam failure recovery response BFRR is received (416) from the network entity (104), the BFRR indicating a candidate beam for recovering from a beam failure associated with the BFD, the candidate beam being based on the information for the CBP.
  2. 2. The method of claim 1, further comprising: -receiving (404) from the network entity (104) a configuration for at least one of: A first parameter of the CBP is enabled, A first set of reference signals RS for the BFD, A second set of RSs for the CBP, At least one uplink resource for said BFRQ, The search space for the BFRR is used, The number of reported beams in BFRQ, The reported quantity of each beam in BFRQ, or The number of time instances of the reported beam in the BFRQ.
  3. 3. The method of any of claims 1-2, wherein the transmitting (412) the BFRQ comprises: the BRFQ is sent via at least one of: Media access control, MAC, control element, CE, or Physical uplink control channel, PUCCH, resources.
  4. 4. The method of any one of claims 1 to 3, wherein the CBP is associated with at least one of spatial domain beam prediction or time domain beam prediction, and Wherein the BFRQ indicates at least one of: a serving cell index indicating a serving cell associated with the BFD; a TRP index indicating failure of a transmission reception point TRP associated with the BFD; An indicator of the candidate beam; a predefined DL RS index from a configured set of downlink reference signals DL RS, or Beam quality of the predefined DL RS.
  5. 5. The method of any one of claims 1-3, wherein the BFRQ indicates one or more DL RS indices and an associated beam quality for each of the one or more DL RS indices.
  6. 6. The method of any of claims 1-2, wherein the transmitting (412) the BFRQ comprises: Said BRFQ is sent via a physical random access channel PRACH.
  7. 7. The method of any one of claims 1-6, wherein the BFRR indicates a set of candidate beams including the candidate beam, the set of candidate beams being based on the CBP.
  8. 8. The method of any of claims 6 to 7, wherein the receiving (416) the BFRR comprises at least one of: Receiving the BFRR in a different serving cell than the serving cell associated with the BFD, or The BFRR is received in the same serving cell as the serving cell associated with the BFD.
  9. 9. The method of any one of claims 1 to 8, further comprising: -sending (402) a UE capability message to the network entity (104) indicating a capability of the UE (102) to facilitate the CBP.
  10. 10. The method of any one of claims 1 to 9, further comprising: Receiving (410) a second set of RSs from the network entity (104), and The information for the CBP is generated based on the second set of RSs.
  11. 11. The method of any one of claims 1 to 10, further comprising: receiving (418) a third set of RSs from the network entity (104) via the candidate beam to verify whether the candidate beam meets a beam quality criterion, and -Transmitting (420) a beam report based on the third set of RSs to the network entity (104).
  12. 12. A method of wireless communication at a network entity (104), comprising: Receiving (412) a beam failure recovery request BFRQ from a user equipment, UE, (102) after beam failure detection, BFD, said BFRQ indicating information for candidate beam prediction, CBP, and -Transmitting (416) a beam failure recovery response BFRR to the UE (102), the BFRR indicating at least one candidate beam for recovering from a beam failure associated with the BFD, the at least one candidate beam being based on the information for the CBP.
  13. 13. The method of claim 12, further comprising: -transmitting (404) to the UE (102) a configuration for at least one of: A first parameter of the CBP is enabled, A first set of reference signals RS for the BFD, A second set of RSs for the CBP, At least one uplink resource for said BFRQ, The search space for the BFRR is used, The number of reported beams in BFRQ, The reported quantity of each beam in BFRQ, or The number of BFD instances of the reported beam in the BFRQ.
  14. 14. The method of any of claims 12 to 13, further comprising: -transmitting (410) the second set of RSs to the UE (102), wherein the information for the CBP is based on the second set of RSs.
  15. 15. The method of any of claims 12 to 14, further comprising: a prediction of the at least one candidate beam is generated based on at least one of a spatial domain beam prediction or a temporal domain beam prediction.
  16. 16. The method of any of claims 12 to 15, further comprising: transmitting (418) a third set of RSs to the UE (102) via a set of candidate beams associated with the at least one candidate beam, and A beam report is received (420) from the UE (102) verifying at least one of the candidate beam sets, wherein the at least one of the candidate beam sets meets a beam quality criterion.
  17. 17. An apparatus for wireless communication, the apparatus comprising a memory, a transceiver, and a processor coupled to the memory and the transceiver, the apparatus configured to implement the method of any of claims 1-16.

Description

Method for beam failure recovery based on network machine learning Technical Field The present disclosure relates generally to wireless communications, and more particularly to beam failure recovery based on network side Machine Learning (ML) techniques. Background The third generation partnership project (3 GPP) specifies a radio interface called the fifth generation (5G) New Radio (NR) (5G NR). The architecture of a 5G NR wireless communication system includes a 5G core (5 GC) network, a 5G radio access network (5G-RAN), user Equipment (UE), and the like. The 5G NR architecture seeks to provide increased data rates, reduced latency, and/or increased capacity compared to previous generation cellular communication systems. Generally, wireless communication systems provide various telecommunication services (e.g., telephony), video, data, messaging, broadcast, etc.) based on multiple access techniques, such as Orthogonal Frequency Division Multiple Access (OFDMA) techniques, that support communication with multiple UEs. Improvements in mobile broadband continue to advance in such wireless communication technologies. For example, the beam restoration process may be modified to implement Candidate Beam Prediction (CBP) through a network side Machine Learning (ML) model. Disclosure of Invention The following presents a simplified summary of one or more aspects in order to provide a basic understanding of such aspects. This summary is not an extensive overview of all contemplated aspects. This summary does not identify key or critical elements of all aspects nor delineate the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form as a prelude to the more detailed description that is presented later. Next generation communication systems (e.g., fifth generation (5G) and sixth generation (6G) wireless technologies) use beam management techniques to improve throughput to meet wireless user demands. Beam management techniques are becoming increasingly complex due to the adoption of higher frequency bands, user mobility, and increased antenna counts. Machine Learning (ML) helps reduce the complexity and overhead associated with beam management techniques, such as by means of Candidate Beam Prediction (CBP). CBP by the network-side ML model may reduce overhead associated with Candidate Beam Detection (CBD) Reference Signals (RSs). For example, the network entity may configure CBD RSs for only a subset of the network beams, but not all of the network beams. Furthermore, since CBP is performed by the network-side ML model at the network entity instead of at the UE, overhead is further reduced. However, the beam restoration process may need to be modified to implement CBP by the network side ML model. For example, the modified beam restoration procedure may require control signaling between a network entity and a User Equipment (UE) designated for implementing the network side ML CBP. The modified beam recovery procedure may also require specifying the contents of the beam failure recovery request (BFRQ) and the Beam Failure Recovery Response (BFRR). Aspects of the present disclosure address the above and other deficiencies by implementing CBP techniques based on a network-side ML model. In one example, the UE detects beam failure. The UE sends BFRQ indicating information that the network entity can use to predict the candidate beam. In response, the network entity transmits BFRR indicating the predicted candidate beam that the UE should use. In some implementations, the network entity may send a set of reference signals for the UE to verify or authenticate the predicted candidate beam, and the UE may report such authentication to the network entity. Thus, the UE starts using the predicted candidate beam so that the UE can recover from the beam failure. In this way, the UE does not have to identify the candidate beams, but after the UE fails to identify a beam, the UE sends a report to the network entity including the beam quality of a subset of the network beams. The network entity may predict the network beam based on the report. According to some aspects, the UE sends BFRQ to the network entity after Beam Failure Detection (BFD). BFRQ indicates information for CBP. The UE receives BFRR from the network entity, the BFRR indicating a candidate beam for recovering from the beam failure associated with BFD. The candidate beams are based on information for CBP. According to some aspects, the network entity receives BFRQ from the UE after BFD. BFRQ indicates information for CBP. The network entity sends BFRR to the UE, the BFRR indicating at least one candidate beam for recovering from the beam failure associated with BFD. At least one candidate beam is based on information for CBP. Drawings Fig. 1 illustrates a diagram of a wireless communication system including a plurality of User Equipments (UEs) and network entities communicating via one or more cells, acco