Devices and methods for communication
Patent Information
- Application Number
- PCT/CN2024/084933
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-03-29
- Publication Date
- 2025-10-02
Smart Images

Figure CN2024084933_02102025_PF_FP_ABST
Abstract
Description
DEVICES AND METHODS FOR COMMUNICATION
[0001] FIELDS
[0002] Example embodiments of the present disclosure generally relate to the field of communication techniques and in particular, to devices, methods and medium for artificial intelligence / machine learning (AI / ML) based radio link failure (RLF) prediction.BACKGROUND
[0003] With the development of AI and ML techniques, various measurements and events in the network may be predicted. By analyzing and forecasting resource management, measurement results, RLF, handover (HO) failures, and measurement events, mobility performance and network efficiency may be enhanced.SUMMARY
[0004] In general, embodiments of the present disclosure provide devices, methods and medium for AI / ML based RLF prediction.
[0005] In a first aspect, there is provided a terminal device. The terminal device comprises: a processor configured to cause the terminal device to: obtain radio link information associated with a radio link of the terminal device; and generate a set of prediction results for at least one predicted RLF based on the radio link information by using a set of AI / ML models. A prediction result comprises information about a predicted RLF.
[0006] In a second aspect, there is provided a network device. The network device comprises: a processor configured to cause the network device to: receive, from a terminal device, at least one portion of a set of prediction results for at least one predicted RLF. The set of predication results are generated based on a radio link information associated with a radio link of the terminal device by using a set of AI / ML models. A prediction result comprises information about a predicted RLF.
[0007] In a third aspect, there is provided a communication method performed by a terminal device. The method comprises: obtaining radio link information associated with a radio link of the terminal device; and generateing a set of prediction results for at least one predicted RLF based on the radio link information by using a set of AI / ML models. A prediction result comprises information about a predicted RLF.
[0008] In a fourth aspect, there is provided a communication method performed by a network device. The method comprises: receiving, from a terminal device, at least one portion of a set of prediction results for at least one predicted RLF. The set of predication results are generated based on a radio link information associated with a radio link of the terminal device by using a set of AI / ML models. A prediction result comprises information about a predicted RLF.
[0009] In a fifth aspect, there is provided a computer readable medium having instructions stored thereon, the instructions, when executed on at least one processor, causing the at least one processor to carry out the method according to the third, or fourth aspect.
[0010] Other features of the present disclosure will become easily comprehensible through the following description.BRIEF DESCRIPTION OF THE DRAWINGS
[0011] Through the more detailed description of some example embodiments of the present disclosure in the accompanying drawings, the above and other objects, features and advantages of the present disclosure will become more apparent, wherein:
[0012] FIG. 1 illustrates an example communication environment in which example embodiments of the present disclosure can be implemented;
[0013] FIG. 2 illustrates a schematic diagram of T310 expiry in accordance with some embodiments of the present disclosure;
[0014] FIG. 3 illustrates a signaling flow of predicting RLF in accordance with some embodiments of the present disclosure;
[0015] FIG. 4 illustrates a flowchart of a method implemented at a terminal device according to some example embodiments of the present disclosure;
[0016] FIG. 5 illustrates a flowchart of a method implemented at a network device according to some example embodiments of the present disclosure; and
[0017] FIG. 6 illustrates a simplified block diagram of a device that is suitable for implementing example embodiments of the present disclosure.
[0018] Throughout the drawings, the same or similar reference numerals represent the same or similar element.DETAILED DESCRIPTION
[0019] Principle of the present disclosure will now be described with reference to some example embodiments. It is to be understood that these embodiments are described only for the purpose of illustration and help those skilled in the art to understand and implement the present disclosure, without suggesting any limitation as to the scope of the disclosure. Embodiments described herein can be implemented in various manners other than the ones described below.
[0020] In the following description and claims, unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skills in the art to which this disclosure belongs.
[0021] As used herein, the term ‘terminal device’ refers to any device having wireless or wired communication capabilities. Examples of the terminal device include, but not limited to, user equipment (UE) , personal computers, desktops, mobile phones, cellular phones, smart phones, personal digital assistants (PDAs) , portable computers, tablets, wearable devices, internet of things (IoT) devices, Ultra-reliable and Low Latency Communications (URLLC) devices, Internet of Everything (IoE) devices, machine type communication (MTC) devices, devices on vehicle for V2X communication where X means pedestrian, vehicle, or infrastructure / network, devices for Integrated Access and Backhaul (IAB) , Space borne vehicles or Air borne vehicles in Non-terrestrial networks (NTN) including Satellites and High Altitude Platforms (HAPs) encompassing Unmanned Aircraft Systems (UAS) , eXtended Reality (XR) devices including different types of realities such as Augmented Reality (AR) , Mixed Reality (MR) and Virtual Reality (VR) , the unmanned aerial vehicle (UAV) commonly known as a drone which is an aircraft without any human pilot, devices on high speed train (HST) , or image capture devices such as digital cameras, sensors, gaming devices, music storage and playback appliances, or Internet appliances enabling wireless or wired Internet access and browsing and the like. The ‘terminal device’ can further has ‘multicast / broadcast’ feature, to support public safety and mission critical, V2X applications, transparent IPv4 / IPv6 multicast delivery, IPTV, smart TV, radio services, software delivery over wireless, group communications and IoT applications. It may also incorporate one or multiple Subscriber Identity Module (SIM) as known as Multi-SIM. The term “terminal device” can be used interchangeably with a UE, a mobile station, a subscriber station, a mobile terminal, a user terminal or a wireless device.
[0022] The term “network device” refers to a device which is capable of providing or hosting a cell or coverage where terminal devices can communicate. Examples of a network device include, but not limited to, a Node B (NodeB or NB) , an evolved NodeB (eNodeB or eNB) , a next generation NodeB (gNB) , a transmission reception point (TRP) , a remote radio unit (RRU) , a radio head (RH) , a remote radio head (RRH) , an IAB node, a low power node such as a femto node, a pico node, a reconfigurable intelligent surface (RIS) , and the like.
[0023] The terminal device or the network device may have Artificial intelligence (AI) or Machine learning capability. It generally includes a model which has been trained from numerous collected data for a specific function, and can be used to predict some information.
[0024] The terminal or the network device may work on several frequency ranges, e.g., FR1 (e.g., 450 MHz to 6000 MHz) , FR2 (e.g., 24.25GHz to 52.6GHz) , frequency band larger than 100 GHz as well as Tera Hertz (THz) . It can further work on licensed / unlicensed / shared spectrum. The terminal device may have more than one connection with the network devices under Multi-Radio Dual Connectivity (MR-DC) application scenario. The terminal device or the network device can work on full duplex, flexible duplex and cross division duplex modes.
[0025] The embodiments of the present disclosure may be performed in test equipment, e.g., signal generator, signal analyzer, spectrum analyzer, network analyzer, test terminal device, test network device, channel emulator. In some embodiments, the terminal device may be connected with a first network device and a second network device. One of the first network device and the second network device may be a master node and the other one may be a secondary node. The first network device and the second network device may use different radio access technologies (RATs) . In some embodiments, the first network device may be a first RAT device and the second network device may be a second RAT device. In some embodiments, the first RAT device is eNB and the second RAT device is gNB. Information related with different RATs may be transmitted to the terminal device from at least one of the first network device or the second network device. In some embodiments, first information may be transmitted to the terminal device from the first network device and second information may be transmitted to the terminal device from the second network device directly or via the first network device. In some embodiments, information related with configuration for the terminal device configured by the second network device may be transmitted from the second network device via the first network device. Information related with reconfiguration for the terminal device configured by the second network device may be transmitted to the terminal device from the second network device directly or via the first network device.
[0026] As used herein, the singular forms ‘a’ , ‘an’ and ‘the’ are intended to include the plural forms as well, unless the context clearly indicates otherwise. The term ‘includes’ and its variants are to be read as open terms that mean ‘includes, but is not limited to. ’ The term ‘based on’ is to be read as ‘at least in part based on. ’ The term ‘one embodiment’ and ‘an embodiment’ are to be read as ‘at least one embodiment. ’ The term ‘another embodiment’ is to be read as ‘at least one other embodiment. ’ The terms ‘first, ’ ‘second, ’ and the like may refer to different or same objects. Other definitions, explicit and implicit, may be included below.
[0027] In some examples, values, procedures, or apparatus are referred to as ‘best, ’ ‘lowest, ’ ‘highest, ’ ‘minimum, ’ ‘maximum, ’ or the like. It will be appreciated that such descriptions are intended to indicate that a selection among many used functional alternatives can be made, and such selections need not be better, smaller, higher, or otherwise preferable to other selections.
[0028] As used herein, the term “resource, ” “transmission resource, ” “uplink resource, ” or “downlink resource” may refer to any resource for performing a communication, such as a resource in time domain, a resource in frequency domain, a resource in space domain, a resource in code domain, or any other resource enabling a communication, and the like. In the following, unless explicitly stated, a resource in both frequency domain and time domain will be used as an example of a transmission resource for describing some example embodiments of the present disclosure. It is noted that example embodiments of the present disclosure are equally applicable to other resources in other domains.
[0029] As used herein, a model may be equivalent to at least one of the following: an AI / ML model, a ML model, an AI model, a data-driven, a data processing model, an algorithm, a functionality, a procedure, a process, an entity, a function, a feature, a feature group, a model identifier (ID) , an ID, a functionality ID, a configuration ID, a scenario ID, a site ID, or a dataset ID. As a result, the above terms may be used interchangeably.
[0030] In some embodiments, the model may be represented by or associated with a channel, a resource, a resource set, a reference signal (RS) resource, a RS resource set, a RS port, a set of RS ports, a RS port ID, or a set of RS port IDs.
[0031] In some embodiments, the model may comprise a set of weights values that may be learned during training, for example for a specific architecture or configuration, where a set of weights values may also be called a parameter set.
[0032] In some embodiments, the model may be used to predict a target cell, or measurements of a set of beams of a set of candidate cells in future based on at least historical measurements (e.g., L1-RSRP, L1-SINR) of a set of beams of a set of candidate cells.
[0033] In some embodiments, an input of the ML model (i.e., AI input) may refer to the input of a model and indicate data inputted into the model, which may be equivalent to data.
[0034] In some embodiments, an output of ML model (i.e., AI output) may refers to the output of a model and indicate result (s) outputted by the model, which is equivalent to label / data.
[0035] As used herein, the term “AI / ML model” may refer to a data driven algorithm that applies AI / ML techniques to generate a set of outputs based on a set of inputs. In the context of the present disclosure, the term “AI / ML model” may be interchangeably with the terms “model” , “AI model” and “ML model” .
[0036] The term “UE-side (AI / ML) model” used herein may refer to an AI / ML Model of which inference is performed entirely at the UE. The term “network-side (AI / ML) model” used herein may refer to an AI / ML Model of which inference is performed entirely at the network. The term “one-sided (AI / ML) model” used herein may refer to a UE-side (AI / ML) model or a network-side (AI / ML) model. The term “two-sided (AI / ML) model” used herein may refer to a paired AI / ML Model (s) over which joint inference is performed, where joint inference comprises AI / ML Inference whose inference is performed jointly across the UE and the network, i.e, the first part of inference is firstly performed by UE and then the remaining part is performed by gNB, or vice versa.
[0037] The term “AI / ML model transfer” used herein may refer to a delivery of an AI / ML model over the air interface, either parameters of a model structure known at the receiving end or a new model with parameters. Delivery may contain a full model or a partial model. The term “model download” used herein may refer to model transfer from the network to UE. The term “model upload” used herein may refer to model transfer from UE to the network. The term “federated learning / federated training” used herein may refer to a machine learning technique that trains an AI / ML model across multiple decentralized edge nodes (e.g., UEs, gNBs) each performing local model training using local data samples. The technique requires multiple interactions of the model, but no exchange of local data samples.
[0038] The term “model activation” used herein may refer to enabling an AI / ML model for a specific function. The term “model deactivation” used herein may refer to disabling an AI / ML model for a specific function. The term “model switching” used herein may refer to deactivating a currently active AI / ML model and activating a different AI / ML model for a specific function.
[0039] As used herein, the term “AI / ML model delivery” may refer to a generic term referring to delivery of an AI / ML model from one entity to another entity in any manner. It is to be noted that an entity could mean a network node / function (e.g., gNB, LMF, etc. ) , UE, proprietary server, etc. The term “model registration” used herein may refer to a process of informing the existence of an AI / ML model to the network or to the UE with an identification, along with model description information of the AI / ML model for the network to enable life cycle management (LCM) .
[0040] The term “model update” as used herein may refer to a process of updating the model parameters and / or model structure of a model. The term “model parameter update” as used herein may refer to a process of updating the model parameters of a model.
[0041] Principles and implementations of the present disclosure will be described in detail below with reference to the figures.
[0042] FIG. 1 illustrates a schematic diagram of an example communication environment 100 in which example embodiments of the present disclosure can be implemented. In the communication environment 100, a plurality of communication devices, including a terminal device 110 and a network device 120, can communicate with each other.
[0043] In the example of FIG. 1, the terminal device 110 may be a UE and the network device 120 may be a base station serving the UE. The serving area of the network device 120 may be called a cell 102.
[0044] It is to be understood that the number of devices and their connections shown in FIG. 1 are only for the purpose of illustration without suggesting any limitation. The communication environment 100 may include any suitable number of devices configured to implementing example embodiments of the present disclosure. Although not shown, it would be appreciated that one or more additional devices may be located in the cell 102, and one or more additional cells may be deployed in the communication environment 100. It is noted that although illustrated as a network device, the network device 120 may be another device than a network device. Although illustrated as a terminal device, the terminal device 110 may be other device than a terminal device.
[0045] In the following, for the purpose of illustration, some example embodiments are described with the terminal device 110 operating as a UE and the network device 120 operating as a base station. However, in some example embodiments, operations described in connection with a terminal device may be implemented at a network device or other device, and operations described in connection with a network device may be implemented at a terminal device or other device.
[0046] In some example embodiments, a link from the network device 120 to the terminal device 110 is referred to as a downlink (DL) , while a link from the terminal device 110 to the network device 120 is referred to as an uplink (UL) . In DL, the network device 120 is a transmitting (TX) device (or a transmitter) and the terminal device 110 is a receiving (RX) device (or a receiver) . In UL, the terminal device 110 is a TX device (or a transmitter) and the network device 120 is a RX device (or a receiver) .
[0047] The communications in the communication environment 100 may conform to any suitable standards including, but not limited to, Global System for Mobile Communications (GSM) , Long Term Evolution (LTE) , LTE-Evolution, LTE-Advanced (LTE-A) , New Radio (NR) , Wideband Code Division Multiple Access (WCDMA) , Code Division Multiple Access (CDMA) , GSM EDGE Radio Access Network (GERAN) , Machine Type Communication (MTC) and the like. The embodiments of the present disclosure may be performed according to any generation communication protocols either currently known or to be developed in the future. Examples of the communication protocols include, but not limited to, the first generation (1G) , the second generation (2G) , 2.5G, 2.75G, the third generation (3G) , the fourth generation (4G) , 4.5G, the fifth generation (5G) communication protocols, 5.5G, 5G-Advanced networks, or the sixth generation (6G) networks.
[0048] In the communication environment 100, an AI / ML model can be applied to different scenarios to achieve better performances.
[0049] In some embodiments, the AI / ML model may be a two-sided model, which comprises a first part for using by the terminal device 110 and a second part for using the network device 120. The first part may be used to generate an intermediate result from an initial result and the second part may be used to generate a reconstructed result from the intermediate result. In the following, the first part may be referred to as a terminal part, UE side part, or UE part, which can be used interchangeably. The second part may be referred to as a network part, network (NW) side part, or NW part, which can be used interchangeably.
[0050] In some embodiments, the terminal device 110 can perform AI / ML based RRM measurement and event prediction. For example, cell-level measurement prediction including intra and inter-frequency may be performed using two-sided AI model (i.e., UE side and NW side model) . Inter-cell Beam-level measurement prediction for L3 Mobility may be performed using two-sided AI model. For another example, HO failure / RLF prediction and measurement events prediction may be performed using UE side model.
[0051] Taking RLF prediction as an example, the RLF causes will be determined first.
[0052] The UE shall set the rlf-Cause in the VarRLF-Report. The following will be described in detail.
[0053] If the UE declares RLF due to T310 expiry, the rlf-Cause is set as t310-Expiry. FIG. 2 illustrates a schematic diagram of T310 expiry in accordance with some embodiments of the present disclosure. When the UE detects the physical link problem, i.e., the UE receives consecutive N310 Out-of-sync indication from the lower layer, the UE will start a T310 timer. Within the T310 timer duration, if the UE receives consecutive N311 in-sync indication from the lower layer, the UE will stop the T310 timer. If the UE receives the RRCReconfiguration WithSync message from the cell group, the UE will stop the T310 timer. If the UE triggers the Radio Resource Control (RRC) re-establishment procedure, the UE will stop the T310 timer. Otherwise, if the T310 expires, if the T310 is configured at the Master Cell Group (MCG) and the UE is not under the security mode, the UE will access RRC_IDLE mode. Otherwise, if the UE is under the security mode, the UE will trigger the RRC re-establishment procedure. Otherwise, if the T310 is configured at the Secondary Cell Group (SCG) , the UE will trigger RLF.
[0054] In some cases, the UE declares RLF due to the random access problem indication from MCG Medium Access Control (MAC) . If the random access procedure was initiated for beam failure recovery (BFR) , the rlf-Cause is set as beamFailureRecoveryFailure; else, the rlf-Cause is set as randomAccessProblem.
[0055] Specifically, a MAC entity shall perform beam failure detection for each Serving Cell configured. In some cases, a Serving Cell is configured with two Beam Failure Detection-Reference Signal (BFD-RS) sets.
[0056] If beam failure instance indication for a BFD-RS set has been received from lower layers, the beamFailureDetectionTimer of the BFD-RS set is started or restarted, and BFI_COUNTER of the BFD-RS set is incremented by 1. If BFI_COUNTER of the BFD-RS set >= beamFailureInstanceMaxCount, a BFR is triggered for this BFD-RS set of the Serving Cell.
[0057] If the BFR is triggered for both BFD-RS sets of the SpCell and the BFR procedure is not successfully completed for any of the BFD-RS sets, a Random Access procedure is initiated on the SpCell.
[0058] If the Serving Cell is SpCell and the Random Access procedure initiated for BFR of both BFD-RS sets of SpCell is successfully completed, BFI_COUNTER of each BFD-RS set of SpCell is set to 0, and the BFR procedure is considered successfully completed.
[0059] If the beamFailureDetectionTimer of this BFD-RS is set to expire, or if beamFailureDetectionTimer, beamFailureInstanceMaxCount, or any of the reference signals used for beam failure detection are reconfigured by upper layers or by the BFD-RS Indication MAC CE associated with a BFD-RS set of the Serving Cell, or if the reference signal (s) associated with a BFD-RS set of the Serving Cell used for beam failure detection is changed, BFI_COUNTER of the BFD-RS set is set to 0.
[0060] If a PDCCH addressed to C-RNTI indicating uplink grant for a new transmission is received for the HARQ process used for the transmission of the Enhanced BFR MAC CE or Truncated Enhanced BFR MAC CE which contains beam failure recovery information of this BFD-RS set of the Serving Cell, BFI_COUNTER of the BFD-RS set is set to 0, and the Beam Failure Recovery procedure is considered successfully completed for this BFD-RS set and cancel all the triggered BFRs of this BFD-RS set of the Serving Cell.
[0061] If the the Serving Cell is SCell and the SCell is deactivated as specified in clause 5.9, BFI_COUNTER of each BFD-RS set of SCell is set to 0. The Beam Failure Recovery procedure is considered successfully completed and all the triggered BFRs of all BFD-RS sets of the Serving Cell are cancelled.
[0062] In some other cases, the Serving Cell is not configured with two BFD-RS sets. If beam failure instance indication has been received from lower layers, the beamFailureDetectionTimer is started or restarted, BFI_COUNTER is incremented by 1. Further, if BFI_COUNTER >= beamFailureInstanceMaxCount, if the Serving Cell is SCell, a BFR for this Serving Cell is triggered; else if the Serving Cell is PSCell and, the SCG is deactivated. If beam failure of the PSCell has not been indicated to upper layers since the SCG was deactivated or since the deactivated SCG was last reconfigured with BFD-RS, beam failure of the PSCell is indicated to upper layers. If the Serving Cell is not SCell, a Random Access procedure is initiated on the SpCell.
[0063] It is noted that after beam failure is indicated to upper layers, the UE may stop the beamFailureDetectionTimer and lower layer beam failure indication while BFI_COUNTER >= beamFailureInstanceMaxCount for the deactivated SCG.
[0064] If the beamFailureDetectionTimer expires, or if beamFailureDetectionTimer, beamFailureInstanceMaxCount, or any of the reference signals used for beam failure detection is reconfigured by upper layers associated with this Serving Cell, or if the reference signal (s) associated with this Serving Cell used for beam failure detection is changed, BFI_COUNTER is set to 0.
[0065] If the Serving Cell is SpCell and the Random Access procedure initiated for SpCell beam failure recovery is successfully completed, BFI_COUNTER is set to 0, the beamFailureRecoveryTimer is stopped (if configured) , and the Beam Failure Recovery procedure is considered successfully completed. Else, a Random Access procedure is initiated on the SpCell.
[0066] If the beamFailureDetectionTimer expires, or if beamFailureDetectionTimer, beamFailureInstanceMaxCount, or any of the reference signals used for beam failure detection is reconfigured by upper layers associated with this Serving Cell, or if the reference signal (s) associated with this Serving Cell used for beam failure detection is changed, BFI_COUNTER is set to 0.
[0067] If the Serving Cell is SpCell and the Random Access procedure initiated for SpCell beam failure recovery is successfully completed, BFI_COUNTER is set to 0, the beamFailureRecoveryTimer is stopped (if configured) , and the Beam Failure Recovery procedure is considered successfully completed.
[0068] The above describes RLF causes such as T310 expiry, beam failure recovery failure, random access problem. There are some other RLF causes which will be described below.
[0069] If the UE declares RLF due to the reaching of maximum number of retransmissions from the MCG RLC, the rlf-Cause is set as rlc-MaxNumRetx. If the UE declares RLF due to consistent uplink LBT failures, the rlf-Cause is set as lbtFailure. If the IAB-MT declares RLF due to the reception of a BH RLF indication on BAP entity, the rlf-Cause is set as bh-rlfRecoveryFailure.
[0070] Assuming a Radio Link Control Service Data Unit (RLC SDU) or an RLC SDU segment is considered for retransmission, the transmitting side of the AM RLC entity shall perform the following actions. If the RLC SDU or RLC SDU segment is considered for retransmission for the first time, the RETX_COUNT associated with the RLC SDU is set to zero. Else, if it (the RLC SDU or the RLC SDU segment that is considered for retransmission) is not pending for retransmission already and the RETX_COUNT associated with the RLC SDU has not been incremented due to another negative acknowledgment in the same STATUS PDU, the RETX_COUNT is incremented. If RETX_COUNT = maxRetxThreshold, it is indicated to upper layers that max retransmission has been reached.
[0071] If the UE declares RLF due to T312 expiry, the rlf-Cause is set as t312-Expiry. Assuming T312 is configured in MCG. Upon triggering a measurement report for a configured measurement identity with T312 set to true, while T310 in PCell is running.
[0072] If T312 is configured in SCG and set to true, upon triggering a measurement report for a configured measurement identity, while T310 in PSCell is running.
[0073] Upon at least one of the following is met, the rlf-Cause is set as t312-Expiry: upon receiving N311 consecutive in-sync indications from lower layers for the SpCell, receiving RRCReconfiguration with reconfiguration WithSync for that cell group; upon reception of MobilityFromNRCommand; upon initiating the connection re-establishment procedure; upon the reconfiguration of rlf-TimersAndConstant; upon initiating the MCG failure information procedure; upon conditional reconfiguration execution i.e. when applying a stored RRCReconfiguration message including reconfiguration WithSync for that cell group; upon the expiry of T310 in corresponding SpCell; and upon SCG release, if the T312 is kept in SCG.
[0074] If T312 is kept in MCG, the MCG failure information procedure or the connection re-establishment procedure is initiated. If T312 is kept in SCG, E-UTRAN / NR is informed about the SCG RLF by initiating the SCG failure information procedure.
[0075] AI / ML models can be used to predict RLF based on the above RLF causes. Given that, embodiments of the present disclosure provide a solution for predicting RLF in accordance with some embodiments of the present disclosure. The terminal device obtains radio link information associated with a radio link of the terminal device. The terminal device generates a set of prediction results for at least one predicted RLF based on the radio link information by using a set of AI / ML models. A prediction result comprises information about a predicted RLF. In this way, network performance may be optimized based on RLF prediction results, thereby reducing communication interruptions and maintaining network stability and reliability.
[0076] Reference is now made to FIG. 3, which illustrates a signaling flow 300 of predicting RLF in accordance with some embodiments of the present disclosure. For the purposes of discussion, the signaling flow 300 will be discussed with reference to FIG. 1, for example, by using the terminal device 110 and the network device 120.
[0077] In some example embodiments, the terminal device 110 may transmit 302 capability information to the network device 120. The network device 120 may receive 304 the capability information. The capability information is related to AI / ML based prediction for RLF. The capability information may comprise correspondence between one or more cause values of RLF and one or more AI / ML models, i.e., how many / which cause failure of RLF can be supported by each AI / ML model.
[0078] The capability information may comprise a maximum duration for which a RLF for a cause value can be predicted. For each cause value, the capability information may comprise a maximum number of AI / ML models stored for a same cause value, and a maximum number of AI / ML models for which life cycle management (LCM) can be performed for a same cause value. In total, the capability information may comprise a maximum number of AI / ML models stored for different cause values, and a maximum number of AI / ML models for which LCM can be performed for different cause values.
[0079] Alternatively, or additionally, the capability information may comprise one or more cause values of RLF to which an AI / ML model can be applied and LCM can be performed, a capability of the terminal device to perform AI / ML based RLF prediction and traditional RLF for the same cause value at the same time or within a short duration, a capability of the terminal device to perform AI / ML based RLF prediction and traditional RLF for different cause values at the same time or within a short duration, a capability of the terminal device to continue model inference after a predicted RLF is determined for each cause value, or a capability of the terminal device to trigger multiple RLF prediction for multiple cause values if the different cause values of RLF inferenced by different AI / ML models.
[0080] In some example embodiments, in response to a request for the capability information from the network device 120, the terminal device 110 may transmit the capability information to the network device 120. For example, the ue-CapabilityAIML-RequestList contains a UE-CapabilityAIML-Request with aiml-Type set to RlfPredict. If the terminal device 110 supports AI / ML based measurement report, a UE-CapabilityAIML-Container of a type UE-AIML-Capability is included in the ue-CapabilityAIML-ContainerList and with the aiml-Type set to RlfPredict (UE autonomously / proactive / dynamic report AI / ML based capability / UAI) .
[0081] Alternatively, or additionally, in response to activation of AI / ML based prediction for RLF, the terminal device 110 may transmit the capability information to the network device 120. For example, the terminal device 110 may be triggered to report the related capability information when it decides to activate AI / ML based RlfPredict. In details, such information may be included within UE Assistance Information or newly defined capability reporting message.
[0082] Alternatively, or additionally, in response to an update of capability of the terminal device in AI / ML based prediction for RLF, the terminal device 110 may transmit the capability information to the network device 120.
[0083] The above describes the terminal device 110 may transmit the capability information to the network device 120. To activate AI / ML based RLF prediction, the terminal device 110 may be configured to report model training data if the model training is performed at the network device 120.
[0084] Continuing with reference to FIG. 3, the network device 120 may transmit 306 a data collection configuration to the terminal device 110. The terminal device 110 may receive 308 the data collection configuration. Then, the terminal device 110 may collect data for training.
[0085] In some example embodiments, the terminal device 110 may first obtain at least one data set of a first data set (also referred to as a data-set A) or a second data set (also referred to as a data-set B) for training the set of AI / ML models. An entry in the first data set may be associated with a historical RLF at the terminal device 110. An entry in the second data set may be associated with a near failure state where a RLF is approaching to occur at the terminal device 110. The near failure state may include a near RLF state.
[0086] In some example embodiments, the terminal device 110 may obtain the first data set. The entry in the first data set may comprise at least one of: a cause value of the historical RLF, a time at which the historical RLF is declared, respective identifications of one or more cells, an indication of a frequency at which the terminal device operates, an active bandwidth part (BWP) numerology, a configuration for uplink and downlink communications in time domain, a geographical position of the terminal device, a moving speed of the terminal device, a moving direction of the terminal device, or a measurement result of a radio link management reference signal.
[0087] In some example embodiments, the terminal device 110 may a configuration of a first data collection timer (also referred to as a timer A) from the network device 120. Then, the terminal device 110 may start to collect entries in the first data set upon starting of the first data collection timer and terminate collecting entries upon expiration of the first data collection timer.
[0088] In some example embodiments, if it is determined that a number of entries in the first data set is below a minimum entry threshold, the terminal device 110 may transmit, to the network device 120, an indication of data insufficiency of the first data set.
[0089] In some example embodiments, if it is determined that a number of entries in the first data set exceeds a maximum entry threshold, the terminal device 110 may transmit, to a network device 120, an indication of a large data set to request a user plane transmission with a specific logical channel identification. Alternatively, or additionally, the terminal device 110 may determine a subset of the first data set such that a number of entries in the subset is equal to the maximum entry threshold.
[0090] For example, when the timer A starts, the terminal device 110 may start to record each entry of the data-set A. When the timer A expires, the terminal device 110 may stop to record each entry of the data-set A. If the number of entries within the data-set A is less than the configured minimum entry threshold, the terminal device 110 may not report the data-set A directly, instead, the terminal device 110 may send an indication towards the network device 120 to indicate the entry of data-set A is not enough. If the number of entries within the data-set A is greater than the configured maximum entry threshold, the terminal device 110 may transmit the indication of the large data set or segment the large data-set A so that keep the number of first N entries of which N equals to the configured maximum entry threshold. Furthermore, the terminal device 110 may be configured with an extension timer, e.g., the timer B. When the timer B starts, the terminal device 110 may continue to record each entry of data-set A, when the number of entries reaches the configured minimum entry threshold, the terminal device 110 may stop the timer B. When the timer B expires, the terminal device 110 may stop to record each entry of data-set A.
[0091] In some example embodiments, the terminal device 110 may obtain the second data set. The entry in the second data set may comprise at least one of: a cause value of the RLF corresponding to the near failure state, a time at which the near failure state is declared, respective identifications of one or more cells, an indication of a frequency at which the terminal device operates, an active bandwidth part (BWP) numerology, a configuration for uplink and downlink communications in time domain, a geographical position of the terminal device, a moving speed of the terminal device, a moving direction of the terminal device, a measurement result of a radio link management reference signal, or an indication on whether the corresponding RLF occurs after the near failure state is declared.
[0092] In some example embodiments, the terminal device 110 may receive, from a network device 120, a configuration of a second data collection timer (also referred to as a timer B) . Then, the terminal device 110 may start to collect entries in the second data set upon starting of the second data collection timer and terminate collecting entries upon expiration of the second data collection timer.
[0093] In some example embodiments, if it is determined that a number of entries in the second data set is below a minimum entry threshold, the terminal device 110 may transmit, to the network device 120, an indication of data insufficiency of the second data set.
[0094] In some example embodiments, if it is determined that a number of entries in the second data set exceeds a maximum entry threshold, the terminal device 110 may transmit, to the network device 120, an indication of a large data set to request a user plane transmission with a specific logical channel identification. Alternatively, or additionally, the terminal device 110 may determine a subset of the second data set such that a number of entries in the subset is equal to the maximum entry threshold.
[0095] For example, when the timer B starts, the terminal device 110 may start to record each entry of the data-set B. When the timer B expires, the terminal device 110 may stop to record each entry of the data-set B. If the number of entries within the data-set B is less than the configured minimum entry threshold, the terminal device 110 may not report the data-set A directly, instead, the terminal device 110 may send an indication towards the network device 120 to indicate the entry of data-set B is not enough. If the number of entries within the data-set B is greater than the configured maximum entry threshold, the terminal device 110 may transmit the indication of the large data set or segment the large data-set B so that keep the number of first N entries of which N equals to the configured maximum entry threshold. Furthermore, the terminal device 110 may be configured with an extension timer, e.g., the timer C. When the timer C starts, the terminal device 110 may continue to record each entry of data-set B, when the number of entries reaches the configured minimum entry threshold, the terminal device 110 may stop the timer C. When the timer C expires, the terminal device 110 may stop to record each entry of data-set B.
[0096] In some example embodiments, in response to receiving, from the network device 120, an indication to activate training data collection, the terminal device 110 may start to collect entries in the at least one data set. In response to receiving, from the network device 120, an indication to activate training data reporting, the terminal device 110 may terminate collecting entries. The indication mentioned in this case may be carrier via at least one of: Uplink Control Information / Downlink Control Information (UCI / DCI) , UL MAC CE / DL MAC CE or RRC messages.
[0097] For example, instead of configuring the timers mentioned above, the network device 120 may dynamically indicate towards the terminal device 110 when to record data set-A and data set-B. Specifically, the network device 120 may indicate to activate RLF model training data record. When receiving this indication, the terminal device 110 may start to record the cases mentioned above for data set-A and data set-B. The network device 120 may indicate to activate RLF model training data report. When receiving this indication, the terminal device 110 may stop to record the cases mentioned above for data set-A and data set-B, instead, the terminal device 110 may report the recorded data set-Aand / or data set-B towards the network device 120.
[0098] In some example embodiments, the terminal device 110 may transmit, to the network device 120, the at least one data set via at least one of: a signaling radio bearer (SRB) defined for training data collection, a user plane transmission with a specific data radio bearer (DRB) , a radio bearer specific to AI / ML, or a logical channel specific to AI / ML.
[0099] In some example embodiments, if it is determined that a near failure criterion for a cause value is met, the terminal device 110 may declare a near failure state where a RLF with the cause value is approaching to occur at the terminal device 110.
[0100] In some example embodiments, if the cause value is T310 expiry, the near failure criterion may comprise that a number of consecutive out-of-sync indications reaches a maximum value lower than N310. For example, if the number of consecutive out-of-sync indication reaches the newly configured maximum value, then it may be regarded as one condition to declare rlf-near-T310expiry.
[0101] Alternatively, or additionally, if the cause value is T310 expiry, the near failure criterion may comprise that a number of consecutive out-of-sync indications reaches a percentage value of N310. For example, if the number of consecutive out-of-sync indication reaches P1*N310, where P1 represents a percentage value, then it may be regarded as one condition to declare rlf-near-T310expiry.
[0102] Alternatively, or additionally, if the cause value is T310 expiry, the near failure criterion may comprise that a number of consecutive out-of-sync indications reaches a value of N310 minus an offset. For example, if the number of consecutive out-of-sync indication reaches N310-off1, where off1 represents an offset value, then it may be regarded as one condition to declare rlf-new-T310expiry.
[0103] Alternatively, or additionally, if the cause value is T310 expiry, the near failure criterion may comprise that a number of consecutive in-sync indications reaches a maximum value lower than N311. For example, if the number of consecutive in-sync indication reaches the newly configured maximum value, then it may be regarded as one condition to declare rlf-new-T310expiry.
[0104] Alternatively, or additionally, if the cause value is T310 expiry, the near failure criterion may comprise that a number of consecutive in-sync indications reaches a percentage value of N311. For example, if the number of consecutive in-sync indication reaches P2*N311, where P2 represents a percentage value, then it may be regarded as one condition to declare rlf-new-T310expiry.
[0105] Alternatively, or additionally, if the cause value is T310 expiry, the near failure criterion may comprise that a number of consecutive in-sync indications reaches a value of N311 minus an offset. For example, if the number of consecutive in-sync indication reaches N310-off2, where off2 represents an offset value, then it may be regarded as one condition to declare rlf-new-T310expiry.
[0106] Alternatively, or additionally, if the cause value is T310 expiry, the near failure criterion may comprise that a timer which is started upon the terminal device receiving N311 out-of-sync indications from a lower layer expires while T310 does not expire. For example, when receiving N311 out-of-sync indications from its lower layer, the terminal device 110 may start a new timer. When receiving consecutive N311 in-sync indication from its lower layer, the terminal device 110 may stop the new timer. When the new timer expires while T310 does not expire (e.g., the terminal device 110 stops T310 successfully) , it may be regarded as one condition to declare rlf-new-T310expiry.
[0107] Alternatively, or additionally, if the cause value is T310 expiry, the near failure criterion may comprise that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T310. For example, if the duration for receiving consecutive N311 in-sync indication from the lower layer of the terminal device 110 exceed the duration T310*P3, where P3 represents a percentage value, it may be regarded as one condition to declare rlf-new-T310expiry.
[0108] Alternatively, or additionally, if the cause value is T310 expiry, the near failure criterion may comprise that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T310 minus an offset. For example, if the duration for receiving consecutive N311 in-sync indication from the lower layer of the terminal device 110 exceed the duration T310-off3, where off3 represents an offset value, it may be regarded as one condition to declare rlf-new-T310expiry.
[0109] In some example embodiments, if the cause value is T312 expiry, the near failure criterion may comprise that a timer which is started upon the terminal device receiving N311 out-of-sync indications from a lower layer expires while T312 does not expire. For example, when receiving N311 out-of-sync indications from its lower layer, the terminal device 110 starts the new timer. When receiving consecutive N311 in-sync indications from its lower layer, the terminal device 110 stops the new timer. When the new timer expires while T312 does not expire (e.g., the terminal device 110 stops T312 successfully) , it may be regarded as one condition to declare rlf-near-T312expiry.
[0110] Alternatively, or additionally, if the cause value is T312 expiry, the near failure criterion may comprise that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T312. For example, if the duration for receiving consecutive N311 in-sync indication from the lower layer of the terminal device 110 exceeds the duration T312*P4, where P4 represents a percentage value, it may be regarded as one condition to declare rlf-near-T312expiry.
[0111] Alternatively, or additionally, if the cause value is T312 expiry, the near failure criterion may comprise that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T312 minus an offset. For example, if the duration for receiving consecutive N311 in-sycn indication from the lower layer of the terminal device 110 exceeds the duration T312-off4, where off4 represents an offset value, it may be regarded as one condition to declare rlf-near-T312expiry.
[0112] In some example embodiments, if the cause value is beam failure recovery failure, the near failure criterion may comprise that a number of consecutive beam failure indications from a lower layer reaches a maximum value lower than a beam failure instance max count. For example, if the number of consecutive BFI reported from lower layer reaches the configured newly maximum value, it may be regarded as one condition to declare rlf-near-beamfailurecoveryfailure.
[0113] Alternatively, or additionally, if the cause value is beam failure recovery failure, the near failure criterion may comprise that a number of consecutive beam failure indications from a lower layer reaches a percentage value of the beam failure instance max count. For example, if the number of consecutive BFI reported from lower layer reaches P5*beamFailureInstanceMaxCount, where P5 represents a percentage value, then it may be regarded as one condition to declare rlf-near-beamfailurecoveryfailure.
[0114] Alternatively, or additionally, if the cause value is beam failure recovery failure, the near failure criterion may comprise that a number of consecutive beam failure indications from a lower layer reaches a value of the beam failure instance max count minus an offset. For example, if the number of consecutive BFI reported from lower layer reaches beamFailureInstanceMaxCount-off5, where off5 represents an offset value, then it may be regarded as one condition to declare rlf-near-beamfailurecoveryfailure.
[0115] In some example embodiments, if the cause value is RLC retransmission, the near failure criterion may comprise that a number of RLC retransmissions reaches a maximum retransmission threshold. For example, if the number of RLC retransmission number RETX_COUNT reaches the newly configured maximum value, it may be regarded as one condition to declare rlf-nea-rlcmaxretx.
[0116] Alternatively, or additionally, if the cause value is RLC retransmission, the near failure criterion may comprise that a number of RLC retransmissions reaches a percentage value of the maximum retransmission threshold. For example, if the number of RLC retransmission RETX_COUNT from the lower layer reaches P6*maxRetxThreshold, where P6 represents a percentage value, then it may be regarded as one condition to declare rlf-near-rlcmaxretx.
[0117] Alternatively, or additionally, if the cause value is RLC retransmission, the near failure criterion may comprise that a number of RLC retransmissions reaches a value of the maximum retransmission threshold minus an offset. For example, if the number of RLC retransmission RETX_COUNT from the lower layer reaches maxRetxThreshold-off6, where off6 represents an offset value, then it may be regarded as one condition to declare rlf-near-rlcmaxretx.
[0118] The above has described that the terminal device 110 collects the data for training. The network device 120 may train AI / ML models with the data, then the terminal device 110 may use the trained AI / ML models to perform model inference for RLF prediction with specific network configuration. The following will be described in detail.
[0119] Referring back to FIG. 3, the terminal device 110 may transmit 312 the data for training to the network device 120. The network device 120 may receive 314 the data.
[0120] In some example embodiments, the network device 120 may transmit 316 state information about activation or deactivation of the set of AI / ML models for RLF prediction to the terminal device 110. The terminal device 110 may receive 318 the state information. For example, the network device 120 may configure the terminal device 110 when / how to activate / deactivate the corresponding AI / ML model for RLF prediction. The configuration may comprise at least one of the following granularities: per cell, per UE, per cause value of RLF or per model. If the terminal device 110 has the capability to perform AI / ML based prediction for RLF, it may use the trained AI / ML models to perform model inference.
[0121] In some example embodiments, the network device 120 may transmit 316 dynamic state information about activation or deactivation of the set of AI / ML models for RLF prediction to the terminal device 110. The dynamic state information may comprise at least one of: identifications of the set of AI / ML models, one or more cause values of RLF for prediction, an activation duration timer, a number of AI / ML models concurrently activated for a cause value of RLF, or an indication to activate or deactivate the set of AI / ML models for RLF prediction.
[0122] In some example embodiments, the network device 120 may transmit 316 periodical state information about activation or deactivation of the set of AI / ML models for RLF prediction to the terminal device 110. The periodical state information may indicate a periodical configuration for activating or deactivating the set of AI / ML models for RLF prediction. The periodical configuration may comprise a starting timing to activate the set of AI / ML models, such as a specific timing stamp, SFN and slot offset configuration.
[0123] Alternatively, or additionally, the periodical configuration may comprise an activation duration of the set of AI / ML models. As an example, the activation duration may be implemented via an activation timer. If the activation timer expires or the corresponding cause of RLF is predicted, or the real cause of RLF is predicted, the terminal device 110 may deactivate the corresponding model.
[0124] Alternatively, or additionally, the periodical configuration may comprise a deactivation duration of the set of AI / ML models, or a configuration validation timer. As an example, the deactivation may be implemented via a deactivation timer. If the deactivation timer expires, the terminal device 110 may re-activate the corresponding model inference.
[0125] Alternatively, or additionally, the periodical configuration may comprise a configuration validation timer. As an example, a configuration validation timer is set. Within the timer duration, the terminal device 110 may perform the activation / deactivation of AI / ML model inference for RLF prediction based on the configuration. If the timer expires, or optionally, the terminal device 110 successfully performs handover / RRC-Reestablishment / RACH, the terminal device 110 may send an request for configuration updating. Moreover, the configuration may include at least one of the following parameters: a model ID or a cause value of RLF.
[0126] Continue with reference to FIG. 3, the terminal device 110 may predict 320 an RLF. Specifically, the terminal device 110 may obtain radio link information associated with a radio link of the terminal device 110. Then the terminal device 110 may generate a set of prediction results for at least one predicted RLF based on the radio link information by using a set of AI / ML models trained.
[0127] In some example embodiments, the prediction result in the set of prediction results may comprise at least one of: a cause value of the predicted RLF, an indication of a predicted time at which the predicted RLF is to occur, or a predicted possibility that the predicted RLF is to occur at the predicted time.
[0128] In some example embodiments, the radio link information may comprise a number of indications associated with radio link failures counted at a predetermined layer. For example, the number of out-of-sync / in-sync / BFI / RLC retransmission indication counted by the AI / ML handling entity anchored at specific specification layer. The specification layer may include Packet Data Convergence Protocol (PDCP) layer / RRC layer / Service Data Adaptation Protocol (SDAP) layer or newly layer defined in between PDCP layer and RRC layer.
[0129] Alternatively, or additionally, the radio link information may comprise respective identifications of one or more cells, a measurement result of a radio link management reference signal (e.g., the corresponding Reference Signal Received Power (RSRP) / Signal-to-Interference-plus-Noise Ratio (SINR) / down link radio link quality / RSRQ measurement result of Radio Link Monitoring-Reference Signal (RLM-RS) ) , a geographical position of the terminal device, a moving speed of the terminal device 110, a moving direction of the terminal device 110, an indication of a frequency at which the terminal device operates, an active bandwidth part (BWP) numerology, or a configuration for uplink and downlink communications in time domain (e.g., TDD-UL-DL-Config) .
[0130] The network device 120 may configure a prediction reporting configuration, which used to help the terminal device 110 to determine how to report the prediction matrix for the AI / ML model inference result of each round.
[0131] In some example embodiments, in response to a first prediction result in the set of prediction results being generated, the terminal device 110 may start a first timer. In response to expiration of the first timer, the terminal device 110 may transmit at least one portion of the set of prediction results to the network device 120.
[0132] In some example embodiments, a second prediction result in the set of prediction results may comprise a predicted possibility below a first possibility threshold (also referred to as a possibility threshold A) . The predicted possibility may be a prediction that a predicted RLF is to occur at a predicted time. The second prediction result may be not transmitted to the network device 120. For example, the possibility threshold A is used for each cause value / model ID. When the predicted possibility for specific cause value of RLF is below the configured possibility threshold A (for different cause value / model ID, the possibility threshold A may be the same or different) , the terminal device 110 does not need to report the predicted result.
[0133] In some example embodiments, in response to that a predicted possibility of a third prediction result exceeds a second possibility threshold (also referred to as a possibility threshold B) , the terminal device 110 may transmit the third prediction result to the network device 120 without waiting for the expiration of the first timer. For example, the possibility threshold B is used for each cause value / model ID. When the predicted possibility for specific cause value of RLF is above the configured possibility threshold B (for different cause value / model ID, the possibility threshold B may be the same or different) , the terminal device 110 does not need to report the predicted result of other cause values, instead, the terminal device 110 may send the predicted RLF report with corresponding cause value.
[0134] For example, the terminal device 110 may be configured with an entry timer. For each RLF prediction reporting, when a first prediction entry is generated, the terminal device 110 may trigger the entry timer, within a timer duration. If multiple prediction entries are generated, the terminal device 110 may generate a prediction matrix with the multiple entries. When the entry timer expires, the terminal device 110 may perform prediction reporting. The terminal device 110 may determine whether to perform prediction reporting for specific prediction entry based on the configured threshold A and configured threshold B. If the predicted possibility for one specific prediction entry is below the configured threshold A, then the terminal device 110 may not report the corresponding prediction entry. Otherwise, if the predicted possibility for one specific prediction entry is above the configured threshold B, then the terminal device 110 may only report the corresponding prediction entry.
[0135] In some example embodiments, in response to a fourth prediction result indicating a predicted time at which a predicted RLF is to occur is below a time threshold, the terminal device 110 may transmit the fourth prediction result to the network device 120 by using an emergency reporting resource.
[0136] For example, for one specific entry, if the corresponding predicted prior time is below than the prior time threshold A, then the terminal device 110 may not include the corresponding entry into the prediction matrix, instead, the terminal device 110 would use the configured Scheduling Request (SR) resource with corresponding cause value and corresponding predicted time to perform reporting.
[0137] For example, an emergency predicted time threshold A is used for each cause value / model ID. When the predicted prior time for specific cause value of RLF is less than the configured time threshold A (for each cause value / model ID, the time threshold A may be the same or different) , the terminal device 110 may perform emergency reporting for the prediction of the specific cause value (s) .
[0138] For example, emergency reporting resource includes different SR resource corresponding to the combination of different cause value and predicted prior time, e.g. rlf-rlcmaxretx and 1ms prior time correspond to one SR resource.
[0139] Furthermore, an ordinary reporting SR configuration may be used. If the prediction result reporting is carried via UL MAC CE, the corresponding SR configuration may include the specific SR resource used for UL MAC CE transmission resource requesting.
[0140] When the terminal device 110 generates the corresponding prediction matrix, it will perform prediction reporting. In some example embodiments, the at least one portion of the set of prediction results may be transmitted via an uplink medium access control (MAC) control element (CE) with multiple gathering entries. Alternatively, or additionally, when the matrix may be reported via a table format, the at least one portion of the set of prediction results may be transmitted via a RRC message.
[0141] In some example embodiments, if it is determined that a prediction criterion for a predicted RLF is met, the terminal device 110 may generate a prediction result for the predicted RLF. As shown in FIG. 3, the terminal device 110 may transmit 322 a report associated with the prediction result to the network device 120. The network device 120 then may receive the report.
[0142] In some example embodiments, if the predicted RLF has a cause value of T310 expiry, the prediction criterion may comprise that a number of consecutive out-of-sync indications reaches a maximum value lower than N310. For example, if the number of consecutive out-of-sync indication reaches the newly configured maximum value, then one rlf-predicted-T310expiry entry may be generated.
[0143] Alternatively, or additionally, if the predicted RLF has a cause value of T310 expiry, the prediction criterion may comprise that a number of consecutive out-of-sync indications reaches a percentage value of N310. For example, if the number of consecutive out-of-sync indication reaches P10*N310, where P10 represents a percentage value, then one rlf-predicted-T310expiry entry may be generated.
[0144] Alternatively, or additionally, if the predicted RLF has a cause value of T310 expiry, the prediction criterion may comprise that a number of consecutive out-of-sync indications reaches a value of N310 minus an offset. For example, if the number of consecutive out-of-sync indication reaches N310-off10, where off10 represents an offset value, then one rlf-predicted-T310expiry entry may be generated.
[0145] Alternatively, or additionally, if the predicted RLF has a cause value of T310 expiry, the prediction criterion may comprise that a number of consecutive in-sync indications reaches a maximum value lower than N311. For example, if the number of consecutive in-sync indication reaches the newly configured maximum value, then one rlf-predicted-T310expiry entry may be generated.
[0146] Alternatively, or additionally, if the predicted RLF has a cause value of T310 expiry, the prediction criterion may comprise that a number of consecutive in-sync indications reaches a percentage value of N311. For example, if the number of consecutive in-sync indication reaches P20*N311, where P20 represents a percentage value, then one rlf-predicted-T310expiry entry may be generated.
[0147] Alternatively, or additionally, if the predicted RLF has a cause value of T310 expiry, the prediction criterion may comprise that a number of consecutive in-sync indications reaches a value of N311 minus an offset. For example, if the number of consecutive in-sync indication reaches N310-off20, where off20 represents an offset value, then one rlf-predicted-T310expiry entry may be generated.
[0148] Alternatively, or additionally, if the predicted RLF has a cause value of T310 expiry, the prediction criterion may comprise that a timer which is started upon the terminal device receiving an out-sync indications from a lower layer expires while T310 does not expire. For example, when receiving in-sync indication from its lower layer, the terminal device 110 may start a new timer. When receiving consecutive N311 in-sync indication from its lower layer, the terminal device 110 may stop the new timer. When the new timer expires while T310 does not expire (e.g., the terminal device 110 stops T310 successfully) , then one rlf-predicted-T310expiry entry may be generated.
[0149] Alternatively, or additionally, if the predicted RLF has a cause value of T310 expiry, the prediction criterion may comprise that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T310. For example, if the duration for receiving consecutive N311 in-sycn indication from the lower layer of the terminal device 110 exceed the duration T310*P30, where P30 represents a percentage value, then one rlf-predicted-T310expiry entry may be generated.
[0150] Alternatively, or additionally, if the predicted RLF has a cause value of T310 expiry, the prediction criterion may comprise that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T310 minus an offset. For example, if the duration for receiving consecutive N311 in-sycn indication from the lower layer of the terminal device 110 exceed the duration T310-off30, where off30 represents an offset value, then one rlf-predicted-T310expiry entry may be generated.
[0151] In some example embodiments, if the predicted RLF has a cause value of T312 expiry, the prediction criterion may comprise that a timer which is started upon the terminal device 110 receiving N311 out-of-sync indications from a lower layer expires while T312 does not expire. For example, when receiving N311 out-of-sync indications from its lower layer, the terminal device 110 starts the new timer. When receiving consecutive N311 in-sync indication from its lower layer, the terminal device 110 stops the new timer. When the new timer expires while T312 does not expire (e.g., the terminal device 110 stops T312 successfully) , then one rlf-predicted-T310expiry entry may be generated.
[0152] Alternatively, or additionally, if the predicted RLF has a cause value of T312 expiry, the prediction criterion may comprise that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T312. For example, if the duration for receiving consecutive N311 in-sycn indication from the lower layer of the terminal device 110 exceeds the duration T312*P40, where P40 represents a percentage value, then one rlf-predicted-T310expiry entry may be generated.
[0153] Alternatively, or additionally, if the predicted RLF has a cause value of T312 expiry, the prediction criterion may comprise that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T312 minus an offset. For example, if the duration for receiving consecutive N311 in-sycn indication from the lower layer of the terminal device 110 exceeds the duration T312-off40, where off40 represents an offset value, then one rlf-predicted-T310expiry entry may be generated.
[0154] In some example embodiments, if the predicted RLF has a cause value of beam failure recovery failure, the prediction criterion may comprise that a number of consecutive beam failure indications from a lower layer reaches a maximum value lower than a beam failure instance max count. For example, if the number of consecutive BFI reported from lower layer reaches the configured newly maximum value, then one rlf-predicted-T310expiry entry may be generated.
[0155] Alternatively, or additionally, if the predicted RLF has a cause value of beam failure recovery failure, the prediction criterion may comprise that a number of consecutive beam failure indications from a lower layer reaches a percentage value of the beam failure instance max count. For example, if the number of consecutive BFI reported from lower layer reaches P50*beamFailureInstanceMaxCount, where P50 represents a percentage value, then one rlf-predicted-T310expiry entry may be generated.
[0156] Alternatively, or additionally, if the predicted RLF has a cause value of beam failure recovery failure, the prediction criterion may comprise that a number of consecutive beam failure indications from a lower layer reaches a value of the beam failure instance max count minus an offset. For example, if the number of consecutive BFI reported from lower layer reaches beamFailureInstanceMaxCount-off50, where off50 represents an offset value, then one rlf-predicted-T310expiry entry may be generated.
[0157] In some example embodiments, if the predicted RLF has a cause value of a RLC retransmission, the prediction criterion may comprise that a number of RLC retransmissions reaches a maximum retransmission threshold. For example, if the number of RLC retransmission number RETX_COUNT reaches the newly configured maximum value, then one rlf-predicted-T310expiry entry may be generated.
[0158] Alternatively, or additionally, if the predicted RLF has a cause value of a RLC retransmission, the prediction criterion may comprise that a number of RLC retransmissions reaches a percentage value of the maximum retransmission threshold. For example, if the number of RLC retransmission RETX_COUNT from the lower layer reaches P6*maxRetxThreshold, where P6 represents a percentage value, then one rlf-predicted-T310expiry entry may be generated.
[0159] Alternatively, or additionally, if the predicted RLF has a cause value of a RLC retransmission, the prediction criterion may comprise that a number of RLC retransmissions reaches a value of the maximum retransmission threshold minus an offset. For example, if the number of RLC retransmission RETX_COUNT from the lower layer reaches maxRetxThreshold-off6, where off6 represents an offset value, then one rlf-predicted-T310expiry entry may be generated.
[0160] In summary, the embodiments of the present disclosure provide a solution for AI / ML based RLF prediction. Network performance may be optimized based on RLF prediction results, thereby reducing communication interruptions and maintaining network stability and reliability.
[0161] FIG. 4 illustrates a flowchart of a communication method 400 implemented at a terminal device in accordance with some embodiments of the present disclosure. For the purpose of discussion, the method 400 will be described from the perspective of the terminal device 110 in FIG. 1.
[0162] At block 410, the terminal device 110 obtains radio link information associated with a radio link of the terminal device.
[0163] At block 420, the terminal device 110 generates a set of prediction results for at least one predicted radio link failure based on the radio link information by using a set of AI / ML models. A prediction result comprises information about a predicted radio link failure.
[0164] In some example embodiments, the prediction result in the set of prediction results comprises at least one of: a cause value of the predicted RLF, an indication of a predicted time at which the predicted RLF is to occur, or a predicted possibility that the predicted RLF is to occur at the predicted time.
[0165] In some example embodiments, the radio link information comprises at least one of: a number of indications associated with radio link failures counted at a predetermined layer, respective identifications of one or more cells, a measurement result of a radio link management reference signal, a geographical position of the terminal device, a moving speed of the terminal device, a moving direction of the terminal device, an indication of a frequency at which the terminal device operates, an active bandwidth part (BWP) numerology, or a configuration for uplink and downlink communications in time domain.
[0166] In some example embodiments, the terminal device 110 is further caused to: in response to a first prediction result in the set of prediction results being generated, start a first timer; and in response to expiration of the first timer, transmit at least one portion of the set of prediction results to a network device.
[0167] In some example embodiments, a second prediction result in the set of prediction results comprises a predicted possibility below a first possibility threshold, the predicted possibility is a prediction that a predicted RLF is to occur at a predicted time, and the second prediction result is not transmitted to the network device.
[0168] In some example embodiments, the terminal device 110 is further caused to: in response to that a predicted possibility of a third prediction result exceeds a second possibility threshold, transmit the third prediction result to the network device without waiting for the expiration of the first timer.
[0169] In some example embodiments, the terminal device 110 is further caused to: in response to a fourth prediction result indicating a predicted time at which a predicted RLF is to occur is below a time threshold, transmit the fourth prediction result to a network device by using an emergency reporting resource.
[0170] In some example embodiments, the at least one portion of the set of prediction results is transmitted via at least one of: an uplink medium access control (MAC) control element (CE) , or a radio resource control (RRC) message.
[0171] In some example embodiments, the terminal device 110 is caused to: in accordance with a determination that a prediction criterion for a predicted RLF is met, generate a prediction result for the predicted RLF.
[0172] In some example embodiments, the predicted RLF has a cause value of T310 expiry, and the prediction criterion comprises at least one of: that a number of consecutive out-of-sync indications reaches a maximum value lower than N310, that a number of consecutive out-of-sync indications reaches a percentage value of N310, that a number of consecutive out-of-sync indications reaches a value of N310 minus an offset, that a number of consecutive in-sync indications reaches a maximum value lower than N311, that a number of consecutive in-sync indications reaches a percentage value of N311, that a number of consecutive in-sync indications reaches a value of N311 minus an offset, that a timer which is started upon the terminal device receiving N311 out-of-sync indications from a lower layer expires while T310 does not expire, that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T310, or that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T310 minus an offset.
[0173] In some example embodiments, the predicted RLF has a cause value of T312 expiry, and the prediction criterion comprises at least one of: that a timer which is started upon the terminal device receiving N311 out-of-sync indications from a lower layer expires while T312 does not expire, that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T312, or that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T312 minus an offset.
[0174] In some example embodiments, the predicted RLF has a cause value of beam failure recovery failure, and the prediction criterion comprises at least one of: that a number of consecutive beam failure indications from a lower layer reaches a maximum value lower than a beam failure instance max count, that a number of consecutive be am failure indications from a lower layer reaches a percentage value of the beam failure instance max count, or that a number of consecutive beam failure indications from a lower layer reaches a value of the beam failure instance max count minus an offset.
[0175] In some example embodiments, the predicted RLF has a cause value of a radio link control (RLC) retransmission, and the prediction criterion comprises at least one of: that a number of RLC retransmissions reaches a maximum retransmission threshold, that a number of RLC retransmissions reaches a percentage value of the maximum retransmission threshold, or that a number of RLC retransmissions reaches a value of the maximum retransmission threshold minus an offset.
[0176] In some example embodiments, the terminal device 110 is further caused to: receive, from a network device, state information about activation or deactivation of the set of AI / ML models for radio link failure prediction.
[0177] In some example embodiments, the state information comprises at least one of: identifications of the set of AI / ML models, one or more cause values of RLF for prediction, an activation duration timer, a number of AI / ML models concurrently activated for a cause value of RLF, or an indication to activate or deactivate the set of AI / ML models for radio link failure prediction.
[0178] In some example embodiments, the state information indicates a periodical configuration for activating or deactivating the set of AI / ML models for radio link failure prediction, and the periodical configuration comprises at least one of: a starting timing to activate the set of AI / ML models, an activation duration of the set of AI / ML models, a deactivation duration of the set of AI / ML models, or a configuration validation timer.
[0179] In some example embodiments, the terminal device 110 is further caused to: obtain at least one data set of a first data set or a second data set for training the set of AI / ML models, wherein an entry in the first data set is associated with a historical radio link failure at the terminal device, and an entry in the second data set is associated with a near failure state where a radio link failure is approaching to occur at the terminal device.
[0180] In some example embodiments, the entry in the first data set comprises at least one of: a cause value of the historical radio link failure, a time at which the historical radio link failure is declared, respective identifications of one or more cells, an indication of a frequency at which the terminal device operates, an active bandwidth part (BWP) numerology, a configuration for uplink and downlink communications in time domain, a geographical position of the terminal device, a moving speed of the terminal device, a moving direction of the terminal device, or a measurement result of a radio link management reference signal.
[0181] In some example embodiments, the terminal device 110 is further caused to: receive, from a network device, a configuration of a first data collection timer; start to collect entries in the first data set upon starting of the first data collection timer; and terminate collecting entries in the first data set upon expiration of the first data collection timer.
[0182] In some example embodiments, the terminal device 110 is further caused to: in accordance with a determination that a number of entries in the first data set is below a minimum entry threshold, transmit, to a network device, an indication of data insufficiency of the first data set.
[0183] In some example embodiments, the terminal device 110 is further caused to: in accordance with a determination that a number of entries in the first data set exceeds a maximum entry threshold, transmit, to a network device, an indication of a large data set to request a user plane transmission with a specific logical channel identification, or determine a subset of the first data set such that a number of entries in the subset is equal to the maximum entry threshold.
[0184] In some example embodiments, the entry in the second data set comprises at least one of: a cause value of the RLF corresponding to the near failure state, a time at which the near failure state is declared, respective identifications of one or more cells, an indication of a frequency at which the terminal device operates, an active bandwidth part (BWP) numerology, a configuration for uplink and downlink communications in time domain, a geographical position of the terminal device, a moving speed of the terminal device, a moving direction of the terminal device, a measurement result of a radio link management reference signal, or an indication on whether the corresponding RLF occurs after the near failure state is declared.
[0185] In some example embodiments, the terminal device 110 is further caused to: receive, from a network device, a configuration of a second data collection timer; start to collect entries in the second data set upon starting of the second data collection timer; and terminate collecting entries in the second data set upon expiration of the second data collection timer.
[0186] In some example embodiments, the terminal device 110 is further caused to: in accordance with a determination that a number of entries in the second data set is below a minimum entry threshold, transmit, to a network device, an indication of data insufficiency of the second data set.
[0187] In some example embodiments, the terminal device 110 is further caused to: in accordance with a determination that a number of entries in the second data set exceeds a maximum entry threshold, transmit, to a network device, an indication of a large data set to request a user plane transmission with a specific logical channel identification, or determine a subset of the second data set such that a number of entries in the subset is equal to the maximum entry threshold.
[0188] In some example embodiments, the terminal device 110 is further caused to: start to collect entries in the at least one data set in response to receiving, from a network device, an indication to activate training data collection; and terminate collecting entries in the at least one data set in response to receiving, from the network device, an indication to activate training data reporting.
[0189] In some example embodiments, the terminal device 110 is further caused to: transmit, to a network device, the at least one data set via at least one of: a signaling radio bearer (SRB) defined for training data collection, a user plane transmission with a specific data radio bearer (DRB) , a radio bearer specific to AI / ML, or a logical channel specific to AI / ML.
[0190] In some example embodiments, the terminal device 110 is further caused to in accordance with a determination that a near failure criterion for a cause value is met, declare a near failure state where a RLF with the cause value is approaching to occur at the terminal device.
[0191] In some example embodiments, the cause value is T310 expiry, and the near failure criterion comprises at least one of: that a number of consecutive out-of-sync indications reaches a maximum value lower than N310, that a number of consecutive out-of-sync indications reaches a percentage value of N310, that a number of consecutive out-of-sync indications reaches a value of N310 minus an offset, that a number of consecutive in-sync indications reaches a maximum value lower than N311, that a number of consecutive in-sync indications reaches a percentage value of N311, that a number of consecutive in-sync indications reaches a value of N311 minus an offset, that a timer which is started upon the terminal device receiving an out-sync indications from a lower layer expires while T310 does not expire, that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T310, or that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T310 minus an offset.
[0192] In some example embodiments, the cause value is T312 expiry, and the near failure criterion comprises at least one of: that a timer which is started upon the terminal device receivingN311 out-of-sync indications from a lower layer expires while T312 does not expire, that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T312, or that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T312 minus an offset.
[0193] In some example embodiments, the cause value is beam failure recovery failure, and the near failure criterion comprises at least one of: that a number of consecutive beam failure indications from a lower layer reaches a maximum value lower than a beam failure instance max count, that a number of consecutive beam failure indications from a lower layer reaches a percentage value of the beam failure instance max count, or that a number of consecutive beam failure indications from a lower layer reaches a value of the beam failure instance max count minus an offset.
[0194] In some example embodiments, the cause value is RLC retransmission, and the near failure criterion comprises at least one of: that a number of RLC retransmissions reaches a maximum retransmission threshold, that a number of RLC retransmissions reaches a percentage value of the maximum retransmission threshold, or that a number of RLC retransmissions reaches a value of the maximum retransmission threshold minus an offset.
[0195] In some example embodiments, the terminal device 110 is further caused to: transmit, to a network device, capability information related to AI / ML based prediction for radio link failure, wherein the capability information comprises at least one of: correspondence between one or more cause values of RLF and one or more AI / ML model s, a maximum duration for which a radio link failure for a cause value can be predicted, a maximum number of AI / ML models stored for a same cause value, a maximum number of AI / ML models stored for different cause values, a maximum number of AI / ML models for which life cycle management (LCM) can be performed for a same cause value, a maximum number of AI / ML models for which LCM can be performed for different cause values, one or more cause values of RLF to which an AI / ML model can be applied and LCM can be performed, a capability of the terminal device to perform AI / ML based RLF prediction and traditional RLF for the same cause value at the same time or within a short duration, a capability of the terminal device to perform AI / ML based RLF prediction and traditional RLF for different cause values at the same time or within a short duration, a capability of the terminal device to continue model inference after a predicted RLF is determined for each cause value, a capability of the terminal device to trigger multiple RLF prediction for multiple cause values if the different cause values of RLF inferenced by different AI / ML models.
[0196] In some example embodiments, the terminal device 110 is caused to: transmit the capability information to the network device in response to at least one of: a request for the capability information from the network device, activation of AI / ML based prediction for RLF, or an update of capability of the terminal device in AI / ML based prediction for RLF.
[0197] FIG. 5 illustrates a flowchart of a communication method 500 implemented at a network device in accordance with some embodiments of the present disclosure. For the purpose of discussion, the method 500 will be described from the perspective of the network device 120 in FIG. 1.
[0198] At block 510, the network device 120 receives, from a terminal device, at least one portion of a set of prediction results for at least one predicted radio link failure. The set of predication results are generated based on a radio link information associated with a radio link of the terminal device by using a set of AI / ML models. A prediction result comprises information about a predicted radio link failure.
[0199] In some example embodiments, the prediction result in the set of prediction results comprises at least one of: a cause value of the predicted RLF, an indication of a predicted time at which the predicted RLF is to occur, or a predicted possibility that the predicted RLF is to occur at the predicted time.
[0200] In some example embodiments, the network device 120 is further caused to: receive, from a network device, state information about activation or deactivation of the set of AI / ML models for radio link failure prediction.
[0201] In some example embodiments, the state information comprises at least one of: identifications of the set of AI / ML models, one or more cause values of RLF for prediction, an activation duration timer, a number of AI / ML models concurrently activated for a cause value of RLF, or an indication to activate or deactivate the set of AI / ML models for radio link failure prediction.
[0202] In some example embodiments, the state information indicates a periodical configuration for activating or deactivating the set of AI / ML models for radio link failure prediction, and the periodical configuration comprises at least one of: a starting timing to activate the set of AI / ML models, an activation duration of the set of AI / ML models, a deactivation duration of the set of AI / ML models, or a configuration validation timer.
[0203] In some example embodiments, the network device 120 is further caused to: receive, from the terminal device, at least one data set of a first data set or a second data set for training the set of AI / ML models, wherein an entry in the first data set is associated with a historical radio link failure at the terminal device, and an entry in the second data set is associated with a near failure state where a radio link failure is approaching to occur at the terminal device.
[0204] In some example embodiments, the entry in the first data set comprises at least one of: a cause value of the historical radio link failure, a time at which the historical radio link failure is declared, respective identifications of one or more cells, an indication of a frequency at which the terminal device operates, an active bandwidth part (BWP) numerology, a configuration for uplink and downlink communications in time domain, a geographical position of the terminal device, a moving speed of the terminal device, a moving direction of the terminal device, or a measurement result of a radio link management reference signal.
[0205] In some example embodiments, the entry in the second data set comprises at least one of: a cause value of the RLF corresponding to the near failure state, a time at which the near failure state is declared, respective identifications of one or more cells, an indication of a frequency at which the terminal device operates, an active bandwidth part (BWP) numerology, a configuration for uplink and downlink communications in time domain, a geographical position of the terminal device, a moving speed of the terminal device, a moving direction of the terminal device, a measurement result of a radio link management reference signal, or an indication on whether the corresponding RLF occurs after the near failure state is declared.
[0206] In some example embodiments, the network device 120 is further caused to: receive, from the terminal device, capability information related to AI / ML based prediction for radio link failure, wherein the capability information comprises at least one of:correspondence between one or more cause values of RLF and one or more AI / ML models, a maximum duration for which a radio link failure for a cause value can be predicted, a maximum number of AI / ML models stored for a same cause value, a maximum number of AI / ML models stored for different cause values, a maximum number of AI / ML models for which life cycle management (LCM) can be performed for a same cause value, a maximum number of AI / ML models for which LCM can be performed for different cause values, one or more cause values of RLF to which an AI / ML model can be applied and LCM can be performed, a capability of the terminal device to perform AI / ML based RLF prediction and traditional RLF for the same cause value at the same time or within a short duration, a capability of the terminal device to perform AI / ML based RLF prediction and traditional RLF for different cause values at the same time or within a short duration, a capability of the terminal device to continue model inference after a predicted RLF is determined for each cause value, a capability of the terminal device to trigger multiple RLF prediction for multiple cause values if the different cause values of RLF inferenced by different AI / ML models.
[0207] FIG. 6 is a simplified block diagram of a device 600 that is suitable for implementing embodiments of the present disclosure. The device 600 can be considered as a further example implementation of any of the devices as shown in FIG. 1. Accordingly, the device 600 can be implemented at or as at least a part of the terminal device 110 or the network device 120.
[0208] As shown, the device 600 includes a processor 610, a memory 620 coupled to the processor 610, a suitable transceiver 640 coupled to the processor 610, and a communication interface coupled to the transceiver 640. The memory 620 stores at least a part of a program 630. The transceiver 640 may be for bidirectional communications or a unidirectional communication based on requirements. The transceiver 640 may include at least one of a transmitter 642 and a receiver 644. The transmitter 642 and the receiver 644 may be functional modules or physical entities. The transceiver 640 has at least one antenna to facilitate communication, though in practice an Access Node mentioned in this application may have several ones. The communication interface may represent any interface that is necessary for communication with other network elements, such as X2 / Xn interface for bidirectional communications between eNBs / gNBs, S1 / NG interface for communication between a Mobility Management Entity (MME) / Access and Mobility Management Function (AMF) / SGW / UPF and the eNB / gNB, Un interface for communication between the eNB / gNB and a relay node (RN) , or Uu interface for communication between the eNB / gNB and a terminal device.
[0209] The program 630 is assumed to include program instructions that, when executed by the associated processor 610, enable the device 600 to operate in accordance with the embodiments of the present disclosure, as discussed herein with reference to FIGS. 1 to 5. The embodiments herein may be implemented by computer software executable by the processor 610 of the device 600, or by hardware, or by a combination of software and hardware. The processor 610 may be configured to implement various embodiments of the present disclosure. Furthermore, a combination of the processor 610 and memory 620 may form processing means 650 adapted to implement various embodiments of the present disclosure.
[0210] The memory 620 may be of any type suitable to the local technical network and may be implemented using any suitable data storage technology, such as a non-transitory computer readable storage medium, semiconductor-based memory devices, magnetic memory devices and systems, optical memory devices and systems, fixed memory and removable memory, as non-limiting examples. While only one memory 620 is shown in the device 600, there may be several physically distinct memory modules in the device 600. The processor 610 may be of any type suitable to the local technical network, and may include one or more of general purpose computers, special purpose computers, microprocessors, digital signal processors (DSPs) and processors based on multicore processor architecture, as non-limiting examples. The device 600 may have multiple processors, such as an application specific integrated circuit chip that is slaved in time to a clock which synchronizes the main processor.
[0211] According to embodiments of the present disclosure, a terminal device comprising a circuitry is provided. The circuitry is configured to: obtain radio link information associated with a radio link of the terminal device; and generate a set of prediction results for at least one predicted radio link failure based on the radio link information by using a set of AI / ML models. A prediction result comprises information about a predicted radio link failure. According to embodiments of the present disclosure, the circuitry may be configured to perform any method implemented by the terminal device as discussed above.
[0212] According to embodiments of the present disclosure, a network device comprising a circuitry is provided. The circuitry is configured to: receive, from a terminal device, at least one portion of a set of prediction results for at least one predicted radio link failure, wherein the set of predication results are generated based on a radio link information associated with a radio link of the terminal device by using a set of AI / ML models. A prediction result comprises information about a predicted radio link failure. According to embodiments of the present disclosure, the circuitry may be configured to perform any method implemented by the network device as discussed above.
[0213] The term “circuitry” used herein may refer to hardware circuits and / or combinations of hardware circuits and software. For example, the circuitry may be a combination of analog and / or digital hardware circuits with software / firmware. As a further example, the circuitry may be any portions of hardware processors with software including digital signal processor (s) , software, and memory (ies) that work together to cause an apparatus, such as a terminal device or a network device, to perform various functions. In a still further example, the circuitry may be hardware circuits and or processors, such as a microprocessor or a portion of a microprocessor, that requires software / firmware for operation, but the software may not be present when it is not needed for operation. As used herein, the term circuitry also covers an implementation of merely a hardware circuit or processor (s) or a portion of a hardware circuit or processor (s) and its (or their) accompanying software and / or firmware.
[0214] According to embodiments of the present disclosure, a terminal device is provided. The terminal apparatus comprises means for obtaining radio link information associated with a radio link of the terminal device; and means for generateing a set of prediction results for at least one predicted radio link failure based on the radio link information by using a set of AI / ML models. A prediction result comprises information about a predicted radio link failure. In some embodiments, the terminal device may comprise means for performing the respective operations of the method 400. In some example embodiments, the terminal device may further comprise means for performing other operations in some example embodiments of the method 400. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module.
[0215] According to embodiments of the present disclosure, a network device is provided. The network device comprises means for receiving, from a terminal device, at least one portion of a set of prediction results for at least one predicted radio link failure. The set of predication results are generated based on a radio link information associated with a radio link of the terminal device by using a set of AI / ML models. A prediction result comprises information about a predicted radio link failure. In some embodiments, the network device may comprise means for performing the respective operations of the method 400. In some example embodiments, the network device may further comprise means for performing other operations in some example embodiments of the method 400. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module.
[0216] In summary, embodiments of the present disclosure provide the following aspects.
[0217] In an aspect, it is proposed a terminal device comprising: a processor configured to cause the terminal device to: obtain radio link information associated with a radio link of the terminal device; and generate a set of prediction results for at least one predicted radio link failure based on the radio link information by using a set of AI / ML models. A prediction result comprises information about a predicted radio link failure.
[0218] In some embodiments, the prediction result in the set of prediction results comprises at least one of: a cause value of the predicted RLF, an indication of a predicted time at which the predicted RLF is to occur, or a predicted possibility that the predicted RLF is to occur at the predicted time.
[0219] In some embodiments, the radio link information comprises at least one of: a number of indications associated with radio link failures counted at a predetermined layer, respective identifications of one or more cells, a measurement result of a radio link management reference signal, a geographical position of the terminal device, a moving speed of the terminal device, a moving direction of the terminal device, an indication of a frequency at which the terminal device operates, an active bandwidth part (BWP) numerology, or a configuration for uplink and downlink communications in time domain.
[0220] In some embodiments, the terminal device is further caused to: in response to a first prediction result in the set of prediction results being generated, start a first timer; and in response to expiration of the first timer, transmit at least one portion of the set of prediction results to a network device.
[0221] In some embodiments, a second prediction result in the set of prediction results comprises a predicted possibility below a first possibility threshold, the predicted possibility is a prediction that a predicted RLF is to occur at a predicted time, and the second prediction result is not transmitted to the network device.
[0222] In some embodiments, the terminal device is further caused to: in response to that a predicted possibility of a third prediction result exceeds a second possibility threshold, transmit the third prediction result to the network device without waiting for the expiration of the first timer.
[0223] In some embodiments, the terminal device is further caused to: in response to a fourth prediction result indicating a predicted time at which a predicted RLF is to occur is below a time threshold, transmit the fourth prediction result to a network device by using an emergency reporting resource.
[0224] In some embodiments, the at least one portion of the set of prediction results is transmitted via at least one of: an uplink medium access control (MAC) control element (CE) , or a radio resource control (RRC) message.
[0225] In some embodiments, the terminal device is caused to: in accordance with a determination that a prediction criterion for a predicted RLF is met, generate a prediction result for the predicted RLF.
[0226] In some embodiments, the predicted RLF has a cause value of T310 expiry, and the prediction criterion comprises at least one of: that a number of consecutive out-of-sync indications reaches a maximum value lower than N310, that a number of consecutive out-of-sync indications reaches a percentage value of N310, that a number of consecutive out-of-sync indications reaches a value of N310 minus an offset, that a number o f consecutive in-sync indications reaches a maximum value lower than N311, that a number of consecutive in-sync indications reaches a percentage value of N311, that a number of consecutive in-sync indications reaches a value of N311 minus an offset, that a timer which is started upon the terminal device receiving N311 out-of-sync indications from a lower layer expires while T310 does not expire, that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T310, or that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T310 minus an offset.
[0227] In some embodiments, the predicted RLF has a cause value of T312 expiry, and the prediction criterion comprises at least one of: that a timer which is started upon the terminal device receiving N311 out-of-sync indications from a lower layer expires while T312 does not expire, that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T312, or that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T312 minus an offset.
[0228] In some embodiments, the predicted RLF has a cause value of beam failure recovery failure, and the prediction criterion comprises at least one of: that a number of consecutive beam failure indications from a lower layer reaches a maximum value lower than a beam failure instance max count, that a number of consecutive beam failure indications from a lower layer reaches a percentage value of the beam failure instance max count, or that a number of consecutive beam failure indications from a lower layer reaches a value of the beam failure instance max count minus an offset.
[0229] In some embodiments, the predicted RLF has a cause value of a radio link control (RLC) retransmission, and the prediction criterion comprises at least one of: that a number of RLC retransmissions reaches a maximum retransmission threshold, that a number of RLC retransmissions reaches a percentage value of the maximum retransmission threshold, or that a number of RLC retransmissions reaches a value of the maximum retransmission threshold minus an offset.
[0230] In some embodiments, the terminal device is further caused to: receive, from a network device, state information about activation or deactivation of the set of AI / ML models for radio link failure prediction.
[0231] In some embodiments, the state information comprises at least one of: identifications of the set of AI / ML models, one or more cause values of RLF for prediction, an activation duration timer, a number of AI / ML models concurrently activated for a cause value of RLF, or an indication to activate or deactivate the set of AI / ML models for radio link failure prediction.
[0232] In some embodiments, the state information indicates a periodical configuration for activating or deactivating the set of AI / ML models for radio link failure prediction, and the periodical configuration comprises at least one of: a starting timing to activate the set of AI / ML models, an activation duration of the set of AI / ML models, a deactivation duration of the set of AI / ML models, or a configuration validation timer.
[0233] In some embodiments, the terminal device is further caused to: obtain at least one data set of a first data set or a second data set for training the set of AI / ML models, wherein an entry in the first data set is associated with a historical radio link failure at the terminal device, and an entry in the second data set is associated with a near failure state where a radio link failure is approaching to occur at the terminal device.
[0234] In some embodiments, the entry in the first data set comprises at least one of: a cause value of the historical radio link failure, a time at which the historical radio link failure is declared, respective identifications of one or more cells, an indication of a frequency at which the terminal device operates, an active bandwidth part (BWP) numerology, a configuration for uplink and downlink communications in time domain, a geographical position of the terminal device, a moving speed of the terminal device, a moving direction of the terminal device, or a measurement result of a radio link management reference signal.
[0235] In some embodiments, the terminal device is further caused to: receive, from a network device, a configuration of a first data collection timer; start to collect entries in the first data set upon starting of the first data collection timer; and terminate collecting entries in the first data set upon expiration of the first data collection timer.
[0236] In some embodiments, the terminal device is further caused to: in accordance with a determination that a number of entries in the first data set is below a minimum entry threshold, transmit, to a network device, an indication of data insufficiency of the first data set.
[0237] In some embodiments, the terminal device is further caused to: in accordance with a determination that a number of entries in the first data set exceeds a maximum entry threshold, transmit, to a network device, an indication of a large data set to request a user plane transmission with a specific logical channel identification, or determine a subset of the first data set such that a number of entries in the subset i s equal to the maximum entry threshold.
[0238] In some embodiments, the entry in the second data set comprises at least one of: a cause value of the RLF corresponding to the near failure state, a time at which the near failure state is declared, respective identifications of one or more cells, an indication of a frequency at which the terminal device operates, an active bandwidth part (BWP) numerology, a configuration for uplink and downlink communications in time domain, a geographical position of the terminal device, a moving speed of the terminal device, a moving direction of the terminal device, a measurement result of a radio link management reference signal, or an indication on whether the corresponding RLF occurs after the near failure state is declared.
[0239] In some embodiments, the terminal device is further caused to: receive, from a network device, a configuration of a second data collection timer; start to collect entries in the second data set upon starting of the second data collection timer; and terminate collecting entries in the second data set upon expiration of the second data collection timer.
[0240] In some embodiments, the terminal device is further caused to: in accordance with a determination that a number of entries in the second data set is below a minimum entry threshold, transmit, to a network device, an indication of data insufficiency of the second data set.
[0241] In some embodiments, the terminal device is further caused to: in accordance with a determination that a number of entries in the second data set exceeds a maximum entry threshold, transmit, to a network device, an indication of a large data set to request a user plane transmission with a specific logical channel identification, or determine a subset of the second data set such that a number of entries in the subset is equal to the maximum entry threshold.
[0242] In some embodiments, the terminal device is further caused to: start to collect entries in the at least one data set in response to receiving, from a network device, an indication to activate training data collection; and terminate collecting entries in the at least one data set in response to receiving, from the network device, an indication to activate training data reporting.
[0243] In some embodiments, the terminal device is further caused to: transmit, to a network device, the at least one data set via at least one of: a signaling radio bearer (SRB) defined for training data collection, a user plane transmission with a specific data radio bearer (DRB) , a radio bearer specific to AI / ML, or a logical channel specific to AI / ML.
[0244] In some embodiments, the terminal device is further caused to in accordance with a determination that a near failure criterion for a cause value is met, declare a near failure state where a RLF with the cause value is approaching to occur at the terminal device.
[0245] In some embodiments, the cause value is T310 expiry, and the near failure criterion comprises at least one of: that a number of consecutive out-of-sync indications reaches a maximum value lower than N310, that a number of consecutive out-of-sync indications reaches a percentage value of N310, that a number of consecutive out-of-sync indications reaches a value of N310 minus an offset, that a number of consecutive in-sync indications reaches a maximum value lower than N311, that a number of consecutive in-sync indications reaches a percentage value of N311, that a number of consecutive in-sync indications reaches a value of N311 minus an offset, that a timer which is started upon the terminal device receiving an out-sync indications from a lower layer expires while T310 does not expire, that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T310, or that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T310 minus an offset.
[0246] In some embodiments, the cause value is T312 expiry, and the near failure criterion comprises at least one of: that a timer which is started upon the terminal device receiving N311 out-of-sync indications from a lower layer expires while T312 does not expire, that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T312, or that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T312 minus an offset.
[0247] In some embodiments, the cause value is beam failure recovery failure, and the near failure criterion comprises at least one of: that a number of consecutive beam failure indications from a lower layer reaches a maximum value lower than a beam failure instance max count, that a number of consecutive beam failure indications from a lower layer reaches a percentage value of the beam failure instance max count, or that a number of consecutive beam failure indications from a lower layer reaches a value of the beam failure instance max count minus an offset.
[0248] In some embodiments, the cause value is RLC retransmission, and the near failure criterion comprises at least one of: that a number of RLC retransmissions reaches a maximum retransmission threshold, that a number of RLC retransmissions reaches a percentage value of the maximum retransmission threshold, or that a number of RLC retransmissions reaches a value of the maximum retransmission threshold minus an offset.
[0249] In some embodiments, the terminal device is further caused to: transmit, to a network device, capability information related to AI / ML based prediction for radio link failure, wherein the capability information comprises at least one of: correspondence between one or more cause values of RLF and one or more AI / ML models, a maximum duration for which a radio link failure for a cause value can be predicted, a maximum number of AI / ML models stored for a same cause value, a maximum number of AI / ML models stored for different cause values, a maximum number of AI / ML models for which life cycle management (LCM) can be performed for a same cause value, a maximum number of AI / ML models for which LCM can be performed for different cause values, one or more cause values of RLF to which an AI / ML model can be applied and LCM can be performed, a capability of the terminal device to perform AI / ML based RLF prediction and traditional RLF for the same cause value at the same time or within a short duration, a capability of the terminal device to perform AI / ML based RLF prediction and traditional RLF for different cause values at the same time or within a short duration, a capability of the terminal device to continue model inference after a predicted RLF is determined for each cause value, a capability of the terminal device to trigger multiple RLF prediction for multiple cause values if the different cause values of RLF inferenced by different AI / ML models.
[0250] In some embodiments, the terminal device is caused to: transmit the capability information to the network device in response to at least one of: a request for the capability information from the network device, activation of AI / ML based prediction for RLF, or an update of capability of the terminal device in AI / ML based prediction for RLF.
[0251] In an aspect, it is proposed a network device comprising: a processor configured to cause the network device to: receive, from a terminal device, at least one portion of a set of prediction results for at least one predicted radio link failure. The set of predication results are generated based on a radio link information associated with a radio link of the terminal device by using a set of AI / ML models. A prediction result comprises information about a predicted radio link failure.
[0252] In some embodiments, the prediction result in the set of prediction results comprises at least one of: a cause value of the predicted RLF, an indication of a predicted time at which the predicted RLF is to occur, or a predicted possibility that the predicted RLF is to occur at the predicted time.
[0253] In some embodiments, the network device is further caused to: receive, from a network device, state information about activation or deactivation of the set of AI / ML models for radio link failure prediction.
[0254] In some embodiments, the state information comprises at least one of: identifications of the set of AI / ML models, one or more cause values of RLF for prediction, an activation duration timer, a number of AI / ML models concurrently activated for a cause value of RLF, or an indication to activate or deactivate the set of AI / ML models for radio link failure prediction.
[0255] In some embodiments, the state information indicates a periodical configuration for activating or deactivating the set of AI / ML models for radio link failure prediction, and the periodical configuration comprises at least one of: a starting timing to activate the set of AI / ML models, an activation duration of the set of AI / ML models, a deactivation duration of the set of AI / ML models, or a configuration validation timer.
[0256] In some embodiments, the network device is further caused to: receive, from the terminal device, at least one data set of a first data set or a second data set for training the set of AI / ML models, wherein an entry in the first data set is associated with a historical radio link failure at the terminal device, and an entry in the second data set is associated with a near failure state where a radio link failure is approaching to occur at the terminal device.
[0257] In some embodiments, the entry in the first data set comprises at least one of: a cause value of the historical radio link failure, a time at which the historical radio link failure is declared, respective identifications of one or more cells, an indication of a frequency at which the terminal device operates, an active bandwidth part (BWP) numerology, a configuration for uplink and downlink communications in time domain, a geographical position of the terminal device, a moving speed of the terminal device, a moving direction of the terminal device, or a measurement result of a radio link management reference signal.
[0258] In some embodiments, the entry in the second data set comprises at least one of: a cause value of the RLF corresponding to the near failure state, a time at which the near failure state is declared, respective identifications of one or more cells, an indication of a frequency at which the terminal device operates, an active bandwidth part (BWP) numerology, a configuration for uplink and downlink communications in time domain, a geographical position of the terminal device, a moving speed of the terminal device, a moving direction of the terminal device, a measurement result of a radio link management reference signal, or an indication on whether the corresponding RLF occurs after the near failure state is declared.
[0259] In some embodiments, the network device is further caused to: receive, from the terminal device, capability information related to AI / ML based prediction for radio link failure, wherein the capability information comprises at least one of: correspondence between one or more cause values of RLF and one or more AI / ML models, a maximum duration for which a radio link failure for a cause value can be predicted, a maximum number of AI / ML models stored for a same cause value, a maximum number of AI / ML models stored for different cause values, a maximum number of AI / ML models for which life cycle management (LCM) can be performed for a same cause value, a maximum number of AI / ML models for which LCM can be performed for different cause values, one or more cause values of RLF to which an AI / ML model can be applied and LCM can be performed, a capability of the terminal device to perform AI / ML based RLF prediction and traditional RLF for the same cause value at the same time or within a short duration, a capability of the terminal device to perform AI / ML based RLF prediction and traditional RLF for different cause values at the same time or within a short duration, a capability of the terminal device to continue model inference after a predicted RLF is determined for each cause value, a capability of the terminal device to trigger multiple RLF prediction for multiple cause values if the different cause values of RLF inferenced by different AI / ML models.
[0260] In an aspect, a terminal device comprises: at least one processor; and at least one memory coupled to the at least one processor and storing instructions thereon, the instructions, when executed by the at least one processor, causing the device to perform the method implemented by the terminal device discussed above.
[0261] In an aspect, a network device comprises: at least one processor; and at least one memory coupled to the at least one processor and storing instructions thereon, the instructions, when executed by the at least one processor, causing the device to perform the method implemented by the network device discussed above.
[0262] In an aspect, a computer readable medium having instructions stored thereon, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the terminal device discussed above.
[0263] In an aspect, a computer readable medium having instructions stored thereon, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the network device discussed above.
[0264] In an aspect, a computer program comprising instructions, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the terminal device discussed above.
[0265] In an aspect, a computer program comprising instructions, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the network device discussed above.
[0266] Generally, various embodiments of the present disclosure may be implemented in hardware or special purpose circuits, software, logic or any combination thereof. Some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software which may be executed by a controller, microprocessor or other computing device. While various aspects of embodiments of the present disclosure are illustrated and described as block diagrams, flowcharts, or using some other pictorial representation, it will be appreciated that the blocks, apparatus, systems, techniques or methods described herein may be implemented in, as non-limiting examples, hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof.
[0267] The present disclosure also provides at least one computer program product tangibly stored on a non-transitory computer readable storage medium. The computer program product includes computer-executable instructions, such as those included in program modules, being executed in a device on a target real or virtual processor, to carry out the process or method as described above with reference to FIGS. 1 to 6. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, or the like that perform particular tasks or implement particular abstract data types. The functionality of the program modules may be combined or split between program modules as desired in various embodiments. Machine-executable instructions for program modules may be executed within a local or distributed device. In a distributed device, program modules may be located in both local and remote storage media.
[0268] Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may execute entirely on a machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
[0269] The above program code may be embodied on a machine readable medium, which may be any tangible medium that may contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine readable medium may be a machine readable signal medium or a machine readable storage medium. A machine readable medium may include but not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine readable storage medium would include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM) , a read-only memory (ROM) , an erasable programmable read-only memory (EPROM or Flash memory) , an optical fiber, a portable compact disc read-only memory (CD-ROM) , an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0270] Further, while operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve des irable results. In certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are contained in the above discussions, these should not be construed as limitations on the scope of the present disclosure, but rather as descriptions of features that may be specific to particular embodiments. Certain features that are described in the context of separate embodiments may also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment may also be implemented in multiple embodiments separately or in any suitable sub-combination.
[0271] Although the present disclosure has been described in language specific to structural features and / or methodological acts, it is to be understood that the present disclosure defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.
Claims
1.A terminal device comprising:a processor configured to cause the terminal device to:obtain radio link information associated with a radio link of the terminal device; andgenerate a set of prediction results for at least one predicted radio link failure based on the radio link information by using a set of Artificial Intelligence / Machine learning (AI / ML) models, a prediction result comprising information about a predicted radio link failure.2.The terminal device of claim 1, wherein the prediction result in the set of prediction results comprises at least one of:a cause value of the predicted RLF,an indication of a predicted time at which the predicted RLF is to occur, ora predicted possibility that the predicted RLF is to occur at the predicted time.3.The terminal device of claim 1, wherein the radio link information comprises at least one of:a number of indications associated with radio link failures counted at a predetermined layer,respective identifications of one or more cells,a measurement result of a radio link management reference signal,a geographical position of the terminal device,a moving speed of the terminal device,a moving direction of the terminal device,an indication of a frequency at which the terminal device operates,an active bandwidth part (BWP) numerology, ora configuration for uplink and downlink communications in time domain.4.The terminal device of claim 1, wherein the terminal device is further caused to:in response to a first prediction result in the set of prediction results being generated, start a first timer; andin response to expiration of the first timer, transmit at least one portion of the set of prediction results to a network device.5.The terminal device of claim 1, wherein the terminal device is further caused to:in response to a fourth prediction result indicating a predicted time at which a predicted RLF is to occur is below a time threshold, transmit the fourth prediction result to a network device by using an emergency reporting resource.6.The terminal device of claim 4, wherein the at least one portion of the set of prediction results is transmitted via at least one of:an uplink medium access control (MAC) control element (CE) , ora radio resource control (RRC) message.7.The terminal device of claim 1, wherein a prediction result for a predicted RLF is generated in accordance with a determination that a prediction criterion for the predicted RLF is met, the predicted RLF has a cause value of T310 expiry, and the prediction criterion comprises at least one of:that a number of consecutive out-of-sync indications reaches a maximum value lower than N310,that a number of consecutive out-of-sync indications reaches a percentage value of N310,that a number of consecutive out-of-sync indications reaches a value of N310 minus an offset,that a number of consecutive in-sync indications reaches a maximum value lower than N311,that a number of consecutive in-sync indications reaches a percentage value of N311,that a number of consecutive in-sync indications reaches a value of N311 minus an offset,that a timer which is started upon the terminal device receiving N311 out-of-sync indications from a lower layer expires while T310 does not expire,that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T310, orthat a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T310 minus an offset.8.The terminal device of claim 1, wherein a prediction result for a predicted RLF is generated in accordance with a determination that a prediction criterion for the predicted RLF is met, the predicted RLF has a cause value of T312 expiry, and the prediction criterion comprises at least one of:that a timer which is started upon the terminal device receiving N311 out-of-sync indications from a lower layer expires while T312 does not expire,that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T312, orthat a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T312 minus an offset.9.The terminal device of claim 1, wherein a prediction result for a predicted RLF is generated in accordance with a determination that a prediction criterion for the predicted RLF is met, the predicted RLF has a cause value of beam failure recovery failure, and the prediction criterion comprises at least one of:that a number of consecutive beam failure indications from a lower layer reaches a maximum value lower than a beam failure instance max count,that a number of consecutive beam failure indications from a lower layer reaches a percentage value of the beam failure instance max count, orthat a number of consecutive beam failure indications from a lower layer reaches a value of the beam failure instance max count minus an offset.10.The terminal device of claim 1, wherein a prediction result for a predicted RLF is generated in accordance with a determination that a prediction criterion for the predicted RLF is met, the predicted RLF has a cause value of a radio link control (RLC) retransmission, and the prediction criterion comprises at least one of:that a number of RLC retransmissions reaches a maximum retransmission threshold,that a number of RLC retransmissions reaches a percentage value of the maximum retransmission threshold, orthat a number of RLC retransmissions reaches a value of the maximum retransmission threshold minus an offset.11.The terminal device of claim 1, wherein state information about activation or deactivation of the set of AI / ML models for radio link failure prediction indicates a periodical configuration for activating or deactivating the set of AI / ML models for radio link failure prediction, and the periodical configuration comprises at least one of:a starting timing to activate the set of AI / ML models,an activation duration of the set of AI / ML models,a deactivation duration of the set of AI / ML models, ora configuration validation timer.12.The terminal device of claim 1, wherein the terminal device is further caused to:obtain at least one data set of a first data set or a second data set for training the set of AI / ML models,wherein an entry in the first data set is associated with a historical radio link failure at the terminal device, andan entry in the second data set is associated with a near failure state where a radio link failure is approaching to occur at the terminal device.13.The terminal device of claim 12, wherein the terminal device is further caused to:transmit, to a network device, the at least one data set via at least one of:a signaling radio bearer (SRB) defined for training data collection,a user plane transmission with a specific data radio bearer (DRB) ,a radio bearer specific to AI / ML, ora logical channel specific to AI / ML.14.The terminal device of claim 12, wherein a near failure state where a RLF with the cause value is approaching to occur at the terminal device is declared in accordance with a determination that a near failure criterion for a cause value is met, the cause value is T310 expiry, and the near failure criterion comprises at least one of:that a number of consecutive out-of-sync indications reaches a maximum value lower than N310,that a number of consecutive out-of-sync indications reaches a percentage value of N310,that a number of consecutive out-of-sync indications reaches a value of N310 minus an offset,that a number of consecutive in-sync indications reaches a maximum value lower than N311,that a number of consecutive in-sync indications reaches a percentage value of N311,that a number of consecutive in-sync indications reaches a value of N311 minus an offset,that a timer which is started upon the terminal device receiving N311 out-of-sync indications from a lower layer expires while T310 does not expire,that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T310, orthat a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T310 minus an offset.15.The terminal device of claim 12, wherein a near failure state where a RLF with the cause value is approaching to occur at the terminal device is declared in accordance with a determination that a near failure criterion for a cause value is met, the cause value is T312 expiry, and the near failure criterion comprises at least one of:that a timer which is started upon the terminal device receiving N311 out-of-sync indications from a lower layer expires while T312 does not expire,that a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of a percentage of T312, orthat a duration for receiving consecutive N311 in-sync indications from a lower layer exceeds a duration of T312 minus an offset.16.The terminal device of claim 12, wherein a near failure state where a RLF with the cause value is approaching to occur at the terminal device is declared in accordance with a determination that a near failure criterion for a cause value is met, the cause value is beam failure recovery failure, and the near failure criterion comprises at least one of:that a number of consecutive beam failure indications from a lower layer reaches a maximum value lower than a beam failure instance max count,that a number of consecutive beam failure indications from a lower layer reaches a percentage value of the beam failure instance max count, orthat a number of consecutive beam failure indications from a lower layer reaches a value of the beam failure instance max count minus an offset.17.A network device comprising:a processor configured to cause the network device to:receive, from a terminal device, at least one portion of a set of prediction results for at least one predicted radio link failure, wherein the set of predication results are generated based on a radio link information associated with a radio link of the terminal device by using a set of Artificial Intelligence / Machine learning (AI / ML) models, a prediction result comprising information about a predicted radio link failure;transmit, to the terminal device, state information about activation or deactivation of the set of AI / ML models for radio link failure prediction;wherein the state information indicates a periodical configuration for activating or deactivating the set of AI / ML models for radio link failure prediction, and the periodical configuration comprises at least one of:a starting timing to activate the set of AI / ML models,an activation duration of the set of AI / ML models,a deactivation duration of the set of AI / ML models, ora configuration validation timer.18.The network device of claim 17, wherein the network device is further caused to:receive, from the terminal device, at least one data set of a first data set or a second data set for training the set of AI / ML models,wherein an entry in the first data set is associated with a historical radio link failure at the terminal device, andan entry in the second data set is associated with a near failure state where a radio link failure is approaching to occur at the terminal device.19.The network device of claim 18, wherein the entry in the first data set comprises at least one of:a cause value of the historical radio link failure,a time at which the historical radio link failure is declared,respective identifications of one or more cells,an indication of a frequency at which the terminal device operates,an active bandwidth part (BWP) numerology,a configuration for uplink and downlink communications in time domain,a geographical position of the terminal device,a moving speed of the terminal device,a moving direction of the terminal device, ora measurement result of a radio link management reference signal.20.The network device of claim 18, wherein the entry in the second data set comprises at least one of:a cause value of the RLF corresponding to the near failure state,a time at which the near failure state is declared,respective identifications of one or more cells,an indication of a frequency at which the terminal device operates,an active bandwidth part (BWP) numerology,a configuration for uplink and downlink communications in time domain,a geographical position of the terminal device,a moving speed of the terminal device,a moving direction of the terminal device,a measurement result of a radio link management reference signal, oran indication on whether the corresponding RLF occurs after the near failure state is declared.
Citation Information
Patent Citations
Information transmission method and device, communication equipment and storage medium
CN115836545A
Using machine learning to monitor link quality and predict link faults
US20190044824A1
Machine learning fallback model for wireless device
WO2023209673A1