Devices and methods of communication
The terminal device improves model management by evaluating model inferences through delayed handover procedures and post-handover evaluations, addressing incomplete model monitoring issues in mobility-related models.
Patent Information
- Application Number
- PCT/CN2024/107770
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-07-26
- Publication Date
- 2026-01-29
AI Technical Summary
Current solutions for model management in mobility-related models are incomplete and require further development to ensure effective model monitoring and management, particularly in scenarios where handover predictions and time of stay predictions are concerned.
A terminal device evaluates model inference based on predetermined events and conditions, such as delayed handover procedures or post-handover evaluations, to enhance model monitoring accuracy and manage model inputs and outputs effectively.
Enhances the accuracy and efficiency of model monitoring by providing additional time for model evaluation and continuous monitoring, ensuring correct predictions and timely adjustments.
Smart Images

Figure CN2024107770_29012026_PF_FP_ABST
Abstract
Description
DEVICES AND METHODS OF COMMUNICATIONTECHNICAL FIELD
[0001] Embodiments of the present disclosure generally relate to the field of telecommunication, and in particular, to devices and methods of communication for model management.BACKGROUND
[0002] Currently, it has been proposed to introduce artificial intelligence (AI) for mobility enhancement. Accordingly, a mobility related model has been developed. To make sure that a model inference works normally, model management such as model monitoring or other model related operations is required. However, solutions of model management for a mobility related model are still incomplete and need to be further developed.SUMMARY
[0003] In general, embodiments of the present disclosure provide methods, devices and computer storage media of communication for model management.
[0004] In a first aspect, there is provided a terminal device. The terminal device comprises a processor. The processor is configured to cause the terminal device to: transmit a predicted result of a model inference to a network device, the predicted result indicating that a handover is to occur at a first time instance; and in accordance with a determination that a command of the handover is received and a first event for a model monitoring does not occur, evaluate the model inference based on a second event for the model monitoring.
[0005] In a second aspect, there is provided a terminal device. The terminal device comprises a processor. The processor is configured to cause the terminal device to: determine, based on a model inference, that a first time of stay in a target cell of a handover is predicted; store the first time of stay; determine a second time of stay in the target cell after the procedure of the handover; and evaluate the model inference based on a comparison between the first time of stay and the second time of stay.
[0006] In a third aspect, there is provided a terminal device. The terminal device comprises a processor. The processor is configured to cause the terminal device to: transmit, to a network device, at least one of the following: first information of a prediction of a model; or second information indicating update of an input requirement of the model.
[0007] In a fourth aspect, there is provided a terminal device. The terminal device comprises a processor. The processor is configured to cause the terminal device to: determine that a first set of outputs of a model is problematic; perform a first operation for handling the first set of outputs; and perform a model prediction with a second set of outputs of the model.
[0008] In a fifth aspect, there is provided a method of communication. The method comprises: transmitting, at a terminal device, a predicted result of a model inference to a network device, the predicted result indicating that a handover is to occur at a first time instance; and in accordance with a determination that a command of the handover is received and a first event for a model monitoring does not occur, evaluating the model inference based on a second event for the model monitoring.
[0009] In a sixth aspect, there is provided a method of communication. The method comprises: determining, at a terminal device based on a model inference, that a first time of stay in a target cell of a handover is predicted; storing the first time of stay; determining a second time of stay in the target cell after the procedure of the handover; and evaluating the model inference based on a comparison between the first time of stay and the second time of stay.
[0010] In a seventh aspect, there is provided a method of communication. The method comprises: transmitting, at a terminal device and to a network device, at least one of the following: first information of a prediction of a model; or second information indicating update of an input requirement of the model.
[0011] In an eighth aspect, there is provided a method of communication. The method comprises: determining, at a terminal device, that a first set of outputs of a model is problematic; performing a first operation for handling the first set of outputs; and performing a model prediction with a second set of outputs of the model.
[0012] In a ninth aspect, there is provided a computer readable medium having instructions stored thereon. The instructions, when executed on at least one processor, cause the at least one processor to perform the method according to any of the fifth to eighth aspects of the present disclosure.
[0013] Other features of the present disclosure will become easily comprehensible through the following description.BRIEF DESCRIPTION OF THE DRAWINGS
[0014] Through the more detailed description of some 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:
[0015] FIG. 1 illustrates an example communication network in which some embodiments of the present disclosure can be implemented;
[0016] FIG. 2 illustrates a schematic diagram of an AI common framework in which some embodiments of the present disclosure can be implemented;
[0017] FIG. 3 illustrates a signaling chart illustrating an example process of communication for model monitoring in accordance with some embodiments of the present disclosure;
[0018] FIG. 4A illustrates a schematic diagram of an example scenario of model monitoring in accordance with some embodiments of the present disclosure;
[0019] FIG. 4B illustrates a schematic diagram of another example scenario of model monitoring in accordance with some embodiments of the present disclosure;
[0020] FIG. 4C illustrates a schematic diagram of another example scenario of model monitoring in accordance with some embodiments of the present disclosure;
[0021] FIG. 4D illustrates a schematic diagram of another example scenario of model monitoring in accordance with some embodiments of the present disclosure;
[0022] FIG. 5 illustrates a signaling chart illustrating another example process of communication for model monitoring in accordance with some embodiments of the present disclosure;
[0023] FIG. 6 illustrates a signaling chart illustrating an example process of communication for model information reporting in accordance with some embodiments of the present disclosure;
[0024] FIG. 7 illustrates a signaling chart illustrating an example process of communication for model output management in accordance with some embodiments of the present disclosure;
[0025] FIG. 8 illustrates a schematic diagram of an example scenario of model output management in accordance with some embodiments of the present disclosure;
[0026] FIG. 9 illustrates a flowchart of an example method of communication implemented at a terminal device in accordance with some embodiments of the present disclosure;
[0027] FIG. 10 illustrates a flowchart of another example method of communication implemented at a terminal device in accordance with some embodiments of the present disclosure;
[0028] FIG. 11 illustrates a flowchart of another example method of communication implemented at a terminal device in accordance with some embodiments of the present disclosure;
[0029] FIG. 12 illustrates a flowchart of another example method of communication implemented at a terminal device in accordance with some embodiments of the present disclosure; and
[0030] FIG. 13 is a simplified block diagram of a device that is suitable for implementing embodiments of the present disclosure.
[0031] Throughout the drawings, the same or similar reference numerals represent the same or similar element.DETAILED DESCRIPTION
[0032] Principle of the present disclosure will now be described with reference to some 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 limitations as to the scope of the disclosure. The disclosure described herein can be implemented in various manners other than the ones described below.
[0033] 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.
[0034] 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, device on vehicle for V2X communication where X means pedestrian, vehicle, or infrastructure / network, devices for integrated access and backhaul (IAB) , small data transmission (SDT) , mobility, multicast and broadcast services (MBS) , positioning, dynamic / flexible duplex in commercial networks, reduced capability (RedCap) , 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.
[0035] 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) , Network-controlled Repeaters, and the like.
[0036] 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.
[0037] The terminal or the network device may work on several frequency ranges, e.g. FR1 (410 MHz to 7125 MHz) , FR2 (24.25GHz to 71GHz) , frequency band larger than 100GHz as well as Tera Hertz (THz) . It can further work on licensed / unlicensed / shared spectrum. The terminal device may have more than one connections with the network devices under MR-DC application scenario. The terminal device or the network device can work on full duplex, flexible duplex and cross division duplex modes.
[0038] The network device may have the function of network energy saving, self-organizing networks (SON) / minimization of drive tests (MDT) . The terminal may have the function of power saving.
[0039] 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.
[0040] In one embodiment, 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 one embodiment, the first network device may be a first RAT device and the second network device may be a second RAT device. In one embodiment, 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 one embodiment, information A may be transmitted to the terminal device from the first network device and information B may be transmitted to the terminal device from the second network device directly or via the first network device. In one embodiment, 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.
[0041] 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. The term ‘and / or’ indicates that there may be three relationships. For example, A and / or B may indicate cases includes ‘only A’ , ‘both A and B’ , and ‘only B’ . The term ‘at least one of the following items’ or a similar expression thereof refers to any combination of these items, including any combination of a single item or a plurality of items. For example, ‘at least one of A, B, or C’ may represent A, B, C, ‘A and B’ , ‘A and C’ , ‘B and C’ , or ‘A, B and C’ . Other definitions, explicit and implicit, may be included below.
[0042] 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.
[0043] In the context of the present disclosure, the term ‘AI / ML’ may be interchangeably used with ‘ML’ or ‘AI’ . The term ‘AI / ML model’ may be interchangeably used with ‘ML model’ or ‘AI model’ . The term ‘a serving cell’ herein may be interchangeably used with ‘a source cell’ . The term ‘a neighbor cell’ herein may be interchangeably used with ‘a target cell’ or ‘a candidate cell’ .
[0044] In the context of the present disclosure, the term ‘measurement event’ may refer to an event triggering mobility management, for example, Event A1, Event A2, Event A3, Event A4, Event A5, Event A6, Event B1, Event B2, or any other similar events existing or to be developed in future.
[0045] In the context of the present disclosure, the term ‘correct prediction’ herein means that a model inference can work well, e.g., both a predicted event and an actual event are fulfilled, and / or time instances of the predicted event and the actual event is close to each other. The term ‘incorrect prediction’ herein means that a model inference cannot work well, e.g., one of a predicted event and an actual event is not fulfilled, and / or time instances of the predicted event and the actual event is not close to each other.
[0046] In the context of the present disclosure, the term ‘time instance’ herein may refer to a time point or a time range. The term ‘time instance’ may be interchangeably used with ‘time point’ or ‘time range’ . The term ‘measured results’ herein may refer to any suitable real measurement metrics, such as reference signal received power (RSRP) , reference signal received quality (RSRQ) , or signal-to-interference-plus-noise ratio (SINR) . The term ‘signal strength or quality’ herein may refer to any suitable measured values, such as RSRP, RSRQ, or SINR. The term ‘measured results’ herein may refer to real measurement results of cell (serving cell or neighbor cell) or beam or frequency measurements. The term ‘model input’ may indicate a set of measured results (i.e., real measurement results) in an observation window. The term ‘model output’ may indicate a set of predicted results in a prediction window.
[0047] Embodiments of the present disclosure provide solutions of model management for a mobility related model so as to facilitate mobility enhancement. In one aspect, a terminal device may transmit a predicted result of a model inference to a network device, the predicted result indicating that a handover is to occur at a first time instance. In accordance with a determination that a command of the handover is received and a first event for a model monitoring does not occur, the terminal device may evaluate the model inference based on a second event for the model monitoring. In this way, model monitoring for handover prediction in the case of reception of a handover command may be carried out and performance of model monitoring may be enhanced.
[0048] In another aspect, a terminal device may determine, based on a model inference, that a first time of stay in a target cell of a handover is predicted. The terminal device may store the first time of stay, and determine a second time of stay in the target cell after the procedure of the handover. Then the terminal device may evaluate the model inference based on a comparison between the first time of stay and the second time of stay. In this way, model monitoring for time of stay prediction may be carried out.
[0049] In another aspect, a terminal device may transmit, to a network device, at least one of the following: first information of a prediction of a model; and second information indicating update of an input requirement of the model. In this way, model input and / or output information may be reported to a network (NW) .
[0050] In another aspect, upon determination that a first set of outputs of a model is problematic, a terminal device may perform a first operation for handling the first set of outputs, and perform a model prediction with a second set of outputs of the model. In this way, model output may be managed and efficient work of an AI model may be facilitated.
[0051] Principles and implementations of the present disclosure will be described in detail below with reference to the figures.
[0052] EXAMPLE OF COMMUNICATION NETWORK
[0053] FIG. 1 illustrates a schematic diagram of an example communication network 100 in which some embodiments of the present disclosure can be implemented. As shown in FIG. 1, the communication network 100 may include a terminal device 110 and network devices 120 and 130. The network device 120 may provide one or more cells (for convenience, only a cell 121 is shown) to serve the terminal device 110. The network device 130 may provide one or more cells (for convenience, only a cell 131 is shown) to serve the terminal device 110. In the example of FIG. 1, the terminal device 110 is located in the cell 121 and served by the network device 120.
[0054] The terminal device 110 may have a plurality of beams (not shown) , and each of the network devices 120 and 130 may have a plurality of beams (not shown) . A channel (or called as a sub-channel in this case) may be formed between one of the plurality of beams of the terminal device 110 and one of the plurality of beams of the network device 120 or 130. The terminal device 110 may transmit information to the network device 120 or 130, or receive information from the network device 120 or 130 via one or more sub-channels.
[0055] It is to be understood that the number of devices or cells in FIG. 1 is given for the purpose of illustration without suggesting any limitations to the present disclosure. The communication network 100 may include any suitable number of network devices and / or terminal devices and / or other network elements and / or cells adapted for implementing implementations of the present disclosure.
[0056] As shown in FIG. 1, the terminal device 110 and each of the network devices 120 and 130 may communicate with each other via Uu interface. The network device 120 and the network device 130 may communication with each other via Xn interface. The communications in the communication network 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.
[0057] FIG. 2 illustrates a schematic diagram 200 of an AI common framework in which some embodiments of the present disclosure can be implemented. As shown in FIG. 2, training data, monitoring data and inference data may be obtained through an entity of data collection 210.
[0058] With reference to FIG. 2, the training data may be used for model training by an entity of model training 220. An AI model may be trained or updated through the model training. The trained or updated AI model may be stored by an entity of model storage 230.
[0059] With reference to FIG. 2, the monitoring data may be used for model management by an entity of model management 240. The entity of model management 240 may send performance feedback or retraining request to the entity of model training 220 for model retraining. The entity of model management 240 may send model transfer or delivery request to the entity of model storage 230, and the entity of model storage 230 may transfer or delivery the stored AI model to an entity of model inference 250.
[0060] As shown in FIG. 2, the inference data may be used for model inference by the entity of model inference 250. With the inference data as an input of the AI model, an output data set may be obtained by the model inference. The entity of model inference 250 may provide a monitoring output to the entity of model management 240 and receive an indication of model selection / activation / deactivation / switching / fallback from the entity of model management 240.
[0061] So far, life cycle management (LCM) of an AI model is described with reference to FIG. 2. It is to be understood that although the entities 210 to 250 are shown or described as separate entities in FIG. 2, these entities 210-250 may be implemented in one or more physical entities.
[0062] Refer back to FIG. 1, in some embodiments, an AI model (i.e., a mobility related model) for mobility management may be deployed at the terminal device 110. In some embodiments, the AI model may be used to predict occurrence of a measurement event for a handover. For example, an output of the AI model may indicate that the measurement event is to occur or not occur at a future time instance. In some embodiments, the AI model may be used to predict a target cell for a handover. For example, an output of the AI model may indicate that a handover to a target cell is to occur at a future time instance. In some embodiments, the AI model may be used to predict a time of stay in a target cell of a handover. For example, an output of the AI model may indicate the time of stay. It is to be understood that the AI model for mobility management may adopt any other suitable output forms, and the present disclosure does not limit this aspect.
[0063] For all AI based models, a model monitoring plays an important role for AI performance. A typical method of the model monitoring is to compare an output of a model prediction with a ground truth value and determine whether the model prediction is precise. Then for a measurement event prediction or target cell prediction, it is needed to compare a predicted event with an actual event. However, usually a terminal device will receive a handover command around a time instance of the predicted event, and the terminal device may not perform the model monitoring for a period after reception of the handover command if the actual event has not occurred yet. Thus the terminal device may not always derive the actual event for model monitoring, and the model monitoring may not be performed well.
[0064] Further, a model monitoring for a time of stay prediction needs to be designed, and some enhancements on monitoring handling and report are also needed.
[0065] Embodiments of the present disclosure provide solutions of model management for the mobility related model so as to overcome the above and other potential issues. For illustration, the solutions will be detailed below with reference to FIGs. 3 to 7. EXAMPLE IMPLEMENTATION OF MODEL MONITORING FOR HANDOVER PREDICTION
[0066] FIG. 3 illustrates a signaling chart illustrating an example process 300 of communication for model management in accordance with embodiments of the present disclosure. For the purpose of discussion, the process 300 will be described with reference to FIG. 1. The process 300 may involve the terminal device 110 and the network device 120 as illustrated in FIG. 1. It is to be understood that the steps and the order of the steps in FIG. 3 are merely for illustration, and not for limitation. For example, the order of the steps may be changed. Some of the steps may be omitted or any other suitable additional steps may be added. It is assumed that an AI model is deployed at the terminal device 110. The network device 120 provides a serving cell for the terminal device 110, and the network device 130 provides a target cell for the terminal device 110.
[0067] As shown in FIG. 3, the terminal device 110 may transmit 310 a predicted result of a model inference to the network device 120. The predicted result indicates that a handover is to occur at a time instance (for convenience, also referred to as a first time instance herein) . In some embodiments, the terminal device 110 may transmit the predicted result in a measurement report.
[0068] In some embodiments, the terminal device 110 may predict that a measurement event for the handover is to occur at the first time instance. This means that the handover is to occur at the first time instance. In some embodiments, the terminal device 110 may predict that the handover to a target cell is to occur at the first time instance. This also means that the handover is to occur at the first time instance.
[0069] With reference to FIG. 3, upon reception of the predicted result, the network device 120 may transmit 320 a handover command to the terminal device 110. In some embodiments, the network device 120 may transmit a radio resource control (RRC) reconfiguration with synchronization as the handover command. In some embodiments, the terminal device 110 may receive the handover command around the first time instance, e.g., before or at or after the first time instance.
[0070] With reference to FIG. 3, upon reception of the handover command, the terminal device may determine 330 whether a first event for a model monitoring occurs. The first event may refer to an actual event corresponding to the predicted result. In some embodiments, the terminal device may evaluate whether the first event occurs nearby the first time instance or within a time offset to the first time instance.
[0071] In some embodiments, if the predicted result comprises that the measurement event for the handover is predicted to occur at the first time instance, the first event may comprise the measurement event for the handover. That is, for a measurement event prediction, the measurement event may be used as an actual event for model monitoring.
[0072] In some embodiments, if the predicted result comprises that the handover to the target cell is predicted to occur at the first time instance, the first event may comprise that signal quality of the target cell is higher than or equal to a threshold (for convenience, may also referred to as a first threshold herein) . That is, for a target cell prediction, a new event may be defined and used as an actual event for model monitoring. Alternatively, for the target cell prediction, the first event may comprise any existing measurement event for mobility management, e.g., Event A1, A2, A3, A4, A5, A6, B1, or B2.
[0073] As shown in FIG. 3, if the terminal device 110 determines that the first event occurs upon reception of the handover command, the terminal device 110 may determine 340 that the model inference is correct. If the terminal device 110 determines that the first event does not occur upon reception of the handover command, the terminal device 110 may evaluate 350 the model inference based on a second event for the model monitoring. In this way, performance of the model monitoring may be ensured. Some example embodiments of the second event will be described in connection with Embodiments 1 to 3 below.
[0074] Embodiment 1
[0075] In some embodiments, the second event may comprise delaying a handover procedure to evaluate the first event for a duration (for convenience, also referred to as a first duration herein) .
[0076] In other words, when the model monitoring is enabled (i.e., the terminal device 110 will perform a real measurement for the first event) , once the handover is predicted by the model inference, the terminal device 110 may perform the model monitoring. During the model monitoring, if the first event does not occur but the handover command is received, the terminal device 110 may delay to apply the handover procedure, and continue to evaluate the first event for the first duration.
[0077] In some embodiments, the first duration may comprise a duration after reception of the handover command. In some embodiments, the duration may be pre-configured or pre-defined. FIG. 4A illustrates a schematic diagram 400A of an example scenario of model monitoring in accordance with some embodiments of the present disclosure. As shown in FIG. 4A, a predicted result of a model inference indicates that a handover is to occur at a time instance T1. If the first event (i.e., actual event) occurs within time offsets 401 and 402 (i.e., between time instances T2 and T3) to the time instance T1, the model inference is considered as a correct prediction. As shown in FIG. 4A, upon reception of a handover command at a time instance T4, the first event does not occur. In this case, the terminal device 110 may delay to apply a handover procedure and continue to evaluate the first event for a duration since the time instance T4, e.g., until a time instance T5.
[0078] In some embodiments, the first duration may comprise a duration from the reception of the handover command to a time instance (for convenience, also referred to as a second time instance herein) associated with a requirement of the model monitoring. In some embodiments, the requirement of the model monitoring may comprise a time offset to the first time instance of the predicted event. The second time instance may refer to a time instance corresponding to the time offset after the first time instance. Continuing to refer to FIG. 4A, if the first event does not occur upon reception of a handover command at a time instance T4, the terminal device 110 may delay to apply a handover procedure and continue to evaluate the first event for a duration from the time instance T4 to the time instance T3.
[0079] In some embodiments, the first duration may comprise a duration from the reception of the handover command to the first time instance of the predicted event. For example, if the handover command is received before the first time instance, the terminal device 110 may delay to apply the handover procedure until the first time instance of the predicted event. Continuing to refer to FIG. 4A, if the first event does not occur upon reception of a handover command at a time instance T6, the terminal device 110 may delay to apply the handover procedure and continue to evaluate the first event for a duration from the time instance T6 to the time instance T1.
[0080] In some embodiments, if the first event occurs during the first duration, the terminal device 110 may determine that the model inference is correct (i.e., correct prediction) . In some embodiments, if the first event occurs during the first duration, the terminal device 110 may apply the handover procedure immediately.
[0081] In some embodiments, if the first event does not occur during the first duration, the terminal device 110 may determine that the model inference is incorrect. In some embodiments, if the first event does not occur during the first duration, the terminal device 110 may still apply the handover procedure. In some embodiments, if the first event does not occur during the first duration, the terminal device 110 may transmit, to the network device 120, an indication for rejecting the handover command, and cancel the handover procedure.
[0082] In this way, more time for the model monitoring may be provided and accuracy of the model monitoring may be improved.
[0083] Embodiment 2
[0084] In some embodiments, the second event may comprise that evaluating, after applying the handover procedure, a radio link failure (RLF) or a further handover (also referred to as a subsequent handover herein) for a duration (for convenience, also referred to as a second duration herein) .
[0085] In other words, as long as the handover command is received, the terminal device 110 may apply the handover procedure no matter whether the model monitoring is going on or not. However, if the first event does not occur during the model monitoring before applying the handover procedure, the terminal device 110 may evaluate whether the RLF or the further handover occurs for the second duration after handover.
[0086] In some embodiments, the second duration may comprise a duration since reception of the handover command. In some embodiments, the duration may be pre-configured or pre-defined. FIG. 4B illustrates a schematic diagram 400B of another example scenario of model monitoring in accordance with some embodiments of the present disclosure. As shown in FIG. 4B, a predicted result of a model inference indicates that a handover is to occur at a time instance T1. If the first event (i.e., actual event) occurs within time offsets 401 and 402 (i.e., between time instances T2 and T3) to the time instance T1, the model inference is considered as a correct prediction. As shown in FIG. 4B, upon reception of a handover command at a time instance T4, the terminal device 110 may apply a handover procedure at the time instance T4. As the first event does not occur during model monitoring before applying the handover procedure, the terminal device 110 may evaluate whether a RLF or a further handover occurs for a duration since the time instance T4, e.g., until a time instance T7.
[0087] In some embodiments, the second duration may comprise a duration since completion of the handover procedure. In some embodiments, the duration may be pre-configured or pre-defined. Continuing to refer to FIG. 4B, it is assumed that the handover procedure is completed at a time instance T8. As the first event does not occur during model monitoring before applying the handover procedure, the terminal device 110 may evaluate whether the RLF or the further handover occurs for a duration since the time instance T8, e.g., until a time instance T9.
[0088] In some embodiments, if the RLF or the further handover does not occur during the second duration, the terminal device 110 may determine that the model inference (i.e., the previous model inference) is correct. If the RLF or the further handover occurs during the second duration, the terminal device 110 may determine that the model inference (i.e., the previous model inference) is incorrect.
[0089] In this way, upon reception of a handover command, a model monitoring may continue to be performed based on evaluation of a RLF or subsequent handover, and thus accuracy of the model monitoring may be improved.
[0090] Embodiment 3
[0091] In some embodiments, the second event may comprise that evaluating, after applying the handover procedure, the first event once or for a third duration by considering a source cell of the handover as a serving cell for the first event and a target cell of the handover as a neighboring cell for the first event.
[0092] That is, if the first event does not occur upon reception of the handover command, the first event may be continuously evaluated once or for the third duration after handover. During the continuous evaluation, a source cell (i.e., an old serving cell) of the handover is considered as a serving cell for the first event and a target cell of the handover (i.e., a new serving cell) is considered as a neighboring cell for the first event.
[0093] FIG. 4C illustrates a schematic diagram 400C of another example scenario of model monitoring in accordance with some embodiments of the present disclosure. As shown in FIG. 4C, the terminal device 110 is initially located in the cell 121 of the network device 120. Upon reception of a handover command, the terminal device 110 performs a handover from the cell 121 (i.e., old serving or source cell) to the cell 131 (i.e., new serving or source cell) of the network device 130. During a model monitoring before handover, the first event is evaluated by considering the cell 121 as a serving cell for the first event and the cell 131 as a neighboring cell for the first event. During a model monitoring after handover, the first event is still evaluated by considering the cell 121 (i.e., old serving or source cell) as a serving cell for the first event and the cell 131 (i.e., new serving or source cell) as a neighboring cell for the first event.
[0094] For example, the first event is Event A3 and is represented as an equation (1) below. During the model monitoring before or after handover, Mn is always determined for the cell 131 and Mp is always determined for the cell 121. Mn + Ofn + Ocn -Hys > Mp + Ofp + Ocp + Off (1)
[0095] where Mn denotes a measurement result of a neighboring cell, Ofn denotes a frequency-specific offset value for the neighboring cell, Ocn denotes a cell-specific offset value for the neighboring cell, Hys denotes a hysteresis parameter for the first event, Mp denotes a measurement result of a serving cell (i.e., a special cell (SpCell) ) , Ofp denotes a frequency-specific offset value for the serving cell, Ocp denotes a cell-specific offset value for the serving cell, and Off denotes an offset parameter for the first event.
[0096] Alternatively, the second event may comprise that signal quality of a neighboring cell after the handover is lower than or equal to signal quality of a serving cell after the handover. In other words, an event may be redefined for the model monitoring.
[0097] For example, in the case that the first event is Event A3, the second event may be represented as an equation (2) below. Mn + Ofn + Ocn ± Hys < Mp + Ofp + Ocp + Off (2)
[0098] where Mn denotes a measurement result of a neighboring cell, Ofn denotes a frequency-specific offset value for the neighboring cell, Ocn denotes a cell-specific offset value for the neighboring cell, Hys denotes a hysteresis parameter for the first event, Mp denotes a measurement result of a serving cell (i.e., a special cell (SpCell) ) , Ofp denotes a frequency-specific offset value for the serving cell, Ocp denotes a cell-specific offset value for the serving cell, and Off denotes an offset parameter for the first event.
[0099] That is, during the model monitoring before handover, the terminal device 110 may evaluate whether an event represented in the equation (1) occurs by considering the cell 121 as a serving cell and the cell 131 as a neighboring cell. During the model monitoring after handover, the terminal device 110 may evaluate whether an event represented in the equation (2) occurs by considering the cell 131 as a serving cell and the cell 121 as a neighboring cell.
[0100] In some embodiments, the third duration for the continuous evaluation may comprise a duration since reception of the handover command. In some embodiments, the duration may be pre-configured or pre-defined. In some embodiments, the third duration for the continuous evaluation may comprise a duration since completion of the handover procedure. In some embodiments, the duration may be pre-configured or pre-defined.
[0101] In some embodiments, if the first event occurs during the third duration, the terminal device 110 may determine that the model inference is correct. If the first event does not occur during the third duration, the terminal device 110 may determine that the model inference is incorrect.
[0102] In this way, after a handover procedure, a model monitoring may continue to be performed based on continuous evaluation of a previous event, and thus accuracy of the model monitoring may be improved.
[0103] Refer back to FIG. 3, in some embodiments, the terminal device 110 may determine 360 that the first event occurs and the predicted result is absent. In this case, the terminal device 110 may keep performing the model inference until earlier one of the following: reception of the handover command, or a time offset after a third time instance at which the first event occurs.
[0104] FIG. 4D illustrates a schematic diagram 400D of another example scenario of model monitoring in accordance with some embodiments of the present disclosure. As shown in FIG. 4D, the first event (i.e., actual event) occurs at a time instance T10, but no predicted result is output until the time instance T10. In this case, the terminal device 110 may keep performing AI prediction for the predicted result until a time instance T11 corresponding to a time offset 403 after the time instance T10, or until reception of a handover command at a time instance T12, which is earlier.
[0105] In some embodiments, if the predicted result is output until the reception of the handover command or the time offset, the terminal device 110 may determine that the model inference is correct. If the predicted result is still absent until the reception of the command of the handover or the time offset, the terminal device 110 may determine that the model inference is incorrect.
[0106] In this way, accuracy of model monitoring may be improved.
[0107] Continuing to refer to FIG. 3, the terminal device 110 may perform 370 a model LCM procedure, such as model activation or deactivation, model update, model switch, etc. In some embodiments, the terminal device 110 may perform the model LCM procedure based on a result of the model monitoring (i.e., the model inference is correct or incorrect) . In some embodiments, the terminal device 110 may perform the model LCM procedure based on number of monitoring results. For example, the terminal device 110 may deactivate the model inference if the model inference is incorrect for N times.
[0108] With the process 300, a model monitoring for a handover prediction may be well performed. It is to be understood that operations described above in connection with FIGs. 3 to 4D may be performed separately or in any suitable combinations.
[0109] EXAMPLE IMPLEMENTATION OF MODEL MONITORING FOR TIME OF STAY PREDICATION
[0110] Embodiments of the present disclosure provide a solution of model monitoring for time of stay prediction. This solution will be described in connection with FIG. 5.
[0111] FIG. 5 illustrates a signaling chart illustrating another example process 500 of communication in accordance with embodiments of the present disclosure. For the purpose of discussion, the process 500 will be described with reference to FIG. 1. The process 500 may involve the terminal device 110 and the network device 120 as illustrated in FIG. 1. It is to be understood that the steps and the order of the steps in FIG. 5 are merely for illustration, and not for limitation. For example, the order of the steps may be changed. Some of the steps may be omitted or any other suitable additional steps may be added. It is assumed that one or more models are deployed at the terminal device 110. The terminal device 110 is served by the network device 120, and the network device 130 provides a target cell for the terminal device 110.
[0112] As shown in FIG. 5, the network device 120 may transmit 510, to the terminal device 110, a configuration for a model monitoring for a time of stay prediction. In some embodiments, the configuration may indicate a threshold (for convenience, also referred to as a second threshold herein) used for a comparison between a predicted time of stay (also referred to as a first time of stay herein) and an actual time of stay (also referred to as a second time of stay herein) . It is to be noted that the configuration may comprise any suitable information and the present disclosure does not limit this aspect.
[0113] As shown in FIG. 5, the terminal device 110 may determine 520, based on a model inference, that a time of stay (i.e., the first time of stay) in a target cell of a handover is predicted.
[0114] With reference to FIG. 5, the terminal device 110 may store 530 the first time of stay. In some embodiments, once the first time of stay is predicted, the terminal device 110 may store information of the first time of stay. In some embodiments, the terminal device 110 may perform a procedure of the handover to the target cell without deleting the information of the first time of stay. In other words, after the procedure of the handover, the information of the first time of stay is stored at the terminal device 110.
[0115] With reference to FIG. 5, the terminal device 110 may determine or record 540 a time of stay (i.e., the second time of stay) in the target cell after the procedure of the handover.
[0116] In some embodiments, if the terminal device 110 camps on the target cell, the terminal device 110 may start the recording of the second time of stay. In some embodiments, if the procedure of the handover is completed, the terminal device 110 may start the recording of the second time of stay.
[0117] In some embodiments, if a RLF occurs during camping on the target cell, the terminal device 110 may stop the recording of the second time of stay. The recorded value is considered as a final value of the second time of stay for model monitoring.
[0118] In some embodiments, if a beam failure detection (BFD) occurs during camping on the target cell, the terminal device 110 may stop the recording of the second time of stay. The recorded value is considered as a final value of the second time of stay for model monitoring.
[0119] In some embodiments, if a further handover occurs during camping on the target cell, the terminal device 110 may stop the recording of the second time of stay. The recorded value is considered as a final value of the second time of stay for model monitoring.
[0120] In some embodiments, during camping on the target cell, if the terminal device 110 transits from a connected state to an idle or inactive state while a cell different from the target cell is selected, the terminal device 110 may stop the recording of the second time of stay. The recorded value is considered as a final value of the second time of stay for model monitoring.
[0121] In some embodiments, if the terminal device 110 transits from a connected state to an idle or inactive state while the target cell (i.e., the same cell) is selected during camping on the target cell, the terminal device 110 may keep the recording of the second time of stay. Alternatively, if the terminal device 110 transits from the connected state to the idle or inactive state while the target cell (i.e., the same cell) is selected during camping on the target cell, the terminal device 110 may cancel the evaluating of the model inference, i.e., cancel the model monitoring for this model inference.
[0122] In some embodiments, if a cell reselection occurs during camping on the target cell, the terminal device 110 may stop the recording of the second time of stay. The recorded value is considered as a final value of the second time of stay for model monitoring.
[0123] Continuing to refer to FIG. 5, the terminal device 110 may evaluate 550 the model inference based on the comparison between the first time of stay and the second time of stay.
[0124] In some embodiments, if a difference between the first time of stay and the second time of stay is smaller than or equal to the second threshold, the terminal device 110 may determine that the model inference is correct. In other words, if the second time of stay is nearby the first time of stay or within a range to the first time of stay, the model inference is considered as correct. In some embodiments, if the difference between the first time of stay and the second time of stay is larger than or equal to the second threshold, the terminal device 110 may determine that the model inference is incorrect. In other words, if the second time of stay is not nearby the first time of stay or not within the range to the first time of stay, the model inference is considered as incorrect.
[0125] Continuing to refer to FIG. 5, the terminal device 110 may perform 560 a model LCM procedure, such as model activation or deactivation, model update, model switch, etc. In some embodiments, the terminal device 110 may perform the model LCM procedure based on a result of the model monitoring (i.e., the model inference is correct or incorrect) . In some embodiments, the terminal device 110 may perform the model LCM procedure based on number of monitoring results. For example, the terminal device 110 may deactivate the model inference if the model inference is incorrect for N times.
[0126] With the process 500, a model monitoring for a time of stay prediction may be well performed. It is to be understood that operations described above in connection with FIG. 5 may be performed separately or in any suitable combinations.
[0127] EXAMPLE IMPLEMENTATION OF MODEL INFORMATION REPORTING
[0128] Embodiments of the present disclosure provide a solution of reporting model information. The solution will be described in connection with FIG. 6.
[0129] FIG. 6 illustrates a signaling chart illustrating another example process 600 of communication in accordance with embodiments of the present disclosure. For the purpose of discussion, the process 600 will be described with reference to FIG. 1. The process 600 may involve the terminal device 110 and the network device 120 as illustrated in FIG. 1. It is to be understood that the steps and the order of the steps in FIG. 6 are merely for illustration, and not for limitation. For example, the order of the steps may be changed. Some of the steps may be omitted or any other suitable additional steps may be added. It is assumed that one or more models are deployed at the network device 120. The terminal device 110 is served by the network device 120.
[0130] As shown in FIG. 6, the terminal device 110 may transmit 610, to the network device 120, information (for convenience, also referred to as first information herein) of a prediction of a model. In some embodiments, the first information may comprise a type of the prediction. Alternative or additionally, the first information may comprise an indication of whether the L3 filtering procedure is inside the model.
[0131] In some embodiments, the type of the prediction may indicate that an input of the model is a measured result in a layer 1 (L1) beam level and an output of the model is a predicted result in a L1 beam level. With the predicted result of L1 beam level, a layer 3 (L3) cell level result may be derived. In these embodiments, a L3 filtering procedure is not implemented inside the model.
[0132] In some embodiments, the type of the prediction may indicate that an input of the model is a measured result in a L1 cell level and an output of the model is a predicted result in a L3 cell level. In this case, a L3 cell level result may be directly predicted based on a measured L1 cell level result. In these embodiments, a L3 filtering procedure is implemented inside the model.
[0133] In some embodiments, the type of the prediction may indicate that an input of the model is a measured result in a L3 cell level and an output of the model is a predicted result in a L3 cell level. In these embodiments, a L3 filtering procedure is not implemented inside the model.
[0134] In some embodiments, the type of the prediction may indicate that an input of the model is a measured result in a L1 beam level and an output of the model is a predicted result in a L3 cell level. In this case, a L3 cell level result may be directly predicted based on a measured L1 beam level result. In these embodiments, a L3 filtering procedure is implemented inside the model.
[0135] In some embodiments, the type of the prediction may indicate that an input of the model is a measured result in a L1 beam level and an output of the model is a predicted result in a L3 beam level. In these embodiments, a L3 filtering procedure is implemented inside the model.
[0136] In some embodiments, the type of the prediction may indicate that an input of the model is a measured result in a L3 beam level and an output of the model is a predicted result in a L3 beam level. In these embodiments, a L3 filtering procedure is not implemented inside the model.
[0137] In some embodiments, the terminal device 110 may transmit the first information in a report (i.e., a monitoring report) for a monitoring of the model. In some embodiments, the terminal device 110 may transmit the first information in a report (i.e., a model capability report) for capability of the model. In some embodiments, the terminal device 110 may transmit the first information in a report (i.e., a model applicability report) for applicability of the model. In some embodiments, the terminal device 110 may transmit the first information in a report (i.e., a model activation report) for activation of the model.
[0138] In this way, model input and output information may be reported to NW.
[0139] With reference to FIG. 6, the terminal device 110 may transmit 620, to the network device 120, information (for convenience, also referred to as second information herein) indicating update of an input requirement of the model.
[0140] Generally, NW needs to configure a measurement resource related to one or more model inputs so that a terminal device could perform real measurement for the one or more model inputs. However, if a model is deactivated or updated, it means that a model input requirement has been changed, the configured measurement resource for model input may not be needed anymore. With a reporting of the second information, management of the measurement resource related to model inputs may be facilitated.
[0141] In some embodiments, if the model is deactivated or updated, the terminal device 110 may transmit the second information to the network device 120. In some embodiments, if the input requirement of the model is updated, the terminal device 110 may transmit the second information to the network device.
[0142] In some embodiments, the terminal device 110 may transmit the second information in a request for de-configuring or re-configuring a measurement resource related to the update of the model input requirement. In some embodiments, the request may further comprise information (e.g., a timer) of a duration in which the request is valid.
[0143] For example, for an initiation stage, e.g., when the model is deployed at the terminal device 110 side, the terminal device 110 may need to send assistance information related to model input requirement for acquiring a measurement resource. Alternatively or additionally, the terminal device 110 may carry a valid timer within the assistance information, to let NW be aware that how long the model will work.
[0144] In some embodiments, the terminal device 110 may report the second information without requesting the de-configuration or re-configuration of the measurement resource. Modification of the measurement resource may be up to NW implementation.
[0145] It is to be noted that, the term ‘ameasurement resource related to one or more model inputs’ may mean that if an input is cell 1 and cell 2, NW should provide a measurement object corresponding to the cell 1 or 2, so that a terminal device could perform real measurement on the cell 1 or 2.
[0146] With the process 600, model information reporting may be carried out and model management may be facilitated. It is to be understood that operations described above in connection with FIG. 6 may be performed in any suitable combinations.
[0147] EXAMPLE IMPLEMENTATION OF MODEL OUTPUT MANAGEMENT
[0148] Embodiments of the present disclosure provide a solution of managing a set of model outputs. In the solution, during a model monitoring procedure for a cluster-based AI model, if one or more outputs are problematic, a terminal device may need to handle with the problematic output, meanwhile the AI model is still working normally. This solution will be described in connection with FIG. 7.
[0149] In the context of the present disclosure, for a cluster-based AI model, an input is a set of cell or beam results and an output is the same set of cell or beam results for temporal domain prediction and different sets of cell or beam results for spatial domain prediction.
[0150] FIG. 7 illustrates a signaling chart illustrating another example process 700 of communication in accordance with embodiments of the present disclosure. For the purpose of discussion, the process 700 will be described with reference to FIG. 1. The process 700 may involve the terminal device 110 and the network device 120 as illustrated in FIG. 1. It is to be understood that the steps and the order of the steps in FIG. 7 are merely for illustration, and not for limitation. For example, the order of the steps may be changed. Some of the steps may be omitted or any other suitable additional steps may be added. It is assumed that one or more models are deployed at the network device 120. The terminal device 110 is served by the network device 120.
[0151] As shown in FIG. 7, the network device 120 may transmit 710, to the terminal device 110, a configuration for a model output management. In some embodiments, the configuration may indicate a threshold (for convenience, also referred to as a third threshold herein) used for determination of a problematic model output. It is to be noted that the configuration may comprise any suitable information and the present disclosure does not limit this aspect.
[0152] As shown in FIG. 7, the terminal device 110 may determine 720 that a set of outputs (for convenience, also referred to as a first set of outputs herein) of a model is problematic. The first set of outputs may comprise one or more outputs.
[0153] In some embodiments, the terminal device 110 may determine a predicted result on an output in the first set of outputs based on a model inference, and determine a measured result on the output by performing a real measurement on the output. If a difference between the predicted result and the measured result is larger than or equal to the third threshold, the terminal device 110 may determine that the output is problematic.
[0154] With reference to FIG. 7, upon determination that the first set of outputs is problematic, the terminal device 110 may perform 730 an operation (for convenience, also referred to as a first operation herein) for handling the first set of outputs.
[0155] In some embodiments for the first operation, the terminal device 110 may deactivate the first set of outputs, and transmit information of the first set of outputs to the network device 120. In other words, the terminal device 110 may deactivate the one or more problematic outputs while the model is still working normally, and report the one or more problematic outputs to NW.
[0156] In some embodiments for the first operation, the terminal device 110 may transmit information of the first set of outputs to the network device 120, and receive an indication of deactivating the first set of outputs. Based on the indication, the terminal device 110 may deactivate the first set of outputs.
[0157] In some embodiments for the first operation, the terminal device 110 may disable the first set of outputs. That is, the terminal device 110 may still inference the one or more problematic outputs, but not use them for a mobility related procedure. In this case, the terminal device 110 may continue to monitor the first set of outputs. If there is no problem for an output in the first set of outputs, the terminal device 110 may enable the output, i.e., resume the output.
[0158] In some embodiments, the information of the first set of outputs may indicate a frequency that is problematic, e.g., an information element (IE) ‘ARFCN-ValueNR’ . In some embodiments, the information of the first set of outputs may indicate a cell identity (ID) that is problematic, e.g., a physical cell identity (PCI) , or an IE ‘cell identity’ , or both a public land mobile network (PLMN) and IE ‘cell identity’ . In some embodiments, the information of the first set of outputs may indicate a beam index that is problematic, e.g., a channel status information (CSI) index, or a synchronization signal and physical broadcast channel block (SSB) index.
[0159] With reference to FIG. 7, upon determination that the first set of outputs is problematic, the terminal device 110 may perform 740 a model prediction by using another set of outputs (for convenience, also referred to as a second set of outputs herein) of the model. The second set of outputs may comprise one or more outputs. The second set of outputs is normal (i.e., not problematic) .
[0160] FIG. 8 illustrates a schematic diagram 800 of an example scenario of model output management in accordance with some embodiments of the present disclosure. As shown in FIG. 8, inputs of an AI model 810 are cell 1, cell 2 and cell 3, and outputs of the AI model 810 are cell 4 and cell 5. After model monitoring by performing real measurements on cell 4 and cell 5, cell 4 is determined as problematic. As a result, cell 4 may be deactivated, or cell 4 may be not used. Cell 5 is normally used.
[0161] With the process 700, model output may be managed and efficient work of an AI model may be facilitated. It is to be understood that operations described above in connection with FIG. 7 may be performed in any suitable combinations.
[0162] It is also to be understood that operations or processes described above in connection with FIGs. 3 to 8 may be performed separately or in any suitable combination.
[0163] EXAMPLE IMPLEMENTATION OF METHODS
[0164] Corresponding to the above process, embodiments of the present disclosure provide methods of communication implemented at a terminal device and a network device. These methods will be described below with reference to FIGs. 9 to 12.
[0165] FIG. 9 illustrates a flowchart of an example method 900 of communication implemented at a terminal device in accordance with some embodiments of the present disclosure. For example, the method 900 may be performed at the terminal device 110 as shown in FIG. 1. For the purpose of discussion, in the following, the method 900 will be described with reference to FIG. 1. It is to be understood that the method 900 may include additional blocks not shown and / or may omit some blocks as shown, and the scope of the present disclosure is not limited in this regard. The terminal device 110 is served by the network device 120.
[0166] At block 910, the terminal device 110 may transmit a predicted result of a model inference to the network device 120. The predicted result indicates that a handover is to occur at a first time instance.
[0167] At block 920, the terminal device 110 may determine that a command of the handover is received and a first event for a model monitoring does not occur.
[0168] In some embodiments, the predicted result may comprise that a measurement event for the handover is predicted to occur at the first time instance, and the first event may comprise the measurement event for the handover.
[0169] In some embodiments, the predicted result may comprise that the handover to a target cell is predicted to occur at the first time instance. The first event may comprise that signal quality of the target cell is higher than or equal to a first threshold. Alternatively, the first event may comprise a measurement event for the handover.
[0170] At block 930, the terminal device 110 may evaluate the model inference based on a second event for the model monitoring.
[0171] In some embodiments, the second event may comprise delaying a procedure of the handover to evaluate the first event for a first duration. In some embodiments, the first duration may comprise one of the following: a duration after reception of the command of the handover; a duration from the reception of the command of the handover to a second time instance associated with a requirement of the model monitoring; or a duration from the reception of the command of the handover to the first time instance.
[0172] In some embodiments, the terminal device 110 may evaluate the model inference by:in accordance with a determination that the first event occurs during the first duration, determining that the model inference is correct; and in accordance with a determination that the first event does not occur during the first duration, determining that the model inference is incorrect.
[0173] In some embodiments, in accordance with a determination that the first event occurs during the first duration, the terminal device 110 may apply the procedure of the handover. In some embodiments, in accordance with a determination that the first event does not occur during the first duration, the terminal device 110 may apply the procedure of the handover. In some embodiments, in accordance with a determination that the first event does not occur during the first duration, the terminal device 110 may transmit, to the network device 120, an indication for rejecting the command of the handover, and cancel the procedure of the handover.
[0174] In some embodiments, the second event may comprise that evaluating, after applying a procedure of the handover, a radio link failure or a further handover for a second duration. In some embodiments, the second duration may comprise one of the following: a duration since reception of the command of the handover; or a duration since completion of the procedure of the handover.
[0175] In some embodiments, the terminal device 110 may evaluate the model inference by:in accordance with a determination that the radio link failure or the further handover does not occur during the second duration, determining that the model inference is correct; and in accordance with a determination that the radio link failure or the further handover occurs during the second duration, determining that the model inference is incorrect.
[0176] In some embodiments, the second event may comprise that evaluating, after applying a procedure of the handover, the first event for a third duration by considering a source cell of the handover as a serving cell for the first event and a target cell of the handover as a neighboring cell for the first event. Alternatively, the second event may comprise that signal quality of a neighboring cell after the handover is lower than or equal to signal quality of a serving cell after the handover.
[0177] In some embodiments, the third duration may comprise one of the following: a duration since reception of the command of the handover; or a duration since completion of a procedure of the handover. In some embodiments, the terminal device 110 may evaluate the model inference by: in accordance with a determination that the first event occurs during the third duration, determining that the model inference is correct; and in accordance with a determination that the first event does not occur during the third duration, determining that the model inference is incorrect.
[0178] In some embodiments, in accordance with a determination that the first event occurs and the predicted result is absent, the terminal device 110 may keep performing the model inference until earlier one of the following: reception of the command of the handover, or a time offset after a third time instance at which the first event occurs. In accordance with a determination that the predicted result is output until the reception of the command of the handover or the time offset, the terminal device 110 may determine that the model inference is correct. In accordance with a determination that the predicted result is absent until the reception of the command of the handover or the time offset, the terminal device 110 may determine that the model inference is incorrect.
[0179] With the method 900, model monitoring for handover prediction in the case of reception of a handover command may be carried out and performance of model monitoring may be enhanced.
[0180] FIG. 10 illustrates a flowchart of another example method 1000 of communication implemented at a terminal device in accordance with some embodiments of the present disclosure. For example, the method 1000 may be performed at the terminal device 110 as shown in FIG. 1. For the purpose of discussion, in the following, the method 1000 will be described with reference to FIG. 1. It is to be understood that the method 1000 may include additional blocks not shown and / or may omit some blocks as shown, and the scope of the present disclosure is not limited in this regard. The terminal device 110 is served by the network device 120.
[0181] At block 1010, the terminal device 110 may determine, based on a model inference, that a first time of stay in a target cell of a handover is predicted.
[0182] At block 1020, the terminal device 110 may store the first time of stay.
[0183] At block 1030, the terminal device 110 may determine a second time of stay in the target cell after the procedure of the handover.
[0184] In some embodiments, the terminal device 110 may determine the second time of stay by: in accordance with a determination that the terminal device camps on the target cell or the procedure of the handover is completed, starting a recording of the second time of stay; and stopping the recording based on at least one of the following: a radio link failure occurs during camping on the target cell; a beam failure detection occurs during camping on the target cell; a further handover occurs during camping on the target cell; the terminal device 110 transits from a connected state to an idle or inactive state while a cell different from the target cell is selected during camping on the target cell; or a cell reselection occurs during camping on the target cell.
[0185] In some embodiments, in accordance with a determination that the terminal device 110 transits from a connected state to an idle or inactive state while the target cell is selected during camping on the target cell, the terminal device 110 may keep the recording of the second time of stay. Alternatively, the terminal device 110 may cancel the evaluating of the model inference.
[0186] At block 1040, the terminal device 110 may evaluate the model inference based on a comparison between the first time of stay and the second time of stay.
[0187] In some embodiments, the terminal device 110 may evaluate the model inference by:in accordance with a determination that a difference between the first time of stay and the second time of stay is smaller than or equal to a second threshold, determining that the model inference is correct; and in accordance with a determination that the difference between the first time of stay and the second time of stay is larger than the second threshold, determining that the model inference is incorrect.
[0188] With the method 1000, model monitoring for time of stay prediction may be carried out.
[0189] FIG. 11 illustrates a flowchart of another example method 1100 of communication implemented at a terminal device in accordance with some embodiments of the present disclosure. For example, the method 1100 may be performed at the terminal device 110 as shown in FIG. 1. For the purpose of discussion, in the following, the method 1100 will be described with reference to FIG. 1. It is to be understood that the method 1100 may include additional blocks not shown and / or may omit some blocks as shown, and the scope of the present disclosure is not limited in this regard. The terminal device 110 is served by the network device 120.
[0190] At block 1110, the terminal device 110 may transmit, to the network device 120, at least one of the following: first information of a prediction of a model; or second information indicating update of an input requirement of the model.
[0191] In some embodiments, the first information may comprise at least one of the following: a type of the prediction; or an indication of whether a L3 filtering procedure is inside the model.
[0192] In some embodiments, the type may indicate one of the following: an input of the model is a measured result in a L1 beam level and an output of the model is a predicted result in a L1 beam level; an input of the model is a measured result in a L1 or L3 cell level and an output of the model is a predicted result in a L3 cell level; an input of the model is a measured result in a L1 beam level and an output of the model is a predicted result in a L3 cell level; or an input of the model is a measured result in a L1 or L3 beam level and an output of the model is a predicted result in a L3 beam level.
[0193] In some embodiments, the terminal device 110 may transmit the first information by at least one of the following: transmitting the first information in a report for a monitoring of the model; transmitting the first information in a report for capability of the model; transmitting the first information in a report for applicability of the model; or transmitting the first information in a report for activation of the model.
[0194] In some embodiments, the terminal device 110 may transmit the second information by: in accordance with a determination that the model is deactivated or updated, transmitting the second information to the network device; or in accordance with a determination that the input requirement of the model is updated, transmitting the second information to the network device.
[0195] In some embodiments, the terminal device 110 may transmit the second information by: transmitting the second information in a request for de-configuring or re-configuring a measurement resource related to the update of the input requirement. In some embodiments, the request may further comprise information of a duration in which the request is valid.
[0196] With the method 1100, model input and / or output information may be reported to NW.
[0197] FIG. 12 illustrates a flowchart of another example method 1200 of communication implemented at a terminal device in accordance with some embodiments of the present disclosure. For example, the method 1200 may be performed at the terminal device 110 as shown in FIG. 1. For the purpose of discussion, in the following, the method 1200 will be described with reference to FIG. 1. It is to be understood that the method 1200 may include additional blocks not shown and / or may omit some blocks as shown, and the scope of the present disclosure is not limited in this regard. The terminal device 110 is served by the network device 120.
[0198] At block 1210, the terminal device 110 may determine that a first set of outputs of a model is problematic.
[0199] In some embodiments, the terminal device 110 may determine that the first set of outputs of the model is problematic by: determining a predicted result on an output in the first set of outputs based on a model inference; determining a measured result on the output by performing a real measurement on the output; and in accordance with a determination that a difference between the predicted result and the measured result is larger than or equal to a third threshold, determining that the output is problematic.
[0200] At block 1220, the terminal device 110 may perform a first operation for handling the first set of outputs.
[0201] In some embodiments, the first operation may comprise: deactivating the first set of outputs; and transmitting information of the first set of outputs to the network device 120.
[0202] In some embodiments, the first operation may comprise: transmitting information of the first set of outputs to the network device 120; receiving an indication of deactivating the first set of outputs from the network device 120; and deactivating the first set of outputs based on the indication.
[0203] In some embodiments, the first operation may comprise: disabling the first set of outputs; continuing to monitor the first set of outputs; and in accordance with a determination that there is no problem for an output in the first set of outputs, enabling the output.
[0204] At block 1230, the terminal device 110 may perform a model prediction with a second set of outputs of the model.
[0205] With the method 1200, model output may be managed and efficient work of an AI model may be facilitated.
[0206] It is to be understood that operations of the methods 900, 1000, 1100 and 1200 correspond to that described in connection with FIGs. 3 to 8, and thus other details are not repeated here for conciseness.
[0207] EXAMPLE IMPLEMENTATION OF DEVICES
[0208] FIG. 13 is a simplified block diagram of a device 1300 that is suitable for implementing embodiments of the present disclosure. The device 1300 can be considered as a further example implementation of the terminal device 110 or the network device 120 or the network device 130 as shown in FIG. 1. Accordingly, the device 1300 can be implemented at or as at least a part of the terminal device 110 or the network device 120 or the network device 130.
[0209] As shown, the device 1300 includes a processor 1310, a memory 1320 coupled to the processor 1310, a suitable transceiver 1340 coupled to the processor 1310, and a communication interface coupled to the transceiver 1340. The memory 1310 stores at least a part of a program 1330. The transceiver 1340 may be for bidirectional communications or a unidirectional communication based on requirements. The transceiver 1340 may include at least one of a transmitter 1342 or a receiver 1344. The transmitter 1342 and the receiver 1344 may be functional modules or physical entities. The transceiver 1340 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.
[0210] The program 1330 is assumed to include program instructions that, when executed by the associated processor 1310, enable the device 1300 to operate in accordance with the embodiments of the present disclosure, as discussed herein with reference to FIGs. 1 to 12. The embodiments herein may be implemented by computer software executable by the processor 1310 of the device 1300, or by hardware, or by a combination of software and hardware. The processor 1310 may be configured to implement various embodiments of the present disclosure. Furthermore, a combination of the processor 1310 and memory 1320 may form processing means 1350 adapted to implement various embodiments of the present disclosure.
[0211] The memory 1320 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 1320 is shown in the device 1300, there may be several physically distinct memory modules in the device 1300. The processor 1310 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 1300 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.
[0212] In some embodiments, a terminal device comprises a circuitry configured to: transmit a predicted result of a model inference to a network device, the predicted result indicating that a handover is to occur at a first time instance; and in accordance with a determination that a command of the handover is received and a first event for a model monitoring does not occur, evaluate the model inference based on a second event for the model monitoring.
[0213] In some embodiments, a terminal device comprises a circuitry configured to: determine, based on a model inference, that a first time of stay in a target cell of a handover is predicted; store the first time of stay; determine a second time of stay in the target cell after the procedure of the handover; and evaluate the model inference based on a comparison between the first time of stay and the second time of stay.
[0214] In some embodiments, a terminal device comprises a circuitry configured to: transmit, to a network device, at least one of the following: first information of a prediction of a model; or second information indicating update of an input requirement of the model.
[0215] In some embodiments, a terminal device comprises a circuitry configured to: determine that a first set of outputs of a model is problematic; perform a first operation for handling the first set of outputs; and perform a model prediction with a second set of outputs of the model.
[0216] 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.
[0217] 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.
[0218] 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 12. 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.
[0219] 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.
[0220] 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.
[0221] 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 desirable 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.
[0222] 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:transmit a predicted result of a model inference to a network device, the predicted result indicating that a handover is to occur at a first time instance; andin accordance with a determination that a command of the handover is received and a first event for a model monitoring does not occur, evaluate the model inference based on a second event for the model monitoring.2.The terminal device of claim 1, wherein the predicted result comprises that a measurement event for the handover is predicted to occur at the first time instance, andwherein the first event comprises the measurement event for the handover.3.The terminal device of claim 1, wherein the predicted result comprises that the handover to a target cell is predicted to occur at the first time instance, andwherein the first event comprises that signal quality of the target cell is higher than or equal to a first threshold, or the first event comprises a measurement event for the handover.4.The terminal device of claim 1, wherein the second event comprises delaying a procedure of the handover to evaluate the first event for a first duration.5.The terminal device of claim 4, wherein the first duration comprises one of the following:a duration after reception of the command of the handover;a duration from the reception of the command of the handover to a second time instance associated with a requirement of the model monitoring; ora duration from the reception of the command of the handover to the first time instance.6.The terminal device of claim 4, wherein the terminal device is caused to evaluate the model inference by:in accordance with a determination that the first event occurs during the first duration, determining that the model inference is correct; andin accordance with a determination that the first event does not occur during the first duration, determining that the model inference is incorrect.7.The terminal device of claim 4, wherein the terminal device is further caused to:in accordance with a determination that the first event occurs during the first duration, apply the procedure of the handover; orin accordance with a determination that the first event does not occur during the first duration, apply the procedure of the handover; orin accordance with a determination that the first event does not occur during the first duration, transmit, to the network device, an indication for rejecting the command of the handover, and cancel the procedure of the handover.8.The terminal device of claim 1, wherein the second event comprises that evaluating, after applying a procedure of the handover, a radio link failure or a further handover for a second duration.9.The terminal device of claim 8, wherein the second duration comprises one of the following:a duration since reception of the command of the handover; ora duration since completion of the procedure of the handover.10.The terminal device of claim 8, wherein the terminal device is caused to evaluate the model inference by:in accordance with a determination that the radio link failure or the further handover does not occur during the second duration, determining that the model inference is correct; andin accordance with a determination that the radio link failure or the further handover occurs during the second duration, determining that the model inference is incorrect.11.The terminal device of claim 1, wherein the second event comprises that evaluating, after applying a procedure of the handover, the first event for a third duration by considering a source cell of the handover as a serving cell for the first event and a target cell of the handover as a neighboring cell for the first event, orwherein the second event comprises that signal quality of a neighboring cell after the handover is lower than or equal to signal quality of a serving cell after the handover.12.The terminal device of claim 11, wherein the third duration comprises one of the following:a duration since reception of the command of the handover; ora duration since completion of a procedure of the handover.13.The terminal device of claim 11, wherein the terminal device is caused to evaluate the model inference by:in accordance with a determination that the first event occurs during the third duration, determining that the model inference is correct; andin accordance with a determination that the first event does not occur during the third duration, determining that the model inference is incorrect.14.The terminal device of claim 1, wherein the terminal device is further caused to:in accordance with a determination that the first event occurs and the predicted result is absent, keep performing the model inference until earlier one of the following: reception of the command of the handover, or a time offset after a third time instance at which the first event occurs;in accordance with a determination that the predicted result is output until the reception of the command of the handover or the time offset, determine that the model inference is correct; andin accordance with a determination that the predicted result is absent until the reception of the command of the handover or the time offset, determine that the model inference is incorrect.15.A terminal device, comprising:a processor configured to cause the terminal device to:determine, based on a model inference, that a first time of stay in a target cell of a handover is predicted;store the first time of stay;determine a second time of stay in the target cell after the procedure of the handover; andevaluate the model inference based on a comparison between the first time of stay and the second time of stay.16.The terminal device of claim 15, wherein the terminal device is caused to determine the second time of stay by:in accordance with a determination that the terminal device camps on the target cell or the procedure of the handover is completed, starting a recording of the second time of stay; andstopping the recording based on at least one of the following:a radio link failure occurs during camping on the target cell;a beam failure detection occurs during camping on the target cell;a further handover occurs during camping on the target cell;the terminal device transits from a connected state to an idle or inactive state while a cell different from the target cell is selected during camping on the target cell; ora cell reselection occurs during camping on the target cell.17.The terminal device of claim 16, wherein the terminal device is further caused to:in accordance with a determination that the terminal device transits from a connected state to an idle or inactive state while the target cell is selected during camping on the target cell,keep the recording of the second time of stay; orcancel the evaluating of the model inference.18.The terminal device of claim 15, wherein the terminal device is caused to evaluate the model inference by:in accordance with a determination that a difference between the first time of stay and the second time of stay is smaller than or equal to a second threshold, determining that the model inference is correct; andin accordance with a determination that the difference between the first time of stay and the second time of stay is larger than the second threshold, determining that the model inference is incorrect.19.A terminal device, comprising:a processor configured to cause the terminal device to:transmit, to a network device, at least one of the following:first information of a prediction of a model; orsecond information indicating update of an input requirement of the model.20.The terminal device of claim 19, wherein the first information comprises at least one of the following:a type of the prediction; oran indication of whether a layer 3 (L3) filtering procedure is inside the model.21.The terminal device of claim 20, wherein the type indicates one of the following:an input of the model is a measured result in a layer 1 (L1) beam level and an output of the model is a predicted result in a L1 beam level;an input of the model is a measured result in a L1 or L3 cell level and an output of the model is a predicted result in a L3 cell level;an input of the model is a measured result in a L1 beam level and an output of the model is a predicted result in a L3 cell level; oran input of the model is a measured result in a L1 or L3 beam level and an output of the model is a predicted result in a L3 beam level.22.The terminal device of claim 20, wherein the terminal device is caused to transmit the first information by at least one of the following:transmitting the first information in a report for a monitoring of the model;transmitting the first information in a report for capability of the model;transmitting the first information in a report for applicability of the model; ortransmitting the first information in a report for activation of the model.23.The terminal device of claim 19, wherein the terminal device is caused to transmit the second information by:in accordance with a determination that the model is deactivated or updated, transmitting the second information to the network device; orin accordance with a determination that the input requirement of the model is updated, transmitting the second information to the network device.24.The terminal device of claim 23, wherein the terminal device is caused to transmit the second information by:transmitting the second information in a request for de-configuring or re-configuring a measurement resource related to the update of the input requirement.25.The terminal device of claim 24, wherein the request further comprises information of a duration in which the request is valid.26.A terminal device, comprising:a processor configured to cause the terminal device to:determine that a first set of outputs of a model is problematic;perform a first operation for handling the first set of outputs; andperform a model prediction with a second set of outputs of the model.27.The terminal device of claim 26, wherein the terminal device is caused to determine that the first set of outputs of the model is problematic by:determining a predicted result on an output in the first set of outputs based on a model inference;determining a measured result on the output by performing a real measurement on the output; andin accordance with a determination that a difference between the predicted result and the measured result is larger than or equal to a third threshold, determining that the output is problematic.28.The terminal device of claim 26, wherein the first operation comprises:deactivating the first set of outputs; andtransmitting information of the first set of outputs to a network device.29.The terminal device of claim 26, wherein the first operation comprises:transmitting information of the first set of outputs to a network device;receiving, from the network device, an indication of deactivating the first set of outputs; anddeactivating the first set of outputs based on the indication.30.The terminal device of claim 26, wherein the first operation comprises:disabling the first set of outputs;continuing to monitor the first set of outputs; andin accordance with a determination that there is no problem for an output in the first set of outputs, enabling the output.31.A method of communication, comprising:transmitting, at a terminal device, a predicted result of a model inference to a network device, the predicted result indicating that a handover is to occur at a first time instance; andin accordance with a determination that a command of the handover is received and a first event for a model monitoring does not occur, evaluating the model inference based on a second event for the model monitoring.32.A method of communication, comprising:determining, at a terminal device based on a model inference, that a first time of stay in a target cell of a handover is predicted;storing the first time of stay;determining a second time of stay in the target cell after the procedure of the handover; andevaluating the model inference based on a comparison between the first time of stay and the second time of stay.33.A method of communication, comprising:transmitting, at a terminal device and to a network device, at least one of the following:first information of a prediction of a model; andsecond information indicating update of an input requirement of the model.34.A method of communication, comprising:determining, at a terminal device, that a first set of outputs of a model is problematic;performing a first operation for handling the first set of outputs; andperforming a model prediction with a second set of outputs of the model.
Citation Information
Patent Citations
Cell handover method and apparatus, device, and storage medium
WO2022261834A1
Trajectory data collection in mobile telecommunication systems
WO2023052010A1
Method, device and computer storage medium of communication
WO2024026777A1
Apparatus and method for mobility with ai / ML channel prediction
WO2024096920A1