Communication method and device

Abstract

The present application relates to a communication method and device. The method comprises: when a trigger condition is satisfied, a first communication device executes a behavior corresponding to the trigger condition, wherein the trigger condition comprises a performance monitoring event of a model. By means of embodiments of the present application, the performance of the model can be efficiently monitored.

Description

Communication method and device TECHNICAL FIELD

[0001] The present application relates to the field of communication, and more particularly, to a communication method and device. BACKGROUND

[0002] In a communication system, a terminal can know the strength or quality of a radio signal of a current serving cell and surrounding neighbor cells by measurement, and then report these contents to a network in a measurement report. The network can make relevant decisions such as handover according to these contents. The measurement and reporting of performance indicators of a model can also refer to the measurement and reporting process of a radio signal.

[0003] SUMMARY

[0004] Embodiments of the present application provide a communication method and device, which can efficiently monitor the performance of a model.

[0005] Embodiments of the present application provide a communication method, comprising:

[0006] In a case where a trigger condition is met, a first communication device performs a behavior corresponding to the trigger condition, wherein the trigger condition comprises a performance monitoring event of a model.

[0007] Embodiments of the present application provide a communication method, comprising:

[0008] A second communication device sends configuration information, which is used to configure a first communication device to perform a behavior corresponding to a trigger condition in a case where the trigger condition is met, wherein the trigger condition comprises a performance monitoring event of a model.

[0009] Embodiments of the present application provide a first communication device, comprising:

[0010] A processing unit is configured to cause the first communication device to perform a behavior corresponding to a trigger condition in a case where the trigger condition is met, wherein the trigger condition comprises a performance monitoring event of a model.

[0011] Embodiments of the present application provide a second communication device, comprising:

[0012] A transceiver is configured to send configuration information, which is used to configure a first communication device to perform a behavior corresponding to a trigger condition in a case where the trigger condition is met, wherein the trigger condition comprises a performance monitoring event of a model.

[0013] An embodiment of the present application provides a communication device, comprising a transceiver, a processor and a memory. The memory is configured to store a computer program, the transceiver is configured to communicate with other devices, and the processor is configured to invoke and run the computer program stored in the memory, so that the communication device performs the communication method.

[0014] An embodiment of the present application provides a chip for implementing the communication method.

[0015] Specifically, the chip comprises a processor configured to invoke and run a computer program from a memory, so that a device installed with the chip performs the communication method.

[0016] An embodiment of the present application provides a computer readable storage medium configured to store a computer program, which causes a device to perform the communication method when the computer program is run by the device.

[0017] An embodiment of the present application provides a computer program product comprising computer program instructions, which causes a computer to perform the communication method.

[0018] An embodiment of the present application provides a computer program, which causes a computer to perform the communication method when the computer program is run by the computer.

[0019] In the embodiment of the present application, the performance monitoring event of the model included in the triggering condition can efficiently monitor the running performance of the communication device model or the management of the model, and save the signaling overhead of the wireless interface. BRIEF DESCRIPTION OF DRAWINGS

[0020] FIG. 1 is a schematic diagram of an application scenario according to an embodiment of the present application.

[0021] FIG. 2 is a schematic diagram of an application mode of the TTT timer.

[0022] FIG. 3 is a schematic diagram of a measurement model.

[0023] FIG. 4 is a schematic diagram of an LCM structure.

[0024] FIG. 5 is a schematic diagram of a counter.

[0025] FIG. 6 is a schematic flowchart of a communication method according to an embodiment of the present application.

[0026] FIG. 7 is a schematic flowchart of a communication method according to another embodiment of the present application.

[0027] FIG. 8 is a schematic flowchart of a communication method according to an embodiment of the present application.

[0028] FIG. 9 is a schematic flowchart of a communication method according to another embodiment of the present application.

[0029] FIG. 10 is a schematic block diagram of a first communication device according to an embodiment of the present application.

[0030] FIG. 11 is a schematic block diagram of a first communication device according to another embodiment of the present application.

[0031] FIG. 12 is a schematic block diagram of a second communication device according to an embodiment of the present application.

[0032] FIG. 13 is a schematic structural diagram of a communication device according to an embodiment of the present application.

[0033] FIG. 14 is a schematic block diagram of a chip according to an embodiment of the present application.

[0034] FIG. 15 is a schematic block diagram of a communication system according to an embodiment of the present application. DETAILED DESCRIPTION

[0035] The technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings in the embodiments of the present application.

[0036] The technical solutions in the embodiments of the present application can be applied to various communication systems, such as a Long Term Evolution (LTE) system, an Advanced long term evolution (LTE-A) system, a New Radio (NR) system, an evolved system of the NR system, an LTE-based access to unlicensed spectrum (LTE-U) system, an NR-based access to unlicensed spectrum (NR-U) system, a Non-Terrestrial Networks (NTN) system, a Universal Mobile Telecommunication System (UMTS), a Wireless Local Area Networks (WLAN), a Wireless Fidelity (WiFi), a 5th-Generation (5G) system, or other communication systems, etc.

[0037] Generally, a conventional communication system supports a limited number of connections and is easy to implement. However, with the development of communication technology, a mobile communication system will not only support conventional communication, but also support, for example, Device to Device (D2D) communication, Machine to Machine (M2M) communication, Machine Type Communication (MTC), Vehicle to Vehicle (V2V) communication, or Vehicle to everything (V2X) communication, and the like. Embodiments of the present application can also be applied to these communication systems.

[0038] In an embodiment, the communication system in the embodiments of the present application can be applied to a Carrier Aggregation (CA) scenario, can also be applied to a Dual Connectivity (DC) scenario, and can also be applied to a Standalone (SA) network deployment scenario.

[0039] In an embodiment, the communication system in the embodiments of the present application can be applied to an unlicensed spectrum, which can also be considered as a shared spectrum, or can be applied to a licensed spectrum, which can also be considered as a non-shared spectrum.

[0040] Embodiments of the present application describe various embodiments in combination with network devices and terminal devices, wherein the terminal device can also be referred to as User Equipment (UE), access terminal, subscriber unit, subscriber station, mobile station, mobile, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent, or user equipment, etc.

[0041] The terminal device can be a station (STATION, ST) in a WLAN, can be a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA) device, a handheld device with wireless communication function, a computing device, or other processing device connected to a wireless modem, a vehicle-mounted device, a wearable device, a terminal device in a next-generation communication system such as an NR network, or a terminal device in a future evolved Public Land Mobile Network (PLMN) network, etc.

[0042] In the embodiments of the present application, the terminal device can be deployed on land, including indoors or outdoors, handheld, wearable or vehicle-mounted; can also be deployed on the water surface (such as ships, etc.); and can also be deployed in the air (such as airplanes, balloons and satellites, etc.).

[0043] In the embodiments of the present application, the terminal device can be a mobile phone, a tablet computer, a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in self driving, a wireless terminal device in remote medical, a wireless terminal device in smart grid, a wireless terminal device in transportation safety, a wireless terminal device in smart city, or a wireless terminal device in smart home, etc.

[0044] By way of example and without limitation, the terminal device in the embodiments of the present application can also be a wearable device. The wearable device can also be referred to as a wearable smart device, which is a general term for devices that are designed and developed by applying wearable technology to daily wear, such as glasses, gloves, watches, clothing and shoes, etc. The wearable device is a portable device that can be directly worn on the body or integrated into the clothes or accessories of the user. The wearable device is not only a hardware device, but also a device that realizes powerful functions through software support and data interaction and cloud interaction. The general wearable smart device includes devices with complete functions, large size and complete or partial functions independent of smart phones, such as smart watches or smart glasses, etc., and devices that focus on a certain application function and need to cooperate with other devices such as smart phones, such as various smart wristbands and smart jewelry for monitoring vital signs.

[0045] In the embodiments of the present application, the network device can be a device for communicating with the mobile device. The network device can be an access point (AP) in a WLAN, an evolved node B (eNB or eNodeB) in LTE, or a relay station or an access point, or a vehicle-mounted device, a wearable device, and a network device in an NR network (gNB) or a future evolved PLMN network or a network device in an NTN network, etc.

[0046] By way of example and not limitation, in embodiments of the present application, a network device can have a mobile characteristic, for example, the network device can be a mobile device. Alternatively, the network device can be a satellite, a balloon station. For example, the satellite can be a low earth orbit (LEO) satellite, a medium earth orbit (MEO) satellite, a geostationary earth orbit (GEO) satellite, a high elliptical orbit (HEO) satellite, etc. Alternatively, the network device can also be a base station disposed at a location on land, water, etc.

[0047] In embodiments of the present application, a network device can serve a cell, and a terminal device communicates with the network device through a transmission resource (for example, a frequency domain resource, or a spectrum resource) used by the cell. The cell can be a cell corresponding to the network device (for example, a base station), and the cell can belong to a macro base station or a base station corresponding to a small cell. The small cell can include a metro cell, a micro cell, a pico cell, a femto cell, etc., and these small cells have the characteristics of small coverage and low transmit power, and are suitable for providing high-speed data transmission services.

[0048] FIG. 1 illustrates a communication system 100. The communication system includes one network device 110 and two terminal devices 120. In an implementation, the communication system 100 can include multiple network devices 110, and each network device 110 can include other numbers of terminal devices 120 within its coverage, which is not limited in embodiments of the present application.

[0049] In an implementation, the communication system 100 can further include a mobility management entity (MME), an access and mobility management function (AMF), and other network entities, which are not limited in embodiments of the present application.

[0050] The network device can include an access network device and a core network device. That is, the wireless communication system also includes a plurality of core networks for communicating with the access network device. The access network device can be an evolved node B (eNB or e-NodeB) macro base station, a micro base station (also referred to as a "small base station"), a pico base station, an access point (AP), a transmission point (TP), or a new generation Node B (gNodeB) in a long-term evolution (LTE) system, a next radio (NR) system, or an authorized auxiliary access long-term evolution (LAA-LTE) system.

[0051] It should be understood that the devices with communication functions in the network / system in the embodiments of the present application can be referred to as communication devices. For example, the communication system shown in FIG. 1 can include network devices and terminal devices with communication functions. The network devices and terminal devices can be specific devices in the embodiments of the present application, which will not be described here. The communication devices can also include other devices in the communication system, such as network controllers, mobile management entities, and other network entities. The embodiments of the present application do not limit the above.

[0052] It should be understood that the terms "system" and "network" are often used interchangeably in this document. The term "and / or" in this document is only used to describe the association relationship between the associated objects. For example, A and / or B can represent three cases: A alone, A and B together, and B alone. In addition, the character " / " generally represents an "or" relationship between the associated objects.

[0053] It should be understood that the "indication" mentioned in the embodiments of the present application can be direct indication, indirect indication, or an indication with an associated relationship. For example, A indicates B, which can mean that B can be obtained through A; or A indirectly indicates B, for example, A indicates C, and B can be obtained through C; or A and B have an associated relationship.

[0054] In the description of the embodiments of the present application, the term "corresponding" can represent a direct or indirect corresponding relationship between the two, or an associated relationship between the two, or an indication and being indicated, configuration and being configured, and the like.

[0055] For the convenience of understanding the technical solutions of the embodiments of the present application, the related technologies of the embodiments of the present application are described as follows, and the following related technologies can be combined with the technical solutions of the embodiments of the present application in any manner, which all belong to the protection scope of the embodiments of the present application.

[0056] I. RRM measurement

[0057] In the cellular communication system of 3GPP (3rd Generation Partnership Project), the handover of the Radio Resource Control (RRC) connection between different cells (i.e. mobility management) is the core process in the standard specification of the control plane. In order to enable the handover between cells, the UE needs to report a measurement report.

[0058] In the second generation communication system, the measurement report is always reported in a certain period. From the third generation communication system, including, for example, Code Division Multiple Access (CDMA), the fourth generation communication system LTE, the fifth generation communication system NR, the measurement report is distinguished from the reporting mode, which can be mainly divided into three kinds:

[0059] 1. Periodic reporting;

[0060] 2. Reporting based on measurement events;

[0061] 3. Reporting based on measurement events and continuing periodic reporting thereafter.

[0062] No matter which form of reporting of the measurement report, the specific measurement event and / or measurement result, such as the signal strength or signal quality of the cell, can be included in the measurement report; wherein the signal strength is, for example, the Reference Signal Receiving Power (RSRP) in dbm (decibel milliwatt) as the dimension; and the signal quality is, for example, the Reference Signal Receiving Quality (RSRQ) in db (decibel) as the dimension. The reported cell includes the current serving cell and the neighboring cell. The object of measurement can be the same frequency, different frequency or different communication system frequency.

[0063] The triggering of a measurement event can include the following basic elements:

[0064] 1. Measurement result, including the measurement result of the serving cell and / or the neighboring cell, such as the signal strength of the cell.

[0065] 2. Comparison parameters, such as threshold, hysteresis, offset, etc. The dimension of the measurement result is in the standard protocol, the larger the value, the higher the intensity or quality of the signal. Absolute comparison means comparing the measurement value of a cell with a threshold, in which case the measurement result greater than the sum of the threshold and the hysteresis value indicates that the entering condition is met, and the measurement result less than the difference between the threshold and the hysteresis value indicates that the leaving condition is met. Relative comparison usually means comparing the measurement result of a neighbor cell with the measurement result of a serving cell. Before comparison, each cell needs to add its own relevant offset value. For the serving cell, an offset value (Off_event) related to the corresponding event is also added. Finally, when comparing, the hysteresis value (Hys) also needs to be considered. Taking the A3 event as an example, the measurement result, offset value, etc. of the serving cell are marked with s, and the measurement result, offset value, etc. of the neighbor cell are marked with n, then the entering condition of the A3 event can be expressed as:

[0066] Mn+Ofn>Ms+Ofs+Hys+Off_event;

[0067] The leaving condition of the A3 event can be expressed as:

[0068] Mn+Ofn<Ms+Ofs-Hys+Off_event;

[0069] Wherein, Mn represents the measurement result of the neighbor cell, Ofn represents the offset value related to the neighbor cell; Ms represents the measurement result of the serving cell, Ofs represents the offset value related to the serving cell, Hys represents the hysteresis value, and Off_event represents the offset value related to the event.

[0070] 3. Timer indicating the robustness of the measurement result, i.e. the time to trigger (TTT) timer. Fig. 2 is a schematic diagram of the application mode of the TTT timer. When a cell meets the entering condition of an event at T0, the TTT timer will start. When the TTT timer expires, if the cell has always met the entering condition of the event, it means that the cell triggers the measurement event.

[0071] The 3GPP specification protocol describes how the UE performs the intra-frequency or inter-frequency measurement process, and how to measure the sampling according to the beam at Layer 1 (L1) and how to make the measurement event decision according to the network configured parameters. Referring to the schematic diagram of the measurement model in Fig. 3, the process from the measurement result of a single beam to the triggering of the measurement event can be embodied. Among them, the beam selection / combination, the Layer 1 filtering of the beam measurement result, the Layer 3 (L3) filtering of the cell and / or beam measurement result, and the measurement event decision process (measurement reporting evaluation), Layer 3 beam filtering, beam selection reporting, etc. can be controlled by the RRC parameters configured by the network.

[0072] The meanings of several reference points in Figure 3 are as follows:

[0073] A: is the point where UE performs physical layer measurement sampling, in the granularity of beams

[0074] A 1 : UE performs L1 filtering on the measured beam measurement results. In general, the protocol specifies the length of the measurement period under specific RRC configuration. The measurement period specifies that UE should perform at least one sampling, and the beam measurement results after L1 filtering should meet the performance requirements of 3GPP specifications. The specific sampling times of UE within a measurement period at reference point A are specified. In test cases, oversampling of 4-5 is generally used.

[0075] B: A 1 : the obtained beam measurement results of a certain cell are combined to synthesize L1 cell-level measurement results.

[0076] C: L1 cell-level measurement results of a certain cell, in chronological order, through L3 filtering to obtain L3 cell-level measurement results.

[0077] C 1 : the results of the same reference point C obtained by another cell.

[0078] D: the measurement results of the serving cell and / or neighboring cells are used to determine whether a specific measurement event is true according to certain decision conditions (configured by the network). For example, whether the measurement result of the neighboring cell is higher than the measurement result of the PCell of the cell by an offset value (A3 event), etc.

[0079] In addition to the measurement event, the UE can also evaluate whether a radio link failure will occur according to the measurement results of the current serving cell. If the UE upper layer protocol receives N310 consecutive Qout indications from the physical layer, the UE will start a T310 timer, and when the T310 times out, it is considered that a radio link failure (RLF) event has been triggered. The Qout indication is triggered when the reference signal measured at the physical layer is lower than a certain threshold. When T310 is running, if the UE upper layer protocol receives N311 consecutive Qin indications, T310 will stop. The Qin indication is triggered when the reference signal measured at the physical layer is higher than a certain threshold, indicating that the radio link has been restored. In addition to RLF, the UE can also determine whether a handover failure or ping-pong handover event has occurred.

[0080] II. AI / ML research in 3GPP

[0081] 3GPP studied in Release 18 (R18) whether AI / ML algorithm models can be applied in key technologies in the physical layer, including whether channel feedback information (CSI) of the wireless interface can be compressed and decompressed, whether the best beam or beam pair in the spatial or time domain can be predicted, and whether positioning can be predicted.

[0082] For example, in a 3GPP Technical Report (TR), it is described how to use AI / ML algorithms to predict beam measurements for beam management purposes. One of the sub-use cases is a spatial domain prediction, that is, by measuring a subset of beams, using the spatial correlation between beams, to predict the best beam (the strongest wireless signal) or a pair of beams (downlink transmission and reception) in the full set of beams. Another use case is a time domain beam prediction, that is, based on the measurement results of beams (and beams that have been predicted) measured in the past, using the correlation of beams over time, to predict the measurement results of beams in the current time slot. According to the recorded evaluation results, it can be proved that these two use cases are not only technically feasible, but also can provide high performance improvement.

[0083] 3GPP also studied the application of beam measurement prediction techniques to RRM measurements. The main purpose of RRM measurements is for mobility management processes in cellular systems. In addition to predicting RRM measurement results, the model can also be used to determine whether certain mobility events, such as measurement events, will occur and when they will occur.

[0084] In addition to recording the core evaluation methods and results, the TR also records how to manage the AI / ML models on the network side or the UE side, or both sides. These contents are referred to as life cycle management (LCM) in the TR. LCM is a relatively broad term, and its contents include data collection, model training, function / model identification, model transmission, model inference, selection, activation, deactivation, replacement, and fallback of functions / models, function / model monitoring, model updating, UE capability reporting, and other contents. These processes can be seen in the schematic diagram shown in FIG. 4.

[0085] In these LCM procedures, the function / model monitoring procedure usually refers to the network monitoring the AI / ML model running on the UE side, and deciding further management steps, such as deactivating the model or replacing the model, according to the running indicators reported by the UE. In the model monitoring content related to beam management, it can be seen that the indicators reflecting the performance of the model running can be calculated on the UE or network side. If it is on the UE, it means that the UE needs to calculate these performance indicators and then report them to the network. If it is on the network side, the relevant auxiliary information needs to be reported to the network, and then the network calculates the performance indicators according to the auxiliary information.

[0086] The use cases in the study of RRM measurement in 3GPP RAN2 can include:

[0087] 1. RRM measurement prediction;

[0088] 2. Measurement event prediction;

[0089] 3. RLF / HOF event prediction.

[0090] The RRM measurement prediction use case contains three sub-use cases, as shown in Table 1:

[0091] Table 1

[0092] These sub-use cases use the same performance indicator, that is, the average absolute error value between the predicted L3 RSRP and the true L3 RSRP as the prediction accuracy. The so-called true L3 RSRP refers to the measurement result of the L3 RSRP measured by the UE without AI / ML model and the predicted L3 RSRP. If the purpose of predicting the measurement result is to reduce the measurement in a certain domain (time domain, spatial domain or frequency domain), the prediction accuracy is directly related to the proportion of reduced measurement. If the purpose of prediction is to know the future measurement result in advance, the prediction window length and the prediction accuracy are directly related.

[0093] For the prediction of measurement events or RLF / HOF events, 3GPP is still discussing whether to adopt the calculation method of F1 score. F1 score is a general indicator for evaluating the model whose output result is the event decision result.

[0094] In addition to predicting whether a mobility event will occur, the AI / ML model is used to predict when the mobility event will occur. The UE can also determine whether the predicted event occurs by reporting measurements. Since the prediction of mobility events and the measurement behavior itself are ongoing, it is necessary to determine whether a predicted mobility event occurs by looking at the proximity of the predicted mobility event and the mobility event determined based on the measurement report in the time axis. This is because the time of the model prediction and the time of the determination based on the measurement report will generally have a certain gap. Assuming that the maximum gap in time between the two is defined as a parameter MAX_TIME_GAP, the conditions for counting the counters n1, n2, and n3 are as follows:

[0095] The counting condition of the counter n1: A mobility event that occurs in a certain cell predicted by the model does not occur in the cell within a maximum of MAX_TIME_GAP before or after the time point of the occurrence of the mobility event.

[0096] The counting condition of the counter n2: A mobility event that occurs in a certain cell predicted by the model does not occur in the cell within a maximum of MAX_TIME_GAP before or after the time point of the occurrence of the mobility event.

[0097] The counting condition of the counter n3: A mobility event that occurs in a certain cell is not predicted by the model within a maximum of MAX_TIME_GAP.

[0098] FIG. 5 is a case of the counter n2, in which the model can predict a mobility event in advance, and the time point (middle star) can be before or after the time point of the measured event (right star).

[0099] The above timers can be substituted into the following table:

[0100] Table 2: Mobility event counter assignment

[0101] Performance indicators related to mobility events can include the following examples:

[0102] Model prediction accuracy = (n2) / (n1+n2), model false detection rate = 1 - model prediction accuracy

[0103] Model prediction completeness = (n2) / (n3+n2), model missed detection rate = 1 - model prediction completeness

[0104] Model comprehensive performance index F1 = 2 * (1 / (1 / model prediction accuracy + 1 / model prediction integrity))

[0105] Therefore, the value of the F1 score is directly related to the counter result, and the counter result is directly related to the time gap.

[0106] When the model is running, it is important to monitor the performance of the model. This is because the application of AI technology in communication technology is not yet mature, and is currently still limited to the standardization stage, and is rarely seen in actual engineering sites. Monitoring the UE-side model can help the network understand the running status of the UE-side model, so as to make further decisions based on the received information.

[0107] The performance monitoring of the AI / ML model can be regarded as "measuring behavior". Unlike measuring the strength or quality of the wireless signal, the UE measures the performance of the AI / ML model. Based on this understanding, the measurement and reporting of the performance index of the model can also refer to the measurement and reporting process of the wireless signal mentioned above.

[0108] FIG. 6 is a schematic flowchart of a communication method according to an embodiment of the present application. The method can optionally be applied to the system shown in FIG. 1, but is not limited thereto. The method includes at least part of the following content.

[0109] S610, in the case where the trigger condition is satisfied, the first communication device performs a behavior corresponding to the trigger condition, wherein the trigger condition includes a performance monitoring event of the model.

[0110] In the embodiments of the present application, the model can be understood as an AI / ML model, or as an AI / ML function. The first communication device can receive information related to the trigger condition from the second communication device. The trigger condition can also be preset. In the case where the AI / ML model or the AI / ML function meets a certain trigger condition, the first communication device can perform a behavior corresponding to the trigger condition. For example, the first communication device can be a UE, and the second communication device can be a network device (which can be referred to as a network). When managing the AI / ML model or the AI / ML function, in order to reduce the signaling overhead on the wireless interface, the network configures a trigger condition to the UE, for example, a trigger condition, for triggering a behavior associated with the trigger condition, such as activating, deactivating, replacing, or withdrawing the model or function; or for triggering the reporting of the performance index of the model or function.

[0111] In the embodiments of the present application, the trigger condition can include a performance monitoring event. Alternatively, the trigger condition can be used to define the performance monitoring event. The performance indicators of the model or function can include various performance indicators, such as measurement error, prediction accuracy, and the like. According to different performance indicators, one or more performance monitoring events can be set. The performance monitoring event and the behavior of the first communication device can have a corresponding relationship. The first communication device can directly perform the behavior corresponding to the trigger condition when the trigger condition is met. The first communication device can also report event information, performance indicators of the model, and the like to the second communication device when the trigger condition is met, and the second communication device can instruct the first communication device to perform the behavior corresponding to the trigger condition.

[0112] In the embodiments of the present application, the performance monitoring event of the model included in the trigger condition can efficiently monitor the running performance of the communication device model or the management of the model, and save the signaling overhead of the wireless interface.

[0113] In an implementation, the behavior corresponding to the trigger condition includes the behavior corresponding to the performance monitoring event. If there are multiple performance monitoring events, the behaviors corresponding to different performance monitoring events can be different or the same. For example, the behavior corresponding to the event for monitoring the performance deterioration of the model is to deactivate the model. For another example, the behavior corresponding to the event for monitoring the performance improvement of the model is to reactivate the model. For another example, the behavior corresponding to the event for monitoring the excessive measurement error of the model and the event for monitoring the too small prediction accuracy of the model is to deactivate the model.

[0114] In an implementation, the performance of the model includes a measurement error between a predicted measurement result and an actual measurement result, and the performance monitoring event includes one or more of the following:

[0115] The measurement error is greater than or equal to a first threshold value;

[0116] The measurement error is less than or equal to a second threshold value;

[0117] The measurement error of the first model is less than or equal to the measurement error of the second model minus a first offset value;

[0118] The measurement error of the first model is less than a third threshold value, and the measurement error of the second model is greater than a fourth threshold value.

[0119] In the embodiments of the present application, if the model or function is used to predict a measurement result, the performance indicator of the model or function can be a measurement error (or measurement result difference), for example, an average absolute value or a root mean square difference between the predicted measurement result and the actual measurement result. The greater the measurement error, the worse the performance of the model or function; otherwise, the smaller.

[0120] For example, the event for monitoring whether the performance of the model is deteriorating can be that the measurement difference is greater than or equal to a first threshold value (EVENT P1).

[0121] For example, the event for monitoring whether the performance of the model is deteriorating can be that the measurement difference is greater than or equal to a first threshold value (EVENT P1).

[0122] For example, the event for monitoring whether the performance of the model is deteriorating can be that the measurement difference is greater than or equal to a first threshold value (EVENT P1).

[0123] The threshold value or the measurement difference in the above examples can be related to an additional condition. If the model or function is based on historical measurements to predict measurements in a future prediction window, the additional condition can be the length or range of the observation window and / or the prediction window. If the model or function is based on time-domain (or spatial-domain) partial measured measurements to infer measurements, the additional condition can be the percentage of reduction of measurement load. If the model or function is used for measurement prediction between frequencies, the additional condition can be the frequency difference between the frequencies or two specific frequency parameters.

[0124] The measurement in the above examples can be L1 or L3 measurement, which can be a beam or cell level measurement. The measured measurement in the above examples can be a measurement obtained by actual measurement without adopting an AI / ML algorithm. The real measurement can be a measured measurement of a cell to be predicted.

[0125] In an embodiment, the performance of the model includes a prediction accuracy based on one or more determinations of a number of predicted events, a number of actually occurred events, a relationship between the predicted events and the actually occurred events, and the performance monitoring event includes one or more of:

[0126] the prediction accuracy is greater than or equal to a fifth threshold value;

[0127] the prediction accuracy is less than or equal to a sixth threshold value;

[0128] the prediction accuracy of the first model is greater than or equal to the prediction accuracy of the second model plus a second offset value;

[0129] the prediction accuracy of the first model is greater than a seventh threshold value and the prediction accuracy of the second model is less than an eighth threshold value.

[0130] In embodiments of the present application, if the model or function is used to predict whether something will happen, such as a mobility event (a measurement event or RLF or HOF or too short stay event or ping-pong handover event, etc.), the performance indicator of the model or function can be the prediction accuracy (or prediction precision), for example, as follows:

[0131] Indicator 1: N2 / (N1+N2);

[0132] Indicator 2: 2 / (N3+N2);

[0133] Indicator 3: 2*Indicator 1*Indicator 2 / (Indicator 1+Indicator 2)

[0134] Where N1, N2 and N3 represent the following counters, respectively:

[0135] N1: the number of times that the predicted event does not actually occur;

[0136] N3: the actual occurrence of the event that is not predicted;

[0137] N2: the predicted event actually occurs (or the actual occurrence of the event is predicted).

[0138] For example, the event used to monitor whether the performance of the model becomes worse can be that the prediction accuracy is less than or equal to a fifth threshold value (EVENT Q1).

[0139] For another example, the event used to monitor whether the performance of the model becomes better can be that the prediction accuracy is greater than or equal to a sixth threshold value (EVENT Q2).

[0140] For another example, the event used to indicate whether the performance of a model (A) is higher than that of another model (B) can be that the prediction accuracy of model A is greater than or equal to the prediction accuracy of model B plus a second offset value (EVENT Q3); or, the prediction accuracy of model A is greater than or equal to a seventh threshold value, and the prediction accuracy of the model is less than or equal to an eighth threshold value (EVENT Q4).

[0141] The above-mentioned threshold value or prediction accuracy can be associated with an additional condition. This additional condition can be the time advance of the model predicting the mobility event and / or the maximum time span of the allowed interval between the predicted event and the actual occurrence event when defining whether the predicted event actually occurs.

[0142] In an embodiment, the first model and the second model are two models with the same function but applicable to different conditions or configurations. For example, model A can be an example of the first model, and model B can be an example of the second model.

[0143] In the embodiments of the present application, the performance monitoring event and the management action (behavior) of the first communication device can have an association relationship, thereby realizing the conditional control manner. For example, the following examples:

[0144] Table 3

[0145] For example, E1 in the above table can be P1 or Q1 in the above examples, E2 can be P2 or Q2 in the above examples, E3 can be P3 or Q3 in the above examples, and E4 can be P4 or Q4 in the above examples.

[0146] In an embodiment, the trigger condition includes an event entry condition and a time condition, and the event entry condition has a corresponding relationship with the performance monitoring event.

[0147] In an embodiment, if the trigger condition includes the performance monitoring event, the trigger condition is satisfied, including: in the case where the performance monitoring event occurs, the trigger condition is satisfied.

[0148] In one case, the examples of the performance monitoring event described above can be used as examples of the trigger condition. For example, if the measurement error is greater than or equal to a threshold value, the trigger condition is satisfied, and the behavior corresponding to the performance monitoring event can be executed. For example, if it is monitored that the measurement result difference of the model is greater than or equal to a first threshold value, it can be determined that EVENT P1 occurs, that is, the trigger condition is satisfied, and the UE can execute the behavior corresponding to EVENT P1. For another example, if it is monitored that the prediction accuracy of the model is greater than or equal to a sixth threshold value, it can be determined that EVENT Q2 occurs, that is, the trigger condition is satisfied, and the UE can execute the behavior corresponding to EVENT Q2.

[0149] In an embodiment, if the trigger condition includes the performance monitoring event and the time condition, and the time condition includes a timer timeout and / or a counter reaching a maximum number, the trigger condition is satisfied, including: in the case where the performance monitoring event occurs, the first communication device starts the timer and / or the counter; in the case where the timer times out and / or the counter reaches the maximum number, the trigger condition is satisfied.

[0150] In another case, the above examples of performance monitoring events can be examples of event entry conditions. After the event entry condition is satisfied, for example, the prediction accuracy is greater than or equal to the fifth threshold, the performance monitoring event is not directly triggered, but is delayed to be triggered by a timer and / or a counter. After the event entry condition is satisfied, the timer and / or the counter can be started first, and a time condition is continued to be judged. During the timer timing and / or the counter counting, the performance of the model can be continuously monitored and it is judged whether the event entry condition is satisfied. If the time condition is satisfied, the timer is timed out and / or the counter reaches the maximum number of times, the behavior corresponding to the performance monitoring event is executed. For example, if the measured result difference of the model is greater than or equal to the first threshold, it is determined that EVENT P1 occurs. At this time, the behavior corresponding to EVENT P1 is not executed first, but the timer corresponding to EVENT P1 is started first. During the timer timing, the performance of the model can be continuously monitored and it is judged whether EVENT P1 continues to occur. If yes, the behavior corresponding to EVENT P1 is executed again. For another example, if the prediction accuracy of the model is greater than or equal to the sixth threshold, it is determined that EVENT Q2 occurs, that is, the trigger condition is satisfied. The UE can start the timer corresponding to EVENT Q2 first. During the timer timing, the performance of the model can be continuously monitored and it is judged whether EVENT Q2 continues to occur. If yes, the behavior corresponding to EVENT Q2 is executed again.

[0151] In an embodiment, the behavior corresponding to the performance monitoring event includes one or more of the following: deactivating the model; activating the model that has been deactivated; replacing the second model with the first model.

[0152] In the embodiments of the present application, different performance monitoring events can have corresponding behaviors of the first communication device. After a certain performance monitoring event occurs, or after a certain performance monitoring event continues to occur for a period of time, the first communication device can execute the behavior corresponding to the performance monitoring event. For example, the behavior corresponding to EVENT P1 or EVENT Q1 is to deactivate the model. The behavior corresponding to EVENT P2 or EVENT Q2 is to activate the model that has been deactivated. The behavior corresponding to EVENT P1 or EVENT Q is to replace the model B with the model A.

[0153] In an embodiment, in the case where the trigger condition is satisfied, the first communication device executes the behavior corresponding to the trigger condition, including:

[0154] In the case where the trigger condition is satisfied, the first communication device reports event information and / or model performance information;

[0155] The first communication device receives indication information, the indication information being used to instruct the first communication device to perform a behavior corresponding to the trigger condition.

[0156] In the embodiments of the present application, the second communication device can instruct the first communication device to perform a subsequent behavior in the case where the trigger condition of the first communication device is met. For example, if the UE monitors that the performance of Model A meets EVENT P1 and the timer corresponding to the EVENT P1 is expired, the UE can report the EVENT P1 information and the measurement result difference of Model A to the network device, which is greater than or equal to a first threshold. The network device further makes a decision according to the received information. If the network device decides to deactivate Model A, the network device can send indication information to the UE to deactivate Model A. After receiving the indication information, the UE deactivates Model A.

[0157] FIG. 7 is a schematic flowchart of a communication method according to another embodiment of the present application. The method can include one or more features of the above-mentioned methods. In an implementation, the method further includes:

[0158] S710, the first communication device receives configuration information, the configuration information being used to configure a trigger condition.

[0159] In the embodiments of the present application, after the first communication device receives the configuration information from the second communication device, the first communication device can use the control parameters in the configuration information to control the running of the model and / or make a decision on the trigger condition.

[0160] In an implementation, the configuration information includes an event control parameter, the event control parameter being used to determine whether the trigger condition is met.

[0161] In an implementation, the configuration information includes an additional condition parameter, the additional condition parameter being used to control the running of the model.

[0162] In an implementation, the method further includes: when the first communication device performs the performance monitoring of the model, the first communication device uses the additional condition parameter as a control parameter of the model running, and makes a decision on the trigger condition according to the event control parameter configured by the second communication device; and reports event information and / or model performance information after the event trigger is met. For example, if the trigger condition is met due to the occurrence of EVENT P1, the information of EVENT P1 and the measurement error of the model can be reported.

[0163] In the embodiments of the present application, an example of a possible flow of a wireless interface is as follows:

[0164] Method 1: The additional condition related parameter directly affects the calculation of the performance index.

[0165] Step 1: The network configures the control parameters of the triggering condition to the UE, which are used to control and trigger the behavior of the UE or report the performance index of the management model associated with the condition. In addition to the parameters defined by the event (which can be referred to as event control parameters or event parameters), the additional condition parameters are also included in the control parameters.

[0166] Step 2: When the UE performs performance monitoring of the model, the additional condition parameters are used as control parameters for the model running. After obtaining the network performance index, the UE determines whether the triggering condition has been met according to the event parameters configured by the network, such as threshold or offset value. If it is met, the event corresponding to the triggering condition can be triggered (which can also be understood as the event occurring). After the event is triggered, the UE can report the event information and model performance information.

[0167] In an embodiment, the configuration information includes a plurality of additional condition parameters corresponding to a plurality of event control parameters. The event control parameters and the additional condition parameters can be one-to-one correspondence. For example, threshold 1 corresponds to a 40% reduction in measurement load. A group of event control parameters can correspond to a group of condition parameters. The event control parameters include threshold 2 and counter 1, and the event control parameters correspond to, for example, threshold 1 corresponds to a 40% reduction in measurement load and the size of the prediction window t1.

[0168] In an embodiment, the method further includes: the first communication device selects an additional condition parameter as a control parameter for the model running when performing performance monitoring of the model, and determines the triggering condition according to the event control parameter corresponding to the used additional condition parameter; and reports one or more of the event information, the model performance information, and the used additional condition parameter after the event trigger is met.

[0169] In the embodiments of the present application, an example of a possible flow of a wireless interface is as follows:

[0170] Method 2: Allow the UE to select between different and additional condition related parameters. When the network configures the control parameters of the triggering condition, the event control parameters configured and the additional condition parameters configured are one-to-one correspondence.

[0171] Step 1: The network configures the control parameters of the triggering condition to the UE, which are used to control and trigger the behavior of the UE or report the performance index of the management model associated with the condition. In addition to the parameters defined by the event (which can be referred to as event control parameters or event parameters), the additional condition parameters are also included in the control parameters.

[0172] Step 2: UE selects one of the additional condition parameters to run the model when it performs the performance monitoring of the model. After obtaining the performance index of the model, UE can use the event control parameter corresponding to the selected additional condition parameter, such as threshold or offset, to determine whether the triggering condition is met. After the event triggering condition is met, UE can report the event information and / or the model performance information. UE can also report the selected additional condition parameter.

[0173] In an embodiment, the configuration information includes a plurality of optional additional condition parameters.

[0174] In an embodiment, the method further includes: the first communication device selects one of the additional condition parameters as a control parameter for running the model when it performs the performance monitoring of the model, and determines the triggering condition according to the event control parameter configured by the second communication device; and reports one or more of the event information, the model performance information, and the used additional condition parameter after the event triggering condition is met.

[0175] In an embodiment of the present application, an example of a possible flow of a wireless interface is as follows:

[0176] Method 3: UE is allowed to select between different and additional condition related parameters, but the network only configures one set of event control parameters for triggering conditions.

[0177] Step 1: The network configures control parameters for triggering conditions to UE, which are used to control and trigger the behavior of the management model associated with the event or the behavior of reporting the performance index. The control parameters can include event control parameters such as threshold or offset, and a plurality of additional condition parameters.

[0178] Step 2: UE can select one of the additional condition parameters to run the model when it performs the performance monitoring of the model. After obtaining the performance index of the model, UE can determine whether the triggering condition is met based on the configured event triggering parameter. After the event triggering condition is met, UE can report one or more of the event information, the model performance index information, and the selected additional condition parameter. In addition, UE can also report cost information such as model size information, calculation times information, energy consumption information, etc.

[0179] In an embodiment, the event control parameter includes one or more of the following: threshold, offset, counter, timer.

[0180] In an embodiment of the present application, the threshold and offset used by different triggering conditions or performance detection events can be different.

[0181] For example, the threshold value used in the trigger condition of EVENT E1 is a first threshold value. The threshold value used in the trigger condition of EVENT E2 is a second threshold value. The offset value used in the trigger condition of EVENT E3 is a first offset value. The threshold value used in the trigger condition of EVENT E4 includes a third threshold value and a fourth threshold value.

[0182] For example, the threshold value used in the trigger condition of EVENT Q1 is a fifth threshold value. The threshold value used in the trigger condition of EVENT Q2 is a sixth threshold value. The offset value used in the trigger condition of EVENT Q3 is a second offset value. The threshold value used in the trigger condition of EVENT Q4 includes a seventh threshold value and an eighth threshold value.

[0183] The first threshold value to the eighth threshold value in the above examples can be different or partially the same. The first offset value and the second offset value can be the same or different.

[0184] In an embodiment, the additional condition parameter includes one or more of the following: a measured load reduction percentage, a size of a prediction window and / or a size of an observation window, a frequency difference, a time difference of event occurrence.

[0185] For example, the control parameter of an event P1 includes a first threshold value and a maximum count number of a counter, and the additional condition parameter includes a measured load reduction percentage.

[0186] For example, the control parameter of an event P2 includes a second threshold value and a time length of a timer, and the additional condition parameter corresponding to the event control parameter includes a size of a prediction window and a size of an observation window.

[0187] For example, the control parameter of an event Q1 includes three fourth threshold values Thr1, Thr2 and Thr3 and time lengths T1, T2 and T3 of timers respectively associated with the three fourth threshold values. One event control parameter corresponds to one additional condition parameter. The additional condition parameter corresponding to Thr1 and the time length T1 of the timer associated with Thr1 is a frequency difference FD1. The additional condition parameter corresponding to Thr2 and the time length T2 of the timer associated with Thr2 is a frequency difference FD2. The additional condition parameter corresponding to Thr3 and the time length T3 of the timer associated with Thr3 is a frequency difference FD3.

[0188] For example, the control parameter of an event Q2 includes a sixth threshold value Thr6. The multiple optional additional condition parameters include multiple optional time differences of event occurrence Td1, Td2 and Td3.

[0189] The above examples of event control parameters and additional condition parameters, and the specific parameter names in the examples of the corresponding relationship between event control parameters and additional condition parameters, are only examples and not limitations. In actual application scenarios, they can be flexibly adjusted according to needs.

[0190] In embodiments of the present application, the measurement load reduction percentage can be used to reduce the measurement load.

[0191] In embodiments of the present application, the purpose of the model prediction is to improve the performance of mobility. In order to improve the prediction performance, the historical measurements within the observation window do not reduce the measurement load in time domain or spatial domain, and are used to predict the L1 or L3 measurements within the prediction window (future).

[0192] In embodiments of the present application, the model or function can predict the L1 or L3 measurements between different frequencies. The measurements can be beam level or cell level measurements. The performance of the prediction result is related to the frequency difference between the frequencies.

[0193] In an implementation, the measurement load reduction percentage is used to determine the measurement opportunities skipped in time domain or the measurement beams skipped in frequency domain.

[0194] In an implementation, the size of the prediction window is used to determine the length of time for which the model is run to make predictions, and the size of the observation window is used to determine the length of time for which the historical measurements are used.

[0195] In an implementation, the frequency difference is used to determine the frequency difference between the measurement objects in frequency domain.

[0196] In an implementation, the event occurrence time difference is used to determine the difference between the occurrence time of the predicted event and the occurrence time of the actual event.

[0197] In an implementation, the method further comprises: the first communication device reporting cost information after triggering the performance monitoring event, the cost information including one or more of the following: model size information, calculation number information, energy consumption information.

[0198] In embodiments of the present application, the prediction made by the model or function can be time domain prediction. For the examples of the above-mentioned method 1 to method 3, the following adjustments can be made based on the measurement load reduction percentage.

[0199] An example of the steps of method 1 is as follows:

[0200] Step 1: When configuring the model or function, the second communication device, for example, the network, can configure an allowed measurement load reduction percentage for the first communication device, for example, the UE. In the same message, a performance monitoring event (or trigger event) such as P1 can also be configured, and it is specified that when the trigger condition (or event trigger event) of the performance monitoring event is met, the first communication device needs to report the obtained measurement error. The threshold used by the trigger condition and the control parameters for how to perform performance monitoring, such as the period and / or maximum number of periodic monitoring, etc. can also be configured in this message.

[0201] Step 2: The first communication device runs the model according to the allowed measurement load reduction percentage, and performs measurement on the measurement opportunities in the time domain and the measurement beams in the frequency domain. When the model is running, the first communication device can monitor the model according to the parameters configured by the second communication device. If the performance monitoring event, for example, P1, is triggered, it indicates that the performance of the model has dropped to a certain extent, and the first communication device can report the triggered event information, for example, P1, to the network, and can report the latest performance index result in the same message.

[0202] The second communication device can further manage the model, such as activating or deactivating, switching between models, or falling back to an algorithm without model running, by using the received trigger event information and the latest performance index. The second communication device can also directly associate the trigger event and the management action, so as to realize the conditional control mode. For examples of the association between the event type and the possible behavior of the first communication device, refer to Table 1 or Table 2 described above.

[0203] An example of the steps of method 2 is as follows:

[0204] Step 1: When configuring the model or function, the second communication device can configure a range or level of allowed measurement load reduction percentages for the first communication device, such as {40%, 30%, 20%, 10%}. In the same message, a control parameter of a performance monitoring event, for example, P3, can also be configured. In this message, offset values corresponding to the measurement load reduction percentages and how to perform performance monitoring control parameters can also be configured.

[0205] Step 2: The first communication device runs the model according to the allowed measurement load reduction percentage configured by the second communication device, for example, selects a percentage of 30% from multiple measurement load reduction percentages. The first communication device can skip 30% of the measurement opportunities in the time domain or randomly select 70% of the beams to perform measurement. When the model is running, the first communication device can monitor the model according to the parameters corresponding to the selected percentage of 30% configured by the second communication device. If P3 is triggered, it indicates that the performance of a model in the deactivation state is higher than that of another running model.

[0206] Step 3: The first communication device reports the P3 event to the second communication device, and can include the latest performance index of the two models (model A and model B) involved in the P3 event and / or the selected measurement load reduction percentage in the same message. When receiving the message, the second communication device can decide on further actions, such as replacing the model by sending a message to the first communication device.

[0207] An example of the procedure of Method 3 is as follows:

[0208] Step 1: The second communication device can configure a range or levels of allowed measurement load reduction percentage for the first communication device when configuring the model or function, such as {50%~10%} or (40%, 30%, 20%, 10%). In the same message, the control parameters of a performance monitoring event, such as P3, can also be configured, as well as the control parameters of how to perform the performance monitoring.

[0209] Step 2: The first communication device runs the model according to the allowed measurement load reduction percentage of the second communication device (such as choosing 30% from the measurement load reduction percentage, or deciding a value within the allowed range by itself). The first communication device can skip 30% of the measurement opportunities in the time domain or randomly choose 70% of the beams to perform measurement. When the model is running, the first communication device can monitor the model according to the event parameters configured by the second communication device. If P3 is triggered, it means that the performance of one model in the deactivation state is higher than that of another model in the running state.

[0210] Step 3: The first communication device reports the P3 event information and / or the latest performance indicators of the two models, as well as the selected measurement load reduction percentage. The first communication device can also report the cost information, such as the model size information, the calculation times information, the energy consumption information, etc. When receiving the message, the second communication device can decide the further actions, such as instructing the first communication device to replace the model through a message.

[0211] Similar to Method 1, the triggering events in Method 2 and Method 3 can also be used for the management function of the model, as shown in Table 1 and Table 3.

[0212] In the embodiments of the present application, the performance of the model or function is directly related to the size of the prediction window and / or observation window. The above-mentioned Method 1, Method 2 or Method 3 can be used, and the additional condition of “measurement load reduction percentage” in the above example can be replaced by “the size of the prediction window and / or observation window”.

[0213] In the embodiments of the present application, the performance of the model or function is related to the frequency difference between the frequencies. The above-mentioned Method 1, Method 2 or Method 3 can be used, and the additional condition of “measurement load reduction percentage” in the above example can be replaced by “the frequency difference”.

[0214] FIG. 8 is a schematic flowchart of a communication method 800 according to an embodiment of the present application. The method can be optionally applied to the system shown in FIG. 1, but is not limited thereto. The method includes at least part of the following content.

[0215] S810, the second communication device sends configuration information, the configuration information being used for configuring the first communication device to perform a behavior corresponding to a trigger condition in a case where the trigger condition is met, wherein the trigger condition comprises a performance monitoring event of a model.

[0216] In an implementation, the behavior corresponding to the trigger condition comprises a behavior corresponding to the performance monitoring event.

[0217] In an implementation, the trigger condition being met comprises: the trigger condition being met in a case where the performance monitoring event occurs.

[0218] In an implementation, the trigger condition further comprises a time condition, the time condition comprising a timer timeout and / or a counter reaching a maximum number; the performance monitoring event is used to start the timer and / or the counter in a case where the performance monitoring event occurs; the time condition is used to meet the trigger condition in a case where the timer times out and / or the counter reaches the maximum number.

[0219] In an implementation, the performance of the model comprises a measurement error between a predicted measurement result and an actual measurement result, and the performance monitoring event comprises one or more of:

[0220] the measurement error being greater than or equal to a first threshold value;

[0221] the measurement error being less than or equal to a second threshold value;

[0222] the measurement error of the first model being less than or equal to the measurement error of the second model minus a first offset value;

[0223] the measurement error of the first model being less than a third threshold value, and the measurement error of the second model being greater than a fourth threshold value.

[0224] In an implementation, the performance of the model comprises a prediction accuracy, the prediction accuracy being determined based on one or more of a number of predicted events, a number of actually occurred events, and a relationship between the predicted events and the actually occurred events, and the performance monitoring event comprises one or more of:

[0225] the prediction accuracy being greater than or equal to a fifth threshold value;

[0226] the prediction accuracy being less than or equal to a sixth threshold value;

[0227] the prediction accuracy of the first model being greater than or equal to the prediction accuracy of the second model plus a second offset value;

[0228] the prediction accuracy of the first model being greater than a seventh threshold value, and the prediction accuracy of the second model being less than an eighth threshold value.

[0229] In an embodiment, the first model and the second model are two models that are functionally identical but are applicable to different conditions or configurations.

[0230] In an embodiment, the behavior corresponding to the performance monitoring event comprises one or more of the following: deactivating a model; activating a model that has been deactivated; replacing a second model with a first model.

[0231] FIG. 9 is a schematic flowchart of a communication method 900 according to another embodiment of the present application. The method can comprise one or more features of the above-mentioned methods. In an embodiment, the method further comprises:

[0232] S910, the second communication device receives event information and / or model performance information reported by the first communication device in a case where a trigger condition is met;

[0233] S920, the second communication device sends indication information, the indication information being used to instruct the first communication device to perform a behavior corresponding to the trigger condition.

[0234] In an embodiment, the configuration information is used to configure a trigger condition.

[0235] In an embodiment, the configuration information comprises an event control parameter, the event control parameter being used to determine whether a trigger condition is met.

[0236] In an embodiment, the configuration information comprises an additional condition parameter, the additional condition parameter being used to control the running of a model.

[0237] In an embodiment, the method further comprises: the second communication device receiving event information and / or model performance information reported after the performance monitoring event is triggered by the first communication device.

[0238] In an embodiment, the configuration information comprises a plurality of additional condition parameters corresponding to a plurality of event control parameters.

[0239] In an embodiment, the method further comprises: the second communication device receiving one or more of event information, model performance information, and used additional condition parameters reported after the performance monitoring event is triggered by the first communication device.

[0240] In an embodiment, the configuration information comprises a plurality of selectable additional condition parameters.

[0241] In an embodiment, the method further comprises: the second communication device receiving one or more of event information, model performance information, and used additional condition parameters reported after the performance monitoring event is triggered by the first communication device.

[0242] In an embodiment, the event control parameter comprises one or more of: a threshold value, an offset value, a counter, a timer.

[0243] In an embodiment, the additional condition parameter comprises one or more of: a measurement load reduction percentage, a size of a prediction window and / or a size of an observation window, a frequency difference, a time difference of event occurrence.

[0244] In an embodiment, the measurement load reduction percentage is used to determine measurement opportunities skipped in time domain or measurement beams skipped in frequency domain.

[0245] In an embodiment, the size of the prediction window is used to determine a length of time for which the model is run to make a prediction, and the size of the observation window is used to determine a length of time for which historical measurements are used.

[0246] In an embodiment, the frequency difference is used to determine a frequency difference between measurement objects in frequency domain.

[0247] In an embodiment, the time difference of event occurrence is used to determine a difference between a predicted time of event occurrence and an actual time of event occurrence.

[0248] In an embodiment, the method further comprises: the second communication device receiving cost information reported after the first communication device triggers the performance monitoring event, the cost information comprising one or more of: model size information, number of calculations information, energy consumption information.

[0249] Specific examples of the second communication device performing the methods 800, 900 of the present embodiment can be found in the relevant descriptions of the second communication device in the methods 600, 700 described above. For brevity, these will not be described again here.

[0250] Embodiment 1: AI / ML model predicts L1 or L3 measurement results. The measurement results refer to beam level or cell level measurement results. The purpose of model prediction is to reduce measurement load in time domain or spatial domain, i.e. to predict L1 or L3 measurement results based on partial measurements in time domain or spatial domain.

[0251] In this case, the measurement error refers to the absolute value average error or root mean square difference between the predicted L1 or L3 measurement results and the true L1 or L3 measurement results. The predicted measurement results refer to measurement results predicted based on partial or historical measurement results using an AI / ML model. The true measurement results refer to measurement results obtained by actual measurement without using an AI / ML model (but measured based on existing standard specifications). The UE calculates the measurement error within a certain time or number range. The performance monitoring event (or performance triggering event, triggering event, etc.) can be one or more of the following:

[0252] EVENT P1: the measurement error is greater than or equal to a certain threshold;

[0253] EVENT P2: the measurement error is less than or equal to a certain threshold;

[0254] EVENT P3: the measurement error of model A is less than or equal to the measurement error of model B minus an offset value;

[0255] EVENT P4: the measurement error of model A is less than a certain threshold, and the measurement error of model B is greater than a certain threshold.

[0256] In the EVENT P3 or P4, model A and model B refer to two models with the same function but applied to different conditions or configurations. When comparison is needed, the two models are run and monitored in the same pace.

[0257] If the above conditions are considered as the entering conditions of an event, in order to ensure the stability of the event, the network needs to configure a timer or a maximum number, which starts when the entering condition is met. When the timer expires, the event is triggered. Or the counter starts counting when the entering condition is met, and increases by 1 when each (or each pair) measurement error result meets the entering condition. When the counter is greater than or equal to the configured maximum number, the event is triggered.

[0258] When the measurement is predicted in time domain or space domain, if the purpose is to reduce the measurement load, the more the measurement load is reduced, the greater the measurement error is, thereby affecting the triggering of the event. For example, the model makes a prediction in time domain. Suppose through the previous training, the network knows that the measurement load can be reduced to 50% at most, in which case the measurement error of the model prediction is not more than 1db.

[0259] The steps of method 1 are as follows:

[0260] Step 1: When the network configures the model, it configures a UE with a percentage of allowed measurement load reduction, such as 40%. In the same message, a triggering event is configured, such as P1, and it is specified that when P1 meets the condition, the UE needs to report the obtained measurement error. In this message, the threshold of P1 is also configured, as well as the control parameters of how to perform performance monitoring, such as the period and / or the maximum number of periodic monitoring, etc.

[0261] Step 2: UE runs the model according to the allowed measurement load reduction percentage, and skips 40% of the measurement opportunities in time domain or randomly selects 60% of the beams to measure. During the model running, UE monitors the model according to the parameters configured by the network. If the trigger event P1 is triggered, it means that the performance of the model has dropped to a certain extent, and then UE reports the triggered event information to the network, and reports the latest performance index result in the same message.

[0262] The network can further manage the model by using the received trigger event information and the latest performance index, such as activating or deactivating, switching between models, or falling back to an algorithm without model running. The network can also directly associate the trigger event and these management actions to achieve a condition-controlled manner. As shown in Table 4:

[0263] Table 4

[0264] Step of method 2:

[0265] Step 1: When configuring the model, the network configures a range or level of allowed measurement load reduction percentage for UE, such as {40%, 30%, 20%, 10%}. In the same message, a control parameter of trigger event P3 is configured. In this message, offset values corresponding to these measurement load reduction percentages are also configured, as well as control parameters for how to perform performance monitoring, such as periodic monitoring period and / or maximum number of times, etc.

[0266] Step 2: UE runs the model according to the allowed measurement load reduction percentage (such as selecting 30% from it), and skips 30% of the measurement opportunities in time domain or randomly selects 70% of the beams to measure. During the model running, UE monitors the model according to the parameters corresponding to the selected 30% configured by the network. If the trigger event P3 is triggered, it means that the performance of a model in the deactivation state is higher than that of another running model.

[0267] Step 3: UE reports P3 event to the network, and includes the latest performance index of the two models and / or the selected measurement load reduction percentage in the same message. When receiving this message, the network can decide further actions, such as replacing the model by a message to command UE, etc.

[0268] Step of method 3:

[0269] Step 1: Network configures UE with a range or levels of allowed measurement load reduction percentage, such as {50%~10%} or (40%, 30%, 20%, 10%) when configuring the model. In the same message, the network configures the control parameter offset value of the trigger event P3 and how to perform the performance monitoring, such as the period and / or maximum number of times of periodic monitoring.

[0270] Step 2: UE runs the model according to the allowed measurement load reduction percentage (such as choosing 30% from the levels or deciding a value within the allowed range) and skips 30% of the measurement opportunities or randomly selects 70% of the beams to perform measurement in the time domain. When the model is running, the UE monitors the model according to the event parameters configured by the network. If the trigger event P3 is triggered, it means that the performance of one model in the deactivation state is higher than that of another running model.

[0271] Step 3: UE reports P3 event information and / or the latest performance indicators of the two models, as well as the selected measurement load reduction percentage. UE can also report cost information, such as model size information, calculation times information, and energy consumption information. When receiving this message, the network can decide further actions, such as instructing UE to replace the model through a message.

[0272] Similar to Method 1, the trigger events in Method 2 and Method 3 can also be used for model management functions, as shown in Table 3-1.

[0273] Embodiment 2: AI / ML model predicts L1 or L3 measurement results. The measurement results can be beam-level or cell-level measurement results.

[0274] The purpose of model prediction is to improve the performance of mobility (such as handover success rate). In order to improve the prediction performance, the historical measurement results in the observation window generally do not reduce any measurement load in the time domain or spatial domain, which are used to predict L1 or L3 measurement results in the prediction window (future).

[0275] In this case, the performance indicators of the model and the definition of the performance trigger event can follow the measurement error in Embodiment 1.

[0276] The performance of this model is directly related to the size of the prediction window and / or observation window. In the same way, using Method 1, Method 2 or Method 3 in Embodiment 1, as long as the additional condition of “percentage of measurement load reduction” in Embodiment 1 is replaced by “size of the prediction window and / or observation window”. In the specific parameter expression, the prediction window or observation window can be expressed in absolute time length, such as 400ms, 800ms, etc., or expressed as the number of time granularity (such as measurement sampling time). If the two parameters need to be configured at the same time, the ratio between the two can also be used to express.

[0277] Embodiment 3: AI / ML model predicts L1 or L3 measurement results between different frequencies. The measurement results can be beam-level or cell-level measurement results. The performance of the prediction result is related to the frequency difference between the frequencies.

[0278] In this case, the performance indicator of the model and the definition of the performance trigger event can follow the measurement error in Embodiment 1.

[0279] The performance of this model is directly related to the size of the prediction window and / or observation window. In the same way, using Method 1, Method 2 or Method 3 in Embodiment 1, as long as the additional condition of “percentage of measurement load reduction” in Embodiment 1 is replaced by “frequency difference”.

[0280] Embodiment 4: AI / ML model predicts mobility events.

[0281] The performance indicator of the model can be the F1 score introduced in the background art. Obviously, when the counter n2, that is, the number of times the predicted event actually occurs, is higher, the ratio of the F1 score is larger, and this indicator can be simply referred to as prediction accuracy. This indicator is also directly related to the size of the prediction window and the size of the time difference between the predicted event and the actual event for determining the event. Generally, the smaller the prediction window, the smaller the measurement error of the prediction, the higher the reliability of the predicted event, and the smaller the time range of the predicted event.

[0282] In this case, the definition of the trigger event has changed (mainly because the F1 score and the performance are positively related, that is, the percentage can represent the performance). Examples of event definitions:

[0283] EVENT Q1: the prediction accuracy is less than or equal to a certain threshold;

[0284] EVENT Q2: the prediction accuracy is greater than or equal to a certain threshold;

[0285] EVENT Q3: the prediction accuracy of model A is greater than or equal to the prediction accuracy of model B plus an offset value;

[0286] EVENT Q4: the prediction accuracy of model A is greater than a certain threshold value, and the prediction accuracy of model B is less than a certain threshold value.

[0287] In the EVENT Q3 or Q4, model A and model B generally refer to two models with the same function but applicable to different conditions or configurations. When comparison is needed, the operation and monitoring of the two models are consistent.

[0288] If the above conditions are regarded as the entering conditions of the event, in order to ensure the stability of the event, the network can configure a timer or a maximum number. The timer starts when the entering condition is met, and the event is triggered when the timer expires. Or the counter starts counting when the entering condition is met, and increases by 1 when each (or each pair) measurement error result meets the entering condition, and the event is triggered when the counter is greater than or equal to the configured maximum number.

[0289] Similarly, using the method 1, the method 2 and the method 3 in the embodiment 1, as long as the additional condition of "measurement load reduction percentage" in the embodiment 1 is replaced by "the difference between the occurrence time of the prediction window and / or the prediction event and the occurrence time of the actual event". In the specific parameter expression, the two parameters can be expressed by the absolute time length, such as 400 ms, 800 ms, etc., or expressed as the number of a time granularity (such as a measurement sampling time).

[0290] In the embodiments of the present application, the running performance of the UE-side model or the management of the model is efficiently monitored by configuring the measurement event based on the model performance, and the signaling overhead of the wireless interface is saved.

[0291] FIG. 10 is a schematic block diagram of a first communication device 1000 according to an embodiment of the present application. The first communication device 1000 can include:

[0292] The processing unit 1010 is configured to execute a behavior corresponding to a trigger condition in a case where the trigger condition is met, wherein the trigger condition comprises a performance monitoring event of a model.

[0293] In an embodiment, the trigger condition is met, comprising: the trigger condition is met in a case where the performance monitoring event occurs.

[0294] In an embodiment, the trigger condition further comprises a time condition, and the time condition comprises a timer timeout and / or a counter reaching a maximum number; wherein the trigger condition is met, comprising:

[0295] In the case where the performance monitoring event occurs, the first communication device starts a timer and / or a counter;

[0296] In the case where the timer expires and / or the counter reaches a maximum number, the trigger condition is satisfied.

[0297] In an embodiment, the performance of the model comprises a measurement error between the predicted measurement result and the real measurement result, and the performance monitoring event comprises one or more of:

[0298] the measurement error is greater than or equal to a first threshold value;

[0299] the measurement error is less than or equal to a second threshold value;

[0300] the measurement error of the first model is less than or equal to the measurement error of the second model minus a first offset value;

[0301] the measurement error of the first model is less than a third threshold value, and the measurement error of the second model is greater than a fourth threshold value.

[0302] In an embodiment, the performance of the model comprises a prediction accuracy, the prediction accuracy being determined based on one or more of a number of predicted events, a number of actually occurred events, a relationship between the predicted events and the actually occurred events, and the performance monitoring event comprises one or more of:

[0303] the prediction accuracy is greater than or equal to a fifth threshold value;

[0304] the prediction accuracy is less than or equal to a sixth threshold value;

[0305] the prediction accuracy of the first model is greater than or equal to the prediction accuracy of the second model plus a second offset value;

[0306] the prediction accuracy of the first model is greater than a seventh threshold value, and the prediction accuracy of the second model is less than an eighth threshold value.

[0307] In an embodiment, the first model and the second model are two models which are functionally identical but are applicable to different conditions or configurations.

[0308] In an embodiment, the behavior corresponding to the trigger condition comprises a behavior corresponding to the performance monitoring event.

[0309] In an embodiment, the behavior corresponding to the performance monitoring event comprises one or more of: deactivating the model; activating a model which has been deactivated; replacing the second model with the first model.

[0310] FIG. 11 is a schematic flowchart of a first communication device according to another embodiment of the present application. The first communication device can include one or more features of the first communication device described above. In an implementation, the first communication device further includes:

[0311] a first transceiver 1110 configured to report event information and / or model performance information when a trigger condition is satisfied, and receive indication information indicating the first communication device to perform an action corresponding to the trigger condition.

[0312] In an implementation, the method further includes:

[0313] a second transceiver 1120 configured to receive configuration information for configuring the trigger condition.

[0314] In an implementation, the configuration information includes an event control parameter for determining whether the trigger condition is satisfied.

[0315] In an implementation, the configuration information includes an additional condition parameter for controlling running of the model.

[0316] In an implementation, the processing unit is further configured to, when performing the performance monitoring of the model, use the additional condition parameter as a control parameter for running of the model, and determine the trigger condition according to an event control parameter corresponding to the additional condition parameter configured by the second communication device; and report the event information and / or the model performance information after the trigger condition is satisfied.

[0317] In an implementation, the configuration information includes a plurality of additional condition parameters corresponding to a plurality of event control parameters, and the processing unit is further configured to, when performing the performance monitoring of the model, select one additional condition parameter as a control parameter for running of the model, and determine the trigger condition according to an event control parameter corresponding to the selected additional condition parameter; and report one or more of the event information, the model performance information, and the selected additional condition parameter after the trigger condition is satisfied.

[0318] In an implementation, the configuration information includes a plurality of selectable additional condition parameters, and the processing unit is further configured to, when performing the performance monitoring of the model, select one additional condition parameter as a control parameter for running of the model, and determine the trigger condition according to an event control parameter configured by the second communication device; and report one or more of the event information, the model performance information, and the selected additional condition parameter after the trigger condition is satisfied.

[0319] In an implementation, the event control parameter includes one or more of a threshold value, an offset value, a counter, and a timer.

[0320] In an embodiment, the additional condition parameter comprises one or more of: a measurement load reduction percentage, a size of a prediction window and / or a size of an observation window, a frequency difference, a time difference of event occurrence.

[0321] In an embodiment, the measurement load reduction percentage is used to determine a measurement opportunity skipped in time domain or a measurement beam skipped in frequency domain.

[0322] In an embodiment, the size of the prediction window is used to determine a length of time for which a model is run to make a prediction, and the size of the observation window is used to determine a length of time for which historical measurements are used.

[0323] In an embodiment, the frequency difference is used to determine a frequency difference between measurement objects in frequency domain.

[0324] In an embodiment, the time difference of event occurrence is used to determine a difference between a predicted time of occurrence of an event and an actual time of occurrence of the event.

[0325] In an embodiment, the processing unit is further configured to report cost information after triggering the performance monitoring event, the cost information comprising one or more of: model size information, number of calculations information, energy consumption information.

[0326] The first communication device 1000, 1100 of the embodiments of the present application can realize the corresponding functions of the first communication device in the method embodiments described above. The processes, functions, implementation manners and advantages of the respective modules (sub-modules, units or components, etc.) in the first communication device 1000, 1100 can be referred to the corresponding descriptions in the method embodiments described above, and will not be described herein again. It should be noted that the functions described with respect to the respective modules (sub-modules, units or components, etc.) in the first communication device 1000, 1100 of the embodiments of the present application can be realized by different modules (sub-modules, units or components, etc.), or can be realized by the same module (sub-module, unit or component, etc.).

[0327] FIG. 1200 is a schematic block diagram of a second communication device 1200 according to an embodiment of the present application. The second communication device 1200 can comprise a transceiver unit 1210 configured to transmit configuration information, the configuration information being configured to configure a first communication device to perform a behavior corresponding to a trigger condition in a case where the trigger condition is satisfied, wherein the trigger condition comprises a performance monitoring event of a model.

[0328] In an embodiment, the trigger condition is satisfied comprises: in a case where the performance monitoring event occurs, the trigger condition is satisfied.

[0329] In an embodiment, the trigger condition further comprises a time condition, the time condition comprising a timer timeout and / or a counter reaching a maximum number; the performance monitoring event is configured to start the timer and / or the counter upon occurrence of the performance monitoring event; the time condition is configured to satisfy the trigger condition upon the timer timeout and / or the counter reaching the maximum number.

[0330] In an embodiment, the performance of the model comprises a measurement error between the predicted measurement result and the real measurement result, the performance monitoring event comprises one or more of:

[0331] the measurement error is greater than or equal to a first threshold value;

[0332] the measurement error is less than or equal to a second threshold value;

[0333] the measurement error of the first model is less than or equal to the measurement error of the second model minus a first offset value;

[0334] the measurement error of the first model is less than a third threshold value, and the measurement error of the second model is greater than a fourth threshold value.

[0335] In an embodiment, the performance of the model comprises a prediction accuracy, the prediction accuracy being determined based on one or more of a number of predicted events, a number of actually occurred events, a relationship between the predicted events and the actually occurred events, the performance monitoring event comprises one or more of:

[0336] the prediction accuracy is greater than or equal to a fifth threshold value;

[0337] the prediction accuracy is less than or equal to a sixth threshold value;

[0338] the prediction accuracy of the first model is greater than or equal to the prediction accuracy of the second model plus a second offset value;

[0339] the prediction accuracy of the first model is greater than a seventh threshold value, and the prediction accuracy of the second model is less than an eighth threshold value.

[0340] In an embodiment, the first model and the second model are two models of the same function but applicable to different conditions or configurations.

[0341] In an embodiment, the behavior corresponding to the trigger condition comprises the behavior corresponding to the performance monitoring event.

[0342] In an embodiment, the behavior corresponding to the performance monitoring event comprises one or more of: deactivating the model; activating the model that has been deactivated; replacing the second model with the first model.

[0343] In an embodiment, the transceiver unit is further configured to receive event information and / or model performance information reported by the first communication device when the trigger condition is met; and send indication information indicating the first communication device to perform an action corresponding to the trigger condition.

[0344] In an embodiment, the configuration information is used to configure a trigger condition.

[0345] In an embodiment, the configuration information includes an event control parameter used to determine whether the trigger condition is met.

[0346] In an embodiment, the configuration information includes an additional condition parameter used to control the running of the model.

[0347] In an embodiment, the transceiver unit is further configured to receive event information and / or model performance information reported by the first communication device after triggering the performance monitoring event.

[0348] In an embodiment, the configuration information includes a plurality of additional condition parameters corresponding to a plurality of event control parameters, and the transceiver unit is further configured to receive event information, model performance information, and one or more of the used additional condition parameters reported by the first communication device after triggering the performance monitoring event.

[0349] In an embodiment, the configuration information includes a plurality of optional additional condition parameters, and the transceiver unit is further configured to receive event information, model performance information, and one or more of the used additional condition parameters reported by the first communication device after triggering the performance monitoring event.

[0350] In an embodiment, the event control parameter includes one or more of: a threshold value, an offset value, a counter, a timer.

[0351] In an embodiment, the additional condition parameter includes one or more of: a measurement load reduction percentage, a size of a prediction window and / or a size of an observation window, a frequency difference, a time difference of event occurrence.

[0352] In an embodiment, the measurement load reduction percentage is used to determine measurement opportunities skipped in a time domain or measurement beams skipped in a frequency domain.

[0353] In an embodiment, the size of the prediction window is used to determine a length of time for which the model is run to make a prediction, and the size of the observation window is used to determine a length of time for which historical measurement results are observed.

[0354] In an embodiment, the frequency difference is used to determine a frequency difference between measurement objects in a frequency domain.

[0355] In an embodiment, the event occurrence time difference is used to determine a difference between a predicted event occurrence time and an actual event occurrence time.

[0356] In an embodiment, the transceiver unit is further configured to receive cost information reported after the first communication device triggers the performance monitoring event, the cost information comprising one or more of: model size information, calculation number information, energy consumption information.

[0357] The second communication device 1200 of the embodiments of the present application can realize the corresponding functions of the second communication device in the foregoing method embodiments. The corresponding processes, functions, implementation manners, and beneficial effects of each module (sub-module, unit, or component, etc.) in the second communication device 1200 can be referred to the corresponding description in the foregoing method embodiments, which will not be described here. It should be noted that the functions described with respect to each module (sub-module, unit, or component, etc.) in the second communication device 1200 of the embodiments of the present application can be realized by different modules (sub-modules, units, or components, etc.), or can be realized by the same module (sub-module, unit, or component, etc.).

[0358] FIG. 13 is a schematic structural diagram of a communication device 1300 according to an embodiment of the present application. The communication device 1300 comprises a processor 1310, which can invoke and run a computer program from a memory to enable the communication device 1300 to implement the method in the embodiments of the present application.

[0359] In an embodiment, the communication device 1300 can further comprise a memory 1320. The processor 1310 can invoke and run a computer program from the memory 1320 to enable the communication device 1300 to implement the method in the embodiments of the present application.

[0360] The memory 1320 can be a separate device independent of the processor 1310, or can be integrated in the processor 1310.

[0361] In an embodiment, the communication device 1300 can further comprise a transceiver 1330, and the processor 1310 can control the transceiver 1330 to communicate with other devices, specifically, to send information or data to other devices, or to receive information or data sent by other devices.

[0362] The transceiver 1330 can comprise a transmitter and a receiver. The transceiver 1330 can further comprise an antenna, and the number of antennas can be one or more.

[0363] In an embodiment, the communication device 1300 can be a first communication device of the embodiments of the present application, and the communication device 1300 can implement the corresponding procedures implemented by the first communication device in the various methods of the embodiments of the present application. For brevity, details are not repeated here.

[0364] In an embodiment, the communication device 1300 can be a second communication device of the embodiments of the present application, and the communication device 1300 can implement the corresponding procedures implemented by the second communication device in the various methods of the embodiments of the present application. For brevity, details are not repeated here.

[0365] FIG. 14 is a schematic structural diagram of a chip 1400 according to an embodiment of the present application. The chip 1400 includes a processor 1410, which can call and run a computer program from a memory to implement the methods in the embodiments of the present application.

[0366] In an embodiment, the chip 1400 can further include a memory 1420. The processor 1410 can call and run a computer program from the memory 1420 to implement the methods performed by the first communication device or the second communication device in the embodiments of the present application.

[0367] The memory 1420 can be a separate device independent of the processor 1410, or can be integrated in the processor 1410.

[0368] In an embodiment, the chip 1400 can further include an input interface 1430. The processor 1410 can control the input interface 1430 to communicate with other devices or chips, and specifically, can obtain information or data sent by other devices or chips.

[0369] In an embodiment, the chip 1400 can further include an output interface 1440. The processor 1410 can control the output interface 1440 to communicate with other devices or chips, and specifically, can output information or data to other devices or chips.

[0370] In an embodiment, the chip can be applied to the first communication device in the embodiments of the present application, and the chip can implement the corresponding procedures implemented by the first communication device in the various methods of the embodiments of the present application. For brevity, details are not repeated here.

[0371] In an embodiment, the chip can be applied to the second communication device in the embodiments of the present application, and the chip can implement the corresponding procedures implemented by the second communication device in the various methods of the embodiments of the present application. For brevity, details are not repeated here.

[0372] The chip applied to the first communication device and the second communication device can be the same chip or different chips.

[0373] It should be understood that the chip mentioned in the embodiments of the present application can also be referred to as a system chip, a system chip, a chip system or a system on chip, etc.

[0374] The processor mentioned above can be a general-purpose processor, a digital signal processor (DSP), a field programmable gate array (FPGA), an application specific integrated circuit (ASIC) or other programmable logic devices, transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor mentioned above can be a microprocessor or any conventional processor, etc.

[0375] The memory mentioned above can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically EPROM (EEPROM) or a flash memory. The volatile memory can be a random access memory (RAM).

[0376] It should be understood that the above-mentioned memory is an example but not a limiting description, for example, the memory in the embodiments of the present application can also be a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDR SDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synchronous link dynamic random access memory (SLDRAM) and a direct memory bus random access memory (Direct Rambus RAM, DR RAM), etc. That is, the memory in the embodiments of the present application is intended to include but not limited to these and any other suitable type of memory.

[0377] FIG. 15 is a schematic block diagram of a communication system 1500 according to embodiments of the present application. The communication system 1500 includes a first communication device 1510 and a second communication device 1520. The first communication device 1510 is configured to perform a behavior corresponding to a trigger condition in a case where the trigger condition is satisfied, wherein the trigger condition comprises a performance monitoring event of a model. The second communication device 1520 is configured to send configuration information for configuring the first communication device to perform the behavior corresponding to the trigger condition in a case where the trigger condition is satisfied, wherein the trigger condition comprises a performance monitoring event of a model. The first communication device 1510 can be configured to implement the corresponding function of the first communication device in the above-described method, and the second communication device 1520 can be configured to implement the corresponding function of the first communication device in the above-described method. For brevity, details are not repeated here.

[0378] In the above embodiments, all or some of the processes can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or some of the processes can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed by a computer, the computer instructions generate the processes or functions according to the embodiments of the present application. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) manner. The computer readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media. The available medium can be a magnetic medium (for example, floppy disk, hard disk, magnetic tape), an optical medium (for example, DVD), or a semiconductor medium (for example, solid state disk (SSD)) and the like.

[0379] It should be understood that the size of the sequence number of each process described above in various embodiments of the present application does not mean the order of execution, and the execution order of each process should be determined by its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0380] Those skilled in the art can clearly understand the specific working process of the system, device and unit described above for the convenience and brevity of description, which can refer to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0381] The above merely describes specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which shall be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method of communication, comprising: The first communication device performs a behavior corresponding to the trigger condition in a case where the trigger condition is met, wherein the trigger condition comprises a performance monitoring event of a model.

2. The method of claim 1, wherein, The trigger condition is met in a case where the performance monitoring event occurs.

3. The method of claim 1, wherein, The trigger condition further comprises a time condition, the time condition comprising a timer timeout and / or a counter reaching a maximum number; wherein the trigger condition is met in a case where: The first communication device starts a timer and / or a counter in a case where the performance monitoring event occurs; The trigger condition is met in a case where the timer times out and / or the counter reaches the maximum number.

4. The method of any one of claims 1 to 3, wherein, The performance of the model comprises a measurement error between a predicted measurement result and an actual measurement result, and the performance monitoring event comprises one or more of: The measurement error is greater than or equal to a first threshold value; The measurement error is less than or equal to a second threshold value; The measurement error of a first model is less than or equal to the measurement error of a second model minus a first offset value; The measurement error of the first model is less than a third threshold value, and the measurement error of the second model is greater than a fourth threshold value.

5. The method of any one of claims 1 to 4, wherein, The performance of the model comprises a prediction accuracy, the prediction accuracy being determined based on one or more of a number of predicted events, a number of actually occurred events, and a relationship between the predicted events and the actually occurred events, and the performance monitoring event comprises one or more of: The prediction accuracy is greater than or equal to a fifth threshold value; The prediction accuracy is less than or equal to a sixth threshold value; The prediction accuracy of a first model is greater than or equal to the prediction accuracy of a second model plus a second offset value; The prediction accuracy of the first model is greater than a seventh threshold value, and the prediction accuracy of the second model is less than an eighth threshold value.

6. The method of claim 4 or 5, wherein, The first model and the second model are two models that are functionally identical but are applicable to different conditions or configurations.

7. The method of any one of claims 1 to 6, wherein, The behavior corresponding to the trigger condition comprises a behavior corresponding to the performance monitoring event.

8. The method of claim 7, wherein, The behavior corresponding to the performance monitoring event comprises one or more of: deactivating a model; activating a model that has been deactivated; replacing a second model with a first model.

9. The method of any one of claims 1 to 8, wherein, The first communication device performs a behavior corresponding to the trigger condition in a case where the trigger condition is met, comprising: The first communication device reports event information and / or model performance information in a case where the trigger condition is met. The first communication device receives indication information, the indication information being used to instruct the first communication device to perform the behavior corresponding to the trigger condition.

10. The method of any one of claims 1 to 9, wherein, The method further comprises: The first communication device receives configuration information, the configuration information being used to configure a trigger condition.

11. The method of claim 10, wherein, The configuration information comprises an event control parameter, the event control parameter being used to determine whether the trigger condition is met.

12. The method of claim 10 or 11, wherein, The configuration information comprises an additional condition parameter, the additional condition parameter being used to control running of a model.

13. The method of claim 12, wherein, The method further comprises: The first communication device uses the additional condition parameter as a control parameter of model running when performing performance monitoring of the model, and judges the trigger condition according to the event control parameter configured by the second communication device; and reports event information and / or model performance information after the trigger condition is met.

14. The method of claim 10 or 11, wherein, The configuration information includes a plurality of additional condition parameters corresponding to a plurality of event control parameters, and the method further includes: The first communication device uses one additional condition parameter as a control parameter of model running when performing performance monitoring of the model, and judges the trigger condition according to the event control parameter corresponding to the used additional condition parameter; and reports one or more of event information, model performance information, and the used additional condition parameter after the trigger condition is met.

15. The method of claim 10 or 11, wherein, The configuration information includes a plurality of selectable additional condition parameters, and the method further includes: The first communication device uses one additional condition parameter as a control parameter of model running when performing performance monitoring of the model, and judges the trigger condition according to the event control parameter configured by the second communication device; and reports one or more of event information, model performance information, and the used additional condition parameter after the trigger condition is met.

16. The method of any one of claims 11 to 15, wherein, The event control parameter includes one or more of a threshold value, an offset value, a counter, and a timer.

17. The method of any one of claims 12 to 15, wherein, The additional condition parameter includes one or more of a measurement load reduction percentage, a size of a prediction window and / or a size of an observation window, a frequency difference, and a time difference of event occurrence.

18. The method of claim 17, wherein, The measurement load reduction percentage is used to determine a measurement opportunity skipped in a time domain or a measurement beam skipped in a frequency domain; The size of the prediction window is used to determine a length of time for which the model is run to make a prediction, and the size of the observation window is used to determine a length of time for which historical measurement results are observed; The frequency difference is used to determine a frequency difference between measurement objects in a frequency domain; The time difference of event occurrence is used to determine a difference between a predicted time of event occurrence and an actual time of event occurrence.

19. The method of any one of claims 1 to 18, wherein, The method further includes: The first communication device reports cost information after triggering the performance monitoring event, and the cost information includes one or more of model size information, calculation frequency information, and energy consumption information.

20. A method of communication, comprising: The second communication device sends configuration information, and the configuration information is used to configure the first communication device to perform a behavior corresponding to a trigger condition in a case where the trigger condition is met, wherein the trigger condition includes a performance monitoring event of a model.

21. The method of claim 20, wherein, The trigger condition is met, including: the trigger condition is met in a case where the performance monitoring event occurs.

22. The method of claim 20, wherein, The trigger condition further includes a time condition, and the time condition includes a timer timeout and / or a counter reaching a maximum number; the performance monitoring event is used to start the timer and / or the counter in a case where the performance monitoring event occurs; and the time condition is used to meet the trigger condition in a case where the timer times out and / or the counter reaches the maximum number.

23. The method of any one of claims 20-22, wherein, The performance of the model includes a measurement error between a predicted measurement result and an actual measurement result, and the performance monitoring event includes one or more of: the measurement error is greater than or equal to a first threshold value; the measurement error is less than or equal to a second threshold value; the measurement error of the first model is less than or equal to the measurement error of the second model minus a first offset value; the measurement error of the first model is less than a third threshold value, and the measurement error of the second model is greater than a fourth threshold value.

24. The method of any one of claims 20-23, wherein, the performance of the model includes a prediction accuracy, the prediction accuracy is determined based on one or more of a number of predicted events, a number of actually occurred events, a relationship between the predicted events and the actually occurred events, the performance monitoring event includes one or more of the following: the prediction accuracy is greater than or equal to a fifth threshold value; the prediction accuracy is less than or equal to a sixth threshold value; the prediction accuracy of the first model is greater than or equal to the prediction accuracy of the second model plus a second offset value; the prediction accuracy of the first model is greater than a seventh threshold value, and the prediction accuracy of the second model is less than an eighth threshold value.

25. The method of claim 23 or 24, wherein, the first model and the second model are two functionally identical models but applicable to different conditions or configurations.

26. The method of any one of claims 20-25, wherein, the behavior corresponding to the trigger condition includes the behavior corresponding to the performance monitoring event.

27. The method of claim 26, wherein, the behavior corresponding to the performance monitoring event includes one or more of the following: deactivating a model; activating a model that has been deactivated; replacing a second model with a first model.

28. The method of any one of claims 20 to 27, wherein, the method further includes: the second communication device receives the event information and / or the model performance information reported by the first communication device when the trigger condition is met; the second communication device sends indication information, the indication information being used to instruct the first communication device to perform the behavior corresponding to the trigger condition.

29. The method of any one of claims 20 to 28, wherein, the configuration information is used to configure a trigger condition.

30. The method of claim 29, wherein, the configuration information includes an event control parameter, the event control parameter being used to determine whether the trigger condition is met.

31. The method of claim 29 or 30, wherein, the configuration information includes an additional condition parameter, the additional condition parameter being used to control the running of a model.

32. The method of claim 31, wherein, the method further includes: the second communication device receives the event information and / or the model performance information reported by the first communication device after the performance monitoring event is triggered by the first communication device.

33. The method of claim 29 or 30, wherein, the configuration information includes a plurality of additional condition parameters corresponding to a plurality of event control parameters, the method further includes: the second communication device receives one or more of the event information, the model performance information, and the used additional condition parameters reported by the first communication device after the performance monitoring event is triggered by the first communication device.

34. The method of claim 29 or 30, wherein, the configuration information includes a plurality of selectable additional condition parameters, the method further includes: the second communication device receives one or more of the event information, the model performance information, and the used additional condition parameters reported by the first communication device after the performance monitoring event is triggered by the first communication device.

35. The method of any one of claims 30-34, wherein, the event control parameter includes one or more of the following: a threshold value, an offset value, a counter, a timer.

36. The method of any one of claims 31 to 34, wherein, the additional condition parameter includes one or more of the following: a measurement load reduction percentage, a size of a prediction window and / or a size of an observation window, a frequency difference, a time difference of event occurrence.

37. The method of claim 36, wherein, the measurement load reduction percentage is used to determine a measurement opportunity skipped in a time domain or a measurement beam skipped in a frequency domain; The size of the prediction window is used to determine a length of time for which the model is run to make a prediction, and the size of the observation window is used to determine a length of time for which historical measurements are used; The frequency difference is used to determine a difference in frequency between the measurement objects in the frequency domain; The event occurrence time difference is used to determine a difference between a predicted occurrence time of an event and an actual occurrence time of the event.

38. The method of any one of claims 20 to 37, wherein, The method further comprises: The second communication device receives cost information reported by the first communication device after the performance monitoring event is triggered, and the cost information includes one or more of the following: model size information, calculation frequency information, energy consumption information.

39. A first communication device, comprising: The processing unit is configured to, in a case where a trigger condition is met, cause the first communication device to perform a behavior corresponding to the trigger condition, wherein the trigger condition includes a performance monitoring event of a model.

40. A first communications device according to Claim 39, wherein, The trigger condition is met in a case where the performance monitoring event occurs.

41. The first communication device of claim 39, wherein, The trigger condition further includes a time condition, and the time condition includes a timer timeout and / or a counter reaching a maximum number; wherein the trigger condition is met in a case where: The first communication device starts a timer and / or a counter in a case where the performance monitoring event occurs; The trigger condition is met in a case where the timer times out and / or the counter reaches the maximum number.

42. A first communications device according to any one of claims 39 to 41, wherein, The performance of the model includes a measurement error between a predicted measurement result and an actual measurement result, and the performance monitoring event includes one or more of the following: The measurement error is greater than or equal to a first threshold value; The measurement error is less than or equal to a second threshold value; The measurement error of a first model is less than or equal to the measurement error of a second model minus a first offset value; The measurement error of the first model is less than a third threshold value, and the measurement error of the second model is greater than a fourth threshold value.

43. A first communications device according to any one of claims 39 to 42, wherein, The performance of the model includes a prediction accuracy, and the prediction accuracy is determined based on one or more of the following: a number of predicted events, a number of actually occurred events, and a relationship between the predicted events and the actually occurred events, and the performance monitoring event includes one or more of the following: The prediction accuracy is greater than or equal to a fifth threshold value; The prediction accuracy is less than or equal to a sixth threshold value; The prediction accuracy of a first model is greater than or equal to the prediction accuracy of a second model plus a second offset value; The prediction accuracy of the first model is greater than a seventh threshold value, and the prediction accuracy of the second model is less than an eighth threshold value.

44. A first communications device according to claim 42 or 43, wherein, The first model and the second model are two models that have the same function but are applicable to different conditions or configurations.

45. A first communications device according to any one of claims 39 to 44, wherein, The behavior corresponding to the trigger condition includes a behavior corresponding to the performance monitoring event.

46. The first communication device of claim 45, wherein, The behavior corresponding to the performance monitoring event includes one or more of the following: deactivating a model; activating a model that has been deactivated; replacing a second model with a first model.

47. The first communication device of any one of any one of claims 39 to 46, wherein, The first communication device further comprises a first transceiver configured to, in a case where a trigger condition is met, report event information and / or model performance information by the first communication device, and receive indication information, wherein the indication information is used to instruct the first communication device to perform a behavior corresponding to the trigger condition.

48. A first communications device according to any one of claims 39 to 47, wherein, The first communication device further comprises a second transceiver configured to receive configuration information, the configuration information being used to configure a triggering condition.

49. The first communication device of claim 48, wherein, The configuration information comprises an event control parameter, the event control parameter being used to determine whether the triggering condition is met.

50. A first communications device according to claim 48 or 49, wherein, The configuration information comprises an additional condition parameter, the additional condition parameter being used to control the running of the model.

51. The first communication device of claim 50, wherein, The processing unit is further configured to, when performing the performance monitoring of the model, use the additional condition parameter as a control parameter for the running of the model, and determine the triggering condition according to the event control parameter configured by the second communication device; and report event information and / or model performance information after the triggering condition is triggered.

52. A first communications device according to claim 48 or 49, wherein, The configuration information comprises a plurality of additional condition parameters corresponding to a plurality of event control parameters, and the processing unit is further configured to, when performing the performance monitoring of the model, select one additional condition parameter as a control parameter for the running of the model, and determine the triggering condition according to the event control parameter corresponding to the used additional condition parameter; and report one or more of the event information, the model performance information, and the used additional condition parameter after the triggering condition is met.

53. The first communication device of claim 48 or 49, wherein, The configuration information comprises a plurality of selectable additional condition parameters, and the processing unit is further configured to, when performing the performance monitoring of the model, select one additional condition parameter as a control parameter for the running of the model, and determine the triggering condition according to the event control parameter configured by the second communication device; and report one or more of the event information, the model performance information, and the used additional condition parameter after the triggering condition is met.

54. A first communications device according to any one of claims 49 to 53, wherein, The event control parameter comprises one or more of a threshold value, an offset value, a counter, and a timer.

55. A first communications device according to any one of claims 50 to 53, wherein, The additional condition parameter comprises one or more of a measured load reduction percentage, a size of a prediction window and / or a size of an observation window, a frequency difference, and a time difference of event occurrence.

56. The first communication device of claim 55, wherein, The measured load reduction percentage is used to determine a measurement opportunity skipped in a time domain or a measurement beam skipped in a frequency domain; The size of the prediction window is used to determine a length of time for which the model is run to make a prediction, and the size of the observation window is used to determine a length of time for which historical measurement results are observed; The frequency difference is used to determine a frequency difference between measurement objects in a frequency domain; The time difference of event occurrence is used to determine a difference between a predicted time of event occurrence and an actual time of event occurrence.

57. A first communications device according to any one of claims 39 to 56, wherein, The processing unit is further configured to report cost information after the performance monitoring event is triggered, the cost information comprising one or more of model size information, calculation frequency information, and energy consumption information.

58. A second communication device, comprising: The transceiver is configured to send configuration information, the configuration information being used to configure the first communication device to perform a behavior corresponding to a triggering condition in a case where the triggering condition is met, wherein the triggering condition comprises a performance monitoring event of a model.

59. The second communication device of claim 58, wherein, The triggering condition is met in a case where the performance monitoring event occurs.

60. The second communication device of claim 58, wherein, The trigger condition further comprises a time condition, the time condition comprising a timer timeout and / or a counter reaching a maximum number; the performance monitoring event is used to start a timer and / or a counter in a case where the performance monitoring event occurs; the time condition is used to satisfy the trigger condition in a case where the timer times out and / or the counter reaches the maximum number.

61. A second communications device according to any one of claims 58 to 60, wherein, The performance of the model comprises a measurement error between a predicted measurement result and an actual measurement result, the performance monitoring event comprises one or more of: The measurement error is greater than or equal to a first threshold value; The measurement error is less than or equal to a second threshold value; The measurement error of the first model is less than or equal to the measurement error of the second model minus a first offset value; The measurement error of the first model is less than a third threshold value, and the measurement error of the second model is greater than a fourth threshold value.

62. A second communications device according to any one of claims 58 to 61, wherein, The performance of the model comprises a prediction accuracy, the prediction accuracy being determined based on one or more of a number of predicted events, a number of actually occurred events, a relationship between the predicted events and the actually occurred events, the performance monitoring event comprises one or more of: The prediction accuracy is greater than or equal to a fifth threshold value; The prediction accuracy is less than or equal to a sixth threshold value; The prediction accuracy of the first model is greater than or equal to the prediction accuracy of the second model plus a second offset value; The prediction accuracy of the first model is greater than a seventh threshold value, and the prediction accuracy of the second model is less than an eighth threshold value.

63. A second communications device according to claim 61 or 62, wherein, The first model and the second model are two models that are functionally identical but are applicable to different conditions or configurations.

64. A second communications device according to any one of claims 58 to 63, wherein, The behavior corresponding to the trigger condition comprises a behavior corresponding to the performance monitoring event.

65. The second communication device of claim 64, wherein, The behavior corresponding to the performance monitoring event comprises one or more of: Deactivating a model; 66. A second communications device according to any one of claims 58 to 65, wherein, Activating a model that has been deactivated; replacing a second model with a first model. The transceiver is further configured to receive event information and / or model performance information reported by the first communication device in a case where a trigger condition is satisfied.

67. A second communications device according to any one of claims 58 to 66, wherein, The transceiver is further configured to receive event information and / or model performance information reported by the first communication device in a case where a trigger condition is satisfied.

68. The second communication device of claim 67, wherein, The configuration information is used to configure a trigger condition.

69. A second communications device according to claim 67 or 68, wherein, The configuration information comprises an event control parameter, the event control parameter being used to determine whether a trigger condition is satisfied.

70. A second communications device according to Claim 69, wherein, The configuration information comprises an additional condition parameter, the additional condition parameter being used to control the running of a model.

71. The second communication device of claim 67 or 68, wherein, The transceiver is further configured to receive event information and / or model performance information reported by the first communication device after the performance monitoring event is triggered by the first communication device.

72. A second communications device according to claim 67 or 68, wherein, The configuration information comprises a plurality of additional condition parameters corresponding to a plurality of event control parameters, the transceiver is further configured to receive one or more of event information, model performance information, and used additional condition parameters reported by the first communication device after the performance monitoring event is triggered by the first communication device. The configuration information comprises a plurality of selectable additional condition parameters, and the transceiver is further configured to receive one or more of event information, model performance information, and used additional condition parameters reported by the first communication device after the performance monitoring event is triggered by the first communication device.

73. A second communications device according to any one of claims 68 to 72, wherein, The event control parameter comprises one or more of: a threshold value, an offset value, a counter, a timer.

74. A second communications device according to any one of claims 69 to 72, wherein, The additional condition parameter comprises one or more of: a measurement load reduction percentage, a size of a prediction window and / or a size of an observation window, a frequency difference, a time difference of event occurrence.

75. A second communications device according to Claim 74, wherein, The measurement load reduction percentage is used to determine measurement opportunities skipped in time domain or measurement beams skipped in frequency domain; The size of the prediction window is used to determine a length of time for which a model is run to make a prediction, and the size of the observation window is used to determine a length of time for which historical measurement results are used; The frequency difference is used to determine a frequency difference between measurement objects in frequency domain; The time difference of event occurrence is used to determine a difference between a predicted time of event occurrence and an actual time of event occurrence.

76. A second communications device according to any one of claims 58 to 75, wherein, The transceiver is further configured to receive cost information reported after the performance monitoring event is triggered by the first communication device, the cost information comprising one or more of: model size information, number of calculations information, energy consumption information.

77. A communication device, comprising: A transceiver, a processor and a memory, the memory being configured to store a computer program, the transceiver being configured to communicate with other devices, and the processor being configured to invoke and run the computer program stored in the memory, so that the communication device performs the method of any one of claims 1 to 38.

78. A chip comprising: A processor configured to invoke and run a computer program from a memory, so that a device in which the chip is installed performs the method of any one of claims 1 to 38.

79. A computer readable storage medium configured to store a computer program which, when run by a device, causes the device to perform the method of any one of claims 1 to 38.

80. A computer program product comprising computer program instructions which cause a computer to perform the method of any one of claims 1 to 38.

81. A computer program which causes a computer to perform the method of any one of claims 1 to 38.

82. A communication system comprising: a first communication device configured to perform the method of any one of claims 1 to 19; and a second communication device configured to perform the method of any one of claims 20 to 38.

Images