Frequency-domain measurement prediction method and apparatus, and processor-readable storage medium

Abstract

The present application relates to a frequency-domain measurement prediction method and apparatus, and a processor-readable storage medium. The method comprises: acquiring a first actual RRM measurement result of a UE on a first frequency point or frequency point combination; and using the first actual RRM measurement result as an input to a model to perform derivation, in order to obtain a predicted RRM measurement result of the UE on a second frequency point or frequency point combination.

Description

Frequency domain measurement prediction method, device and processor-readable storage medium Cross-references

[0001] This application refers to Chinese Patent Application No. 2024103606680, filed on March 27, 2024, entitled “Frequency Domain Measurement Prediction Method, Device and Processor-Readable Storage Medium”, which is incorporated into this application in its entirety by reference. Technical Field

[0002] The present application relates to the field of communication technology, and in particular to a frequency domain measurement prediction method, device, and processor-readable storage medium. Background Art

[0003] The 3rd Generation Partnership Project (3GPP) proposed the measurement gap (also known as measurement GAP) mechanism to support user equipment (UE) performing radio resource management (RRM) measurements on different frequencies / bands. A measurement gap is a specific time period configured by the network for the UE. During this time, the UE suspends communications on its current operating frequency / band and adjusts its receiver to perform RRM measurements on the frequency / band to be measured.

[0004] However, the above-mentioned method of performing RRM measurement on the frequency point by the UE has the problem of consuming too much measurement overhead. Summary of the Invention

[0005] Based on this, it is necessary to provide a frequency domain measurement prediction method, device and processor-readable storage medium to address the above technical problems, which can reduce the measurement overhead consumed by the UE in performing RRM measurements on the frequency points.

[0006] In a first aspect, an embodiment of the present application provides a frequency domain measurement prediction method, applied to a communication device, the method comprising:

[0007] Obtaining a first measured RRM measurement result of the UE on a first frequency or frequency combination;

[0008] The first measured RRM measurement result is used as an input of the model to perform derivation to obtain a predicted RRM measurement result of the UE at the second frequency point or frequency point combination.

[0009] In one embodiment, the first frequency point or frequency point combination includes at least one of the following: one or more co-frequency frequency points; one or more heterofrequency frequency points; a combination of co-frequency frequency points and heterofrequency frequency points; wherein the co-frequency frequency point is the same frequency relative to the current working frequency point of the UE, and the heterofrequency frequency point is the heterofrequency relative to the current working frequency point of the UE.

[0010] In one embodiment, the second frequency point or frequency point combination includes at least one of the following: one or more same-frequency points; one or more different-frequency points; one or more different-technology points; a combination of at least two of the same-frequency points, different-frequency points and different-technology points.

[0011] In one embodiment, the first measured RRM measurement result or the predicted RRM measurement result includes at least one of the following: a serving cell measurement result; a neighboring cell measurement result; a combination of serving cell and neighboring cell measurement results; or a UE measurement result of a beam in the cell.

[0012] In one embodiment, the type of the first measured RRM measurement result or the predicted RRM measurement result includes at least one of the following: reference signal received power; reference signal received quality; signal-to-noise ratio.

[0013] In one embodiment, the communication device is a network side device, and obtaining a first measured RRM measurement result of the UE on a first frequency point or a frequency point combination includes: receiving the first measured RRM measurement result periodically reported by the UE according to the measurement configuration.

[0014] In one embodiment, the first measured RRM measurement result is used as an input of the model for derivation to obtain a predicted RRM measurement result of the UE at the second frequency point or frequency point combination, including:

[0015] Obtain UE related information;

[0016] The first measured RRM measurement result and UE-related information are used as inputs of the model to perform deduction to obtain a predicted RRM measurement result.

[0017] In one embodiment, obtaining UE related information includes: obtaining UE related information collected by the UE, and / or obtaining UE related information collected by a network side device.

[0018] In one embodiment, obtaining UE-related information collected by the UE includes:

[0019] Receive UE-related information periodically reported by the UE according to the measurement configuration.

[0020] In one embodiment, the UE-related information includes at least one of the following: UE location information; UE speed information.

[0021] In one of the embodiments, the location information of the UE includes at least one of the following: the geographic location coordinates of the UE; location-related information obtained by the UE based on a positioning method; and location-related information obtained by a network-side device based on a positioning method.

[0022] In one embodiment, the UE speed information includes: UE actual speed information, or various parameter information related to the UE speed.

[0023] In one of the embodiments, the method further includes: making a handover decision of the UE according to the predicted RRM measurement result.

[0024] In one of the embodiments, the communication device is a UE, and the method further includes: sending the predicted RRM measurement result to a network-side device.

[0025] In one embodiment, the method further comprises:

[0026] Monitor model execution;

[0027] Perform management actions on the model based on the monitoring results of the model and / or the cost saving information associated with the model.

[0028] In one embodiment, monitoring the model includes at least one of the following:

[0029] Compare the performance indicator values ​​predicted by the application model with the preset / network configured performance indicator values;

[0030] Compare the performance index values ​​predicted by the application model with the performance index values ​​obtained by actual measurement;

[0031] Compare the performance index values ​​predicted by the application model with those predicted by other different models;

[0032] comparing the predicted RRM measurement result with a second measured RRM measurement result at the second frequency point / frequency point combination;

[0033] The predicted RRM measurement result is compared with the RRM measurement result at the second frequency point / frequency point combination predicted by using other different models.

[0034] In one embodiment, the performance indicator value includes at least one of the following: radio link failure rate; handover failure rate;

[0035] Ping-pong switching probability; interruption time length.

[0036] In one embodiment, the overhead saving information includes at least one of the following: interruption time saving time; measurement of GAP usage time; measurement of GAP saving time; measurement of required UE power saving.

[0037] In one embodiment, the management operation includes performing at least one of the following operations:

[0038] Activate the model or model-related functions;

[0039] Deactivate the model or model-related functions;

[0040] Transforming the model into another model with better performance indicators and / or RRM measurements;

[0041] Convert model-related features into other model-related features that provide better performance indicators and / or RRM measurements;

[0042] Roll back the model or model-related functions;

[0043] Update the model or model-related functions;

[0044] Retrain the model;

[0045] Fine-tune the model.

[0046] In one embodiment, before deriving the model using the first measured RRM measurement result as an input, the method further includes:

[0047] Training the model based on the training data; and / or,

[0048] Receive the model sent by the training node. The model is trained by the training node based on the training data.

[0049] In one embodiment, the training node includes at least one of the following: a third-party server; a base station node; an operator node;

[0050] UE; core network node.

[0051] In one of the embodiments, the training data includes at least one of the following: the measured RRM measurement result of the UE on the third frequency point / frequency point combination; the measured RRM measurement result of the UE on the fourth frequency point / frequency point combination; UE-related information collected by the UE; and UE-related information collected by the network side device.

[0052] In one embodiment, the method further comprises:

[0053] The UE sends model-related recommended or preferred GAP configuration related information to the network side device.

[0054] In one embodiment, the GAP configuration related information includes at least one of the following:

[0055] GAP use or non-use for a single or multiple frequency points; the frequency point includes at least one of the same frequency point, different frequency point, and different technology frequency point;

[0056] GAP relaxation on single or multiple frequency points;

[0057] Frequency identifiers used for single or multiple frequency GAPs;

[0058] Frequency identifiers not used for GAP of single or multiple frequencies;

[0059] Frequency identifier for GAP relaxation for single or multiple frequencies.

[0060] In one embodiment, GAP relaxation includes at least one of the following:

[0061] Configure a longer GAP repetition period than the current GAP repetition period;

[0062] Configure a GAP length that is shorter than the current GAP repetition period;

[0063] Punch holes or omit some GAP positions.

[0064] In one embodiment, the UE sends model-related recommended or preferred GAP configuration related information to the NW, including:

[0065] The UE sends model-related recommendations or preferred GAP configuration information to the NW through any of the following signaling:

[0066] Radio resource control signaling;

[0067] Media access control layer control elements;

[0068] Physical layer instructions.

[0069] In one embodiment, the method further comprises:

[0070] The network side device sends the model-related GAP configuration or related instructions to the UE.

[0071] In one embodiment, the GAP configuration or related indication includes at least one of the following:

[0072] Whether the GAP configuration recommended or preferred by the UE is adopted;

[0073] Adoption of GAP configurations recommended or preferred by the UE;

[0074] Model-dependent GAP use or non-use for the same frequency points;

[0075] Model-dependent GAP use or non-use for different frequency points;

[0076] Model-dependent use or non-use of GAP for different technology frequencies;

[0077] Model-dependent GAP relaxation at single or multiple frequency points;

[0078] Model-related GAP configuration.

[0079] In one embodiment, the network side device sends a model-related GAP configuration or related indication to the UE, including:

[0080] The network-side device sends a model-related GAP configuration or related indication to the UE based on at least one of the following information:

[0081] Recommended or preferred GAP configuration information sent by the UE;

[0082] UE or its own monitoring results for the model.

[0083] In one embodiment, the network side device sends a model-related GAP configuration or related indication to the UE, including:

[0084] The network side device sends the model-related GAP configuration or related instructions to the UE through any of the following signaling:

[0085] Radio resource control signaling;

[0086] Media access control layer control elements;

[0087] Physical layer instructions.

[0088] In a second aspect, an embodiment of the present application provides a frequency domain measurement prediction device, the device comprising:

[0089] An acquisition module, configured to obtain a first measured RRM measurement result of the UE on a first frequency or frequency combination;

[0090] The prediction module is configured to perform derivation using the first measured RRM measurement result as an input of the model to obtain a predicted RRM measurement result of the UE at the second frequency point or frequency point combination.

[0091] In a third aspect, an embodiment of the present application provides a communication device, which includes a memory, a transceiver, and a processor; the memory is used to store computer programs; the transceiver is used to send and receive data under the control of the processor; and the processor is used to read the computer program in the memory and execute the steps of the frequency domain measurement prediction method provided in any embodiment of the first aspect above.

[0092] In a fourth aspect, an embodiment of the present application provides a processor-readable storage medium, which stores a program for causing a processor to execute the steps of the frequency domain measurement prediction method provided in any embodiment of the first aspect above.

[0093] The frequency domain measurement prediction method, device and processor-readable storage medium provided in the embodiments of the present application. By obtaining the first measured RRM measurement result of the UE at the first frequency point or frequency point combination, and using the first measured RRM measurement result as the input of the model for deduction, the predicted RRM measurement result of the UE at the second frequency point or frequency point combination is obtained. This is equivalent to using a pre-trained model to infer the RRM measurement result of the UE at the first frequency point or frequency point combination based on the RRM measurement result of the UE at the second frequency point or frequency point combination, so that the RRM measurement of the UE at the second frequency point or frequency point combination is predicted by the model, without consuming measurement overhead such as measurement GAP and UE power, thereby greatly reducing the measurement time overhead consumed by the UE to perform RRM measurements on some frequency points, and ensuring the UE data transmission and reception throughput.

[0094] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the embodiments below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present application. The same reference numerals are used throughout the drawings to denote the same components. In the drawings:

[0096] FIG1 is a schematic diagram of the internal structure of a communication device provided in one embodiment;

[0097] FIG2 is a schematic flow chart of a frequency domain measurement prediction method provided in one embodiment;

[0098] FIG3 is a schematic diagram of a model lifecycle management function module provided in one embodiment;

[0099] FIG4 is a schematic flow chart of a frequency domain measurement prediction method provided in another embodiment;

[0100] FIG5 is a schematic flow chart of a frequency domain measurement prediction method provided in another embodiment;

[0101] FIG6 is an interactive diagram of a frequency domain measurement prediction method provided in one embodiment;

[0102] FIG7 is an interactive diagram of a frequency domain measurement prediction method provided in another embodiment;

[0103] FIG8 is a structural block diagram of a frequency domain measurement prediction device provided in one embodiment;

[0104] FIG9 is a schematic diagram of the internal structure of a communication device provided in another embodiment. DETAILED DESCRIPTION

[0105] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0106] In the embodiments of this application, the term "and / or" describes the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally indicates that the associated objects are in an "or" relationship.

[0107] In the embodiments of the present application, the term "plurality" refers to two or more than two, and other quantifiers are similar.

[0108] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0109] The technical background of this application is first described below.

[0110] In mobile communication systems, due to the mobility of user equipment (UE), UE measurements are crucial. Whether in idle or connected state, UE measurements are required to reselect and switch the serving frequency or frequency combination, ensuring continuous service for the UE and ensuring service continuity and communication quality, enabling the network to provide better services. However, a UE generally has only one receiver, meaning it is impossible for the UE to simultaneously transmit and receive data in the serving cell and perform UE measurements. UE measurements refer to frequency-domain measurements of the UE, specifically Radio Resource Management (RRM) measurements performed by the UE on certain frequencies / bands.

[0111] In related technologies, to enable UEs to measure cells on different frequencies or different technology frequencies, or to measure synchronization signal blocks (SSBs) outside the initial bandwidth part (BWP) on the same frequency, mobile communication systems configure the UE to stop transmitting or receiving data on the operating frequency band for a certain period of time and instead measure RRM information on the desired frequency band. This period is called the measurement gap (GAP). Therefore, during the measurement gap, the UE cannot transmit or receive any information on the stopped frequency band.

[0112] Generally, whether to use the measurement GAP for measurement depends on the UE's capabilities, the UE's active BWP, and the current operating frequency. The following example illustrates this:

[0113] (1) For cross-frequency measurements based on SSB, if the UE reports measurement GAP requirement information, the mobile communication system can provide measurement GAP configuration based on this information. Otherwise, measurement GAP configuration is always provided in the following cases:

[0114] The UE only supports per-UE (single UE) measurement GAP; or,

[0115] The UE supports per-FR (single frequency range) measurement of GAP, and any serving cell is within the frequency range of the measurement object.

[0116] (2) For intra-frequency measurements based on SSB, if the UE reports measurement GAP requirement information, the measurement GAP configuration can be provided based on this information. Otherwise, the measurement GAP configuration is always provided in the following cases:

[0117] In addition to the initial BWP, if any BWP configured by the UE or reduced capability (RedCap) UE does not contain the frequency domain resources of the SSB associated with the initial downlink (DL) BWP, and for RedCap UE, no non-cell defining (NCD) SSB is configured for measurement of the serving cell.

[0118] In summary, in scenarios where GAP assistance is not required, the UE should be able to perform measurements without measuring the GAP. In scenarios where GAP assistance is required, it is stipulated that the UE can only perform measurements within the configured time for measuring the GAP.

[0119] In scenarios where GAP assistance is required, for frequencies or frequency bands where the UE must enable GAP to perform measurements, the UE must suspend communications on the current operating frequency or frequency band and adjust the receiver to the frequency or frequency band where the measurement needs to be performed. This causes data transmission interruption on the original operating frequency or frequency band, which may affect the overall UE data transmission and reception throughput. In addition, based on the GAP configuration (for example, the maximum repetition period of GAP is 160ms), the UE may have to wait for a long time before the next GAP time period for the next measurement, which may also cause measurement reporting delays due to the long measurement delay, leading to handover failure.

[0120] Based on this, the embodiments of the present application provide a frequency domain measurement prediction method, apparatus, and processor-readable storage medium that can solve the above-mentioned problems. Of course, the technical solutions provided in the embodiments of the present application are not limited to solving only the above-mentioned problems, but also have other technical effects, which can be specifically described in the following embodiments.

[0121] Before describing the embodiments of the present application, the application scenarios of the embodiments of the present application are described first.

[0122] The frequency domain measurement prediction method provided in the embodiment of the present application can be applied to any communication device, which can be a UE or a network-side device.

[0123] Figure 1 illustrates the internal structure of a communication device. The communication device includes a processor, memory, a bus interface, and a transceiver. The processor, memory, and transceiver are connected via the bus interface. The processor of the communication device is used to provide computing and control capabilities. The memory of the communication device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The transceiver is used to transmit and receive data under the control of the processor.

[0124] Figure 1 is merely an example of a structural diagram of a communication device and does not limit the communication device to which the present invention is applied. Actual communication devices may include more or fewer components than those shown in the figure, or may combine certain components or have a different component arrangement. For example, a database may be included to store data generated during the key generation process. An input / output interface may also be included for exchanging information between the processor and external devices. A communication interface may also be included for communicating with external terminals via a network connection, and so on.

[0125] Among them, the UE in the embodiment of the present application can be a device that provides voice and / or data connectivity to users, a handheld device with wireless connection function, or other processing devices connected to a wireless modem, etc., which includes but is not limited to smart terminals or network devices.

[0126] Smart terminals include, but are not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices include smart speakers, smart TVs, smart air conditioners, and smart car devices. Portable wearable devices include smart watches, smart bracelets, and head-mounted devices. Network devices include base stations, wireless access points, and other IoT devices.

[0127] Among them, the network side equipment may include but is not limited to base station nodes, communication nodes, core network nodes, operator nodes, etc. within the 3GPP system.

[0128] The frequency domain measurement prediction method provided in the embodiment of the present application is executed by the above-mentioned communication device, and is described below through a detailed embodiment.

[0129] In one embodiment, a frequency domain measurement prediction method is provided, as shown in FIG2 , and the method may include the following steps:

[0130] S101: Obtain a first actually measured RRM measurement result of a UE on a first frequency or a frequency combination.

[0131] In an embodiment of the present application, the first frequency point or frequency point combination may be some frequency points or frequency point combinations with less measurement overhead, or may be some frequency points or frequency point combinations that are easy to measure, wherein the frequency point combination may be multiple frequency point ranges or frequency bands in certain cells, etc.

[0132] In one embodiment, the first frequency point or frequency point combination includes at least one of the following: one or more same-frequency frequency points; one or more different-frequency frequency points; a combination of same-frequency frequency points and different-frequency frequency points.

[0133] The term "one or more co-frequency points" refers to one or more frequencies that are co-frequency with the UE's current operating frequency; the term "one or more off-frequency points" refers to one or more frequencies that are off-frequency with the UE's current operating frequency. A combination of co-frequency points and off-frequency points is a combination of different numbers of co-frequency points and different numbers of off-frequency points. In each combination of co-frequency points and off-frequency points, the number of co-frequency points and the number of off-frequency points can be in any relationship: one-to-many, many-to-one, one-to-one, or many-to-many.

[0134] In the embodiment of the present application, the first frequency point or frequency point combination includes at least one of one or more co-frequency points, one or more inter-frequency points, and a combination of co-frequency points and inter-frequency points, which can form multiple situations, including but not limited to the following situations:

[0135] A same frequency point;

[0136] Multiple co-frequency points;

[0137] An off-frequency point;

[0138] Multiple different frequency points;

[0139] One same-frequency point and one different-frequency point;

[0140] One same-frequency point and multiple different-frequency points;

[0141] Multiple same-frequency points and one different-frequency point;

[0142] Multiple same-frequency points and multiple different-frequency points;

[0143] a same-frequency point and a combination of a same-frequency point and a different-frequency point;

[0144] One same-frequency point and a combination of multiple same-frequency points and different-frequency points;

[0145] Multiple same-frequency points and a combination of one same-frequency point and one different-frequency point;

[0146] Multiple same-frequency points and multiple combinations of same-frequency points and different-frequency points;

[0147] an off-frequency point and a combination of an on-frequency point and an off-frequency point;

[0148] One different-frequency point and a combination of multiple same-frequency points and different-frequency points;

[0149] Multiple different-frequency points and a combination of the same-frequency point and different-frequency points;

[0150] Multiple different-frequency points and a combination of multiple same-frequency points and different-frequency points.

[0151] Based on the first frequency or frequency combination, the communication device obtains a first measured RRM measurement result of the UE on the first frequency or frequency combination. The communication device can be a UE or a network side device (abbreviated as NW).

[0152] For example, when the communication device is a UE, the UE may obtain the first measured RRM measurement result from its own measurement results, or may obtain the first measured RRM measurement result from the NW side. Similarly, when the communication device is a NW, the NW may obtain the first measured RRM measurement result from itself, or may obtain the first measured RRM measurement result from the UE side.

[0153] Taking the communication device as NW as an example, in one embodiment, the NW obtains the first measured RRM measurement result of the UE on the first frequency point or frequency point combination, including: receiving the first measured RRM measurement result periodically reported by the UE according to the measurement configuration.

[0154] The measurement configuration may be a measurement configuration sent by the NW to the UE, and the measurement configuration includes but is not limited to a measurement period and a reporting period for requiring the UE to perform the first measured RRM measurement result. After obtaining the first measured RRM measurement result, the UE periodically sends the first measured RRM measurement result to the NW according to the reporting period indicated in the measurement configuration. In this way, the UE can report the first measured RRM measurement result to the NW in a timely and effective manner.

[0155] The first measured RRM measurement result in the embodiment of the present application represents an RRM measurement result of the UE at the first frequency or frequency combination obtained by actual measurement. For example, the first measured RRM measurement result may be obtained by performing measurement using a GAP relaxation mechanism or based on a model-related GAP configuration.

[0156] In practical applications, the RRM measurement results can be differentiated from the parameter type dimension to include power measurement results, signal quality measurement results, etc. Optionally, the type of the first measured RRM measurement result includes at least one of the following: reference signal received power; reference signal received quality; signal to noise ratio.

[0157] Among them, the Reference Signal Receiving Power (RSRP) is the average value of the signal power received on all resource elements carrying the reference signal within a symbol. It refers to the power of a specific signal transmitted by the base station received by the UE and can be used to measure the signal reception quality of the UE. Among them, the Reference Signal Receiving Quality (RSRQ) is used to describe the quality of the received RF signal and is calculated by the ratio of the signal-to-noise ratio (SNR) and the signal-to-noise ratio (RSRQ). Among them, the SNR refers to the ratio of the strength of the received useful signal to the strength of the received interference signal (noise and interference), reflecting the link quality of the current channel.

[0158] In addition, RRM measurement results can be differentiated from the measurement object dimension, such as cell measurement results or beam measurement results. Therefore, in one embodiment, the first measured RRM measurement result includes at least one of the following: a serving cell measurement result; a neighboring cell measurement result; a combination of serving cell and neighboring cell measurement results; or a UE measurement result for a beam in a cell.

[0159] The serving cell measurement result refers to the measurement result of the cell currently serving the UE, for example, the signal quality of the serving cell, the signal-to-noise ratio of the serving cell, etc.

[0160] The neighboring cell measurement results represent the measurement results of neighboring cells of the cell currently serving the UE, and the combination of the serving cell and neighboring cell measurement results naturally represents the combination of the measurement results of the two cells.

[0161] The UE's measurement result for the beam in the cell may be a measurement result of the beam in its serving cell, for example, the power value of the beam, etc.

[0162] S102: Derivation is performed using the first measured RRM measurement result as an input of a model to obtain a predicted RRM measurement result of the UE at a second frequency point or a frequency point combination.

[0163] Based on the first measured RRM measurement result obtained by the above-mentioned communication device, it is used as input data for deriving the RRM measurement result of the UE at the second frequency point or frequency point combination, and is input into the model for derivation to obtain the predicted RRM measurement result.

[0164] The predicted RRM measurement result is the RRM measurement result of the UE on the second frequency point or frequency point combination.

[0165] The second frequency point or frequency point combination may be completely different from the first frequency point or frequency point combination, or the second frequency point or frequency point combination may overlap with the first frequency point or frequency point combination.

[0166] In one embodiment, the second frequency point or frequency point combination includes at least one of the following: one or more same-frequency points; one or more different-frequency points; one or more different-technology points; a combination of at least two of the same-frequency points, different-frequency points and different-technology points.

[0167] In the embodiment of the present application, the same-frequency frequency point is the same frequency relative to the current working frequency point of the UE, and the different-frequency frequency point is the different frequency relative to the current working frequency point of the UE.

[0168] The one or more co-frequency points in the second frequency point or frequency point combination refer to co-frequency points or frequency point combinations that are different from the first frequency point or frequency point combination. For example, taking frequency as an example, if the co-frequency points in the first frequency point are co-frequency point 1 and co-frequency point 3, then the co-frequency points in the second frequency point refer to co-frequency point 2, co-frequency point 5, and co-frequency point 6, etc. Similarly, the different-frequency points in the one or more different-frequency points in the second frequency point or frequency point combination may be different from the different-frequency points in the first frequency point or frequency point combination.

[0169] However, it should be noted that since the first frequency point or frequency point combination is the frequency point of the model input data, and the second frequency point or frequency point combination is the frequency point of the model output data, there are some cases where the same frequency point (different-frequency frequency points or same-frequency frequency points) corresponds to different input data and output data combinations of the model. In these cases, the frequency points in the second frequency point or frequency point combination may be the same as some of the frequency points in the first frequency point or frequency point combination.

[0170] For example, the first frequency point corresponding to the input data of the model is the same-frequency point A, and the second frequency point corresponding to the output data of the model can be the different-frequency point B; or, the first frequency point corresponding to the input data of the model is the different-frequency point B, and the second frequency point corresponding to the output data of the model can be the different-frequency point C; or, the first frequency point corresponding to the input data of the model is the different-frequency point C, and the second frequency point corresponding to the output data of the model can be the same-frequency point A.

[0171] As can be seen, in the three cases listed above, the same-frequency point A, different-frequency point B, and different-frequency point C can be either the frequency point in the first frequency point or the frequency point in the second frequency point. However, because these same frequency points correspond to frequency point combinations of different input and output data of the model, they are valid.

[0172] The above merely lists several frequency combinations corresponding to input and output data, and is not intended to limit the first frequency or frequency combination, and the second frequency or frequency combination in the embodiments of the present application.

[0173] The one or more different-technology frequencies refer to one or more frequencies using a different radio access technology than the UE's current serving cell. For example, the types of radio access technologies include, but are not limited to, Bluetooth, Wi-Fi, 3G, 4G, LTE, and 5G. The first frequency may be a frequency using the 5G radio access technology, while the second frequency may be a frequency using the 4G radio access technology.

[0174] Among them, the combination of at least two of the same-frequency frequency points, different-frequency frequency points and different-technology frequency points is a combination of any two of these frequency points, or a combination of three, and the embodiments of the present application do not limit this.

[0175] In summary, in the embodiments of the present application, the first frequency or frequency combination corresponding to the model input data and the second frequency or frequency combination corresponding to the model output data can be any combination of two frequencies. That is, the RRM measurement results at the first frequency or frequency combination, obtained by measuring with less overhead, are used as the input data of the model. Through model derivation, the RRM measurement results at the second frequency or frequency combination can be predicted. This is called a predicted RRM measurement result.

[0176] In this embodiment of the present application, the type and content of the predicted RRM measurement result are the same as those of the first measured RRM measurement result. Based on this, in one embodiment, the predicted RRM measurement result includes at least one of the following: a serving cell measurement result; a neighboring cell measurement result; a combination of serving cell and neighboring cell measurement results; or a UE measurement result for a beam in the cell. In one embodiment, the type of the predicted RRM measurement result includes at least one of the following: reference signal received power; reference signal received quality; or signal-to-noise ratio.

[0177] The difference between the predicted RRM measurement result and the first measured RRM measurement result lies in the corresponding frequencies or frequency combinations. That is, the first measured RRM measurement result is an RRM measurement result at a first frequency or frequency combination, and the predicted RRM measurement result is an RRM measurement result at a second frequency or frequency combination. However, for the type of the predicted RRM measurement result and its specific content, please refer to the description of the first measured RRM measurement result above and will not be repeated here.

[0178] The communication device uses the first measured RRM measurement result as an input to a model to derive a predicted RRM measurement result, wherein the model is pre-deployed in the communication device.

[0179] In one embodiment, the model includes but is not limited to an artificial intelligence (AI) model or a machine learning (ML) model. Using the AI / ML model can improve the performance of the communication system in which the communication device is located.

[0180] In an embodiment of the present application, before deploying the AI / ML model on the communication device, measurement results of the UE in different cells, different frequencies, different frequency bands, or different technical frequencies, etc. can be obtained, and machine learning can be performed based on the correlation between these measurement results to obtain the corresponding AI / ML model.

[0181] Based on the obtained AI / ML model, the RRM measurement results of fewer cells, frequencies, or frequency bands (i.e., the first frequency or frequency combination) are used to infer the RRM measurement results of other cells, frequencies, frequency bands, or technologies (i.e., the second frequency or frequency combination), so that less measurement overhead is consumed when the UE performs same-frequency, different-frequency, or different-technology measurements.

[0182] In an embodiment of the present application, the AI / ML model may be a unilateral model, that is, the model is deployed only on one side, such as the NW side or the UE side; or, the AI / ML model may be a bilateral model, that is, the model is deployed on the NW and UE sides, and collaboration between the two sides is required for complete reasoning. Among them, the unilateral model deployed on the UE side, or the part of the bilateral model that belongs to the UE side, may be trained and generated by the UE manufacturer, chip manufacturer or third-party service provider. The unilateral model deployed on the NW side, or the part of the bilateral model that belongs to the NW side, may be trained and generated by network manufacturers / operators, base station / core network manufacturers or third-party service providers.

[0183] For example, if the AI / ML model is deployed on the UE side, the UE uses the first measured RRM measurement result as the input of the model for derivation to obtain the predicted RRM measurement result of the UE at the second frequency point or frequency combination. If the AI / ML model is deployed on the NW side, the NW uses the first measured RRM measurement result as the input of the model for derivation to obtain the predicted RRM measurement result of the UE at the second frequency point or frequency combination.

[0184] In addition, regardless of whether the AI / ML model is deployed unilaterally or bilaterally, the NW and / or UE can perform life cycle management (LCM) on the model, enabling the UE and NW to collaborate on complete reasoning of the AI / ML model.

[0185] As shown in Figure 3, a schematic diagram of the framework of the main modules of LCM for implementing LCM is shown. Among them, the main modules for implementing LCM include: data collection module, model training module, reasoning module, management module and model storage module.

[0186] In actual application, the data collection module can collect training data, monitoring data, inference data, etc. The training data is used for the model training module to train the model, the monitoring data is used for the management module to monitor or manage the model, and the inference data is used for the inference module to perform data inference. For example, in the embodiment of the present application, the first measured RRM measurement result is used as the input of the model for inference. The model storage module is used to store the model trained or updated by the model training module, and based on the model transfer request issued by the management module, the latest stored model is transferred or delivered to the inference module, so that the inference module infers the result. Among them, after the inference module infers the result, it can be output to the management module, so that the management module can refer to these data to manage the model, and the management module can provide model performance feedback to the model training module based on the management result or initiate a retraining request.

[0187] The modules in the LCM device shown in Figure 3 are merely examples. These modules are not limited to specific devices and can be deployed individually or in combination on at least one of the following: the UE, the NW, a third-party server, etc., depending on actual needs. For example, the data collection module and model storage module can be deployed on the UE, the NW, and a third-party server, respectively, while the model training module, inference module, and management module can be deployed on the NW.

[0188] In the frequency domain measurement prediction method provided in the embodiment of the present application, the communication device obtains the first measured RRM measurement result of the UE at the first frequency point or frequency point combination, and uses the first measured RRM measurement result as the input of the model for deduction to obtain the predicted RRM measurement result of the UE at the second frequency point or frequency point combination. This is equivalent to using a pre-trained model to infer the RRM measurement result of the UE at the first frequency point or frequency point combination based on the RRM measurement result of the UE at the second frequency point or frequency point combination, so that the RRM measurement of the UE at the second frequency point or frequency point combination is predicted by the model, without consuming measurement overhead such as measurement GAP and UE power, thereby greatly reducing the measurement time overhead consumed by the UE to perform RRM measurements on some frequency points, and ensuring the UE data transmission and reception throughput.

[0189] In addition, although the UE actually measures the first measured RRM measurement result on the first frequency point or frequency point combination, since the first measured RRM measurement result is only the measurement result on a part of the frequency points or frequency point combinations, it can be understood that compared with the measurements on many cells, frequencies, frequency bands and frequency point combinations that need to be measured, the first frequency point or frequency point combination only represents a small part of the many measurements that need to be measured. Therefore, compared with the measurement time overhead spent on all cells, frequencies, frequency bands and frequency point combinations that need to be measured using measurement GAP measurements, the measurement time overhead consumed by the measurement on the first frequency point or frequency point combination is only a small part. Therefore, the embodiment of the present application infers the RRM measurement results of other cells or frequencies or frequency bands or technologies through RRM measurement results on fewer cells or frequencies or frequency bands, thereby solving the problem of excessive measurement overhead when the UE performs frequency domain measurements, and reducing the time overhead consumed for frequency domain measurements based on existing measurement results.

[0190] In the above embodiment, the first RRM measurement result is used as input data for the model to derive the predicted RRM measurement result. In practical applications, to increase the accuracy of the derivation of the predicted RRM measurement result, the derivation can be performed simultaneously with relevant information of the UE. Based on this, in one embodiment, as shown in FIG4 , the above derivation using the first measured RRM measurement result as input to the model to obtain the predicted RRM measurement result of the UE at the second frequency point or frequency point combination includes steps S201 and S202.

[0191] S201, obtaining UE related information.

[0192] UE related information can be understood as information related to UE attributes or working status, etc.

[0193] Taking the working status of a UE as an example, in one embodiment, the UE-related information includes at least one of the following: UE location information; UE speed information. That is, the UE-related information may be either the UE location information or the UE speed information, or both.

[0194] The location information of the UE may be some information related to the location, such as coordinates, orientation, longitude and latitude, direction, location, relative position relationship, etc.

[0195] In one embodiment, the location information of the UE includes at least one of the following: the geographic location coordinates of the UE; location-related information obtained by the UE based on a positioning method; and location-related information obtained by a network-side device based on a positioning method.

[0196] In an embodiment of the present application, the geographic location coordinates of the UE may include but are not limited to the horizontal coordinate of the UE based on the Cartesian coordinate system, the vertical coordinate of the UE based on the Cartesian coordinate system, the elliptical coordinates of the UE, spherical coordinates, etc., wherein the spherical coordinates can be coordinates in a three-dimensional coordinate system described by radial distance, azimuth and polar angle, or they can be first determined based on Cartesian coordinates and then converted into spherical coordinates as needed.

[0197] In addition to the geographic location coordinates of the UE, the location information of the UE may also be location-related information obtained using some positioning methods, for example, location-related information obtained based on various positioning methods defined within / outside the 3GPP system (such as positioning methods in protocols such as TS37.355). The above may be that the UE obtains the location-related information of the UE using a positioning method, or that the NW obtains the location-related information of the UE using a positioning method, and this embodiment of the present application is not limited to this.

[0198] In one embodiment, the UE speed information includes: UE actual speed information, and / or various parameter information related to the UE speed.

[0199] The actual UE speed information is the speed information of the UE obtained by actual measurement, and can be the movement speed or rotation speed of the UE in motion, or the relative speed between the UE and other objects in a stationary state.

[0200] Various parameter information related to UE speed may include but is not limited to some parameter values ​​converted based on the actual speed information of the UE, such as average speed, acceleration, displacement, and even the classification results of UE high speed, medium speed and low speed obtained by using some calculation methods based on the actual speed information of the UE.

[0201] The various parameters related to the UE speed listed above and the detailed content of the UR location information are just examples. In actual applications, the UE speed information and location information are not limited thereto.

[0202] In one embodiment, obtaining UE-related information in the embodiment of the present application includes: obtaining UE-related information collected by the UE, and / or UE-related information collected by a network-side device.

[0203] That is, the UE-related information may be collected by the UE, the NW, or both.

[0204] For example, when the communication device is a UE, the UE can obtain UE-related information by directly collecting UE-related information itself, or it can receive part of the UE-related information from the NW; similarly, when the communication device is a NW, the NW can obtain UE-related information by directly collecting part of the UE-related information itself, or it can receive UE-related information from the UE side.

[0205] In the case where the NW receives UE-related information from the UE side, in one embodiment, the NW obtains the UE-related information collected by the UE, which may be the UE-related information periodically reported by the UE according to the measurement configuration.

[0206] This measurement configuration is the same as the aforementioned NW obtaining the first measured RRM measurement result from the UE side. It can be issued by the NW to the UE, and includes but is not limited to requiring the UE to perform UE-related information measurement period, reporting period, etc., so that after obtaining the UE-related information, the UE periodically sends the UE-related information to the NW according to the reporting period indicated in this measurement configuration, thereby greatly ensuring the timeliness and effectiveness of the UE reporting the UE-related information to the NW.

[0207] S202: Use the first measured RRM measurement result and UE-related information as inputs of a model to perform derivation to obtain a predicted RRM measurement result.

[0208] Based on the above-obtained UE-related information, the communication device uses the UE-related information and the obtained first measured RRM measurement result as inputs of the model to perform derivation, thereby obtaining a predicted RRM measurement result.

[0209] Similarly, if the communication device is a UE, the UE performs this derivation process to obtain the predicted RRM measurement result; if the communication device is a NW, the NW performs this derivation process to obtain the predicted RRM measurement result.

[0210] In the frequency domain measurement prediction method provided in the embodiment of the present application, the communication device obtains relevant information of the UE, uses the relevant information of the UE and the first measured RRM measurement result of the UE at the first frequency point or frequency point combination as the input of the model for deduction, and obtains the predicted RRM measurement result of the UE at the second frequency point or frequency point combination. Since the UE-related information can reflect the most real and latest states of the UE, the communication device further considers the relevant information of the UE to perform the derivation process on the basis of the first measured RRM measurement result, so that the final predicted RRM measurement result is also closer to the actual needs of the UE, thereby increasing the accuracy of the predicted RRM measurement result derived by the model. In addition, the RRM measurement of the UE at the second frequency point or frequency point combination is predicted by the model, without consuming measurement overhead such as measurement GAP and UE power, thereby greatly reducing the measurement time overhead consumed by the UE to perform RRM measurements on some frequency points, and ensuring the data transmission and reception throughput of the UE.

[0211] Based on the predicted RRM measurement results of the UE on the second frequency or frequency combination derived in the above embodiments, the communication device can perform different operations based on the predicted RRM measurement results. The following describes some embodiments of some operations that can be performed by the communication device after obtaining the predicted RRM measurement results of the UE on the second frequency or frequency combination.

[0212] In one embodiment, the communication device may make a handover decision for the UE based on the predicted RRM measurement result.

[0213] Since the RRM measurement results are frequency domain measurements of the UE's current serving cell, they can reflect the UE's current service processing quality. Based on this, the UE's handover decision can be made based on this. The handover decision here refers to whether the UE's serving cell, frequency point, or frequency band should be switched.

[0214] For example, the NW makes a handover decision for the UE based on the predicted RRM measurement results.

[0215] Of course, in addition to making a handover decision for the UE based on the predicted RRM measurement result, the communication device can also use it independently based on actual needs.

[0216] In addition, when the communication device is a UE, in addition to using it by itself, in one embodiment, the UE may send the predicted RRM measurement result to the network side device.

[0217] The UE sends the predicted RRM measurement result to the NW, so that the NW side can promptly share the RRM measurement result of the UE on the second frequency point or frequency point combination, which facilitates the NW to make a subsequent handover decision or perform other operations based on the measurement result.

[0218] In addition, because the model is deployed on the communication device, after the communication device derives the predicted RRM measurement results, the model can also be monitored and managed. As shown in Figure 5, in one embodiment, the process of monitoring and managing the model by the communication device includes the following:

[0219] S301, monitoring the model.

[0220] In the embodiment of the present application, the purpose of the communication device monitoring and managing the model is to adjust the reasoning performance of the model to make the model performance better. Therefore, in order to manage the model more reasonably, the communication device can perform different management operations based on the monitoring results of the model.

[0221] When the model is monitored, at least one of the predicted RRM measurement result, the performance indicator value, and the actual RRM measurement result of the UE at the second frequency point or the frequency point combination may be monitored.

[0222] The performance indicator value is a key performance indicator (KPI), which can be understood as some KPIs related to the performance of the model. In one embodiment, the performance indicator value includes at least one of the following: radio link failure rate; handover failure rate; ping-pong handover probability; and interruption duration.

[0223] These performance indicators can be preset or configured based on actual network needs. Examples include the radio link failure rate, handover failure rate, ping-pong handover probability, and interruption duration, which can reflect model performance. By combining these indicators for model monitoring, we can accurately quantify model performance and assess whether the model meets performance standards.

[0224] In one embodiment, the above-mentioned monitoring of the model execution includes at least one of the following: comparing the performance indicator value predicted by the application model with the preset / network configured performance indicator value; comparing the performance indicator value predicted by the application model with the actually measured performance indicator value; comparing the performance indicator value predicted by the application model with the performance indicator value predicted by other different models; comparing the predicted RRM measurement result with the second actually measured RRM measurement result at the second frequency point / frequency point combination; comparing the predicted RRM measurement result with the RRM measurement result at the second frequency point / frequency point combination predicted by other different models.

[0225] The above-mentioned model monitoring methods can be divided into two categories for explanation.

[0226] Category 1: Monitoring methods related to performance indicators:

[0227] (1) Compare the performance indicator values ​​predicted by the application model with the performance indicator values ​​of the preset / network configuration.

[0228] The performance indicator value predicted by the application of the model refers to the performance indicator value calculated when the model is used for inference prediction. For example, when the model is used to predict the RRM measurement results at the second frequency or frequency combination, at least one of the radio link failure rate, handover failure rate, ping-pong handover probability, and interruption duration is calculated.

[0229] The performance indicator values ​​of the preset / network configuration naturally refer to the values ​​of the wireless link failure rate, handover failure rate, ping-pong handover probability, and interruption time length of the preset / network configuration.

[0230] For example, the radio link failure rate is calculated using the model to predict the RRM measurement results for the second frequency or frequency combination and compared with the preset / network-configured radio link failure rate. The comparison logic for several other performance indicators is similar.

[0231] (2) Compare the performance index values ​​predicted by the application model with the performance index values ​​obtained by actual measurement.

[0232] The measured performance indicator value refers to at least one of the radio link failure rate, handover failure rate, ping-pong handover probability, and interruption time length calculated when the RRM measurement results are actually measured on the second frequency point or frequency point combination.

[0233] Taking the handover failure rate as an example, the handover failure rate calculated using the model-predicted RRM measurement results at the second frequency or frequency combination is compared with the handover failure rate calculated using the actual RRM measurement results at the second frequency or frequency combination. The comparison logic for the other performance indicators is similar.

[0234] (3) Compare the performance index values ​​predicted by the application model with the performance index values ​​predicted by other different models.

[0235] Other different models refer to any models that are different from the communication device model in the embodiment of the present application, that is, comparison between performance indicator values ​​predicted using different models.

[0236] For example, taking the ping-pong handover probability as an example, the ping-pong handover probability calculated when the model in the embodiment of the present application is used to predict the RRM measurement results at the second frequency point or frequency combination is compared with the ping-pong handover probability calculated when the RRM measurement results at the second frequency point or frequency combination are predicted using a different model. The comparison logic for the other performance indicators is the same.

[0237] Category 2: Monitoring methods related to RRM measurement results:

[0238] (1) Compare the predicted RRM measurement result with the second measured RRM measurement result at the second frequency point / frequency point combination.

[0239] The predicted RRM measurement result is the RRM measurement result on the second frequency point or frequency point combination predicted by the model in the embodiment of the present application. In addition, the second measured RRM measurement result of the UE on the second frequency point / frequency point combination can be obtained by adopting some actual measurement methods. The RRM measurement result actually measured on the second frequency point / frequency point combination can be an RRM measurement result measured based on a relaxed measurement GAP or based on a model-related GAP configuration.

[0240] The predicted RRM measurement result is then compared with the second measured RRM measurement result, that is, the difference between the predicted result and the measured result is compared.

[0241] (2) Compare the predicted RRM measurement results with the RRM measurement results at the second frequency point / frequency point combination predicted by other different models.

[0242] Other models different from the embodiments of the present application are used to infer and predict the RRM measurement results at the second frequency point or frequency point combination, and the RRM measurement results predicted by these other models are compared with the predicted RRM measurement results predicted by the model in the embodiment of the present application, that is, the comparison is the difference between the RRM measurement results obtained by prediction using different models.

[0243] In all monitoring methods in the embodiments of the present application, the specific method of comparison can be to calculate the difference and then compare the difference with a preset threshold to obtain a monitoring result. For example, whether the difference between the two exceeds the threshold. The embodiments of the present application set multiple monitoring methods based on the two dimensions of performance indicators and RRM measurement results. In this way, by comparing data of different dimensions calculated in multiple ways, the monitoring results of the model are obtained, making the monitoring of the model more comprehensive and the monitoring results more accurately reflecting the true capabilities of the model.

[0244] S302: Perform management operations on the model according to the monitoring results of the model and / or the overhead saving information related to the model.

[0245] Based on the monitoring result of the model obtained above, the communication device performs a management operation on the model according to the monitoring result and / or overhead saving information related to the model.

[0246] Among them, if the communication device is a NW, after performing monitoring and obtaining monitoring results, the NW can perform model management based on the monitoring results and / or the overhead saving information related to the model. However, if the communication device is a UE, after performing monitoring and obtaining monitoring results, the UE can perform model management based on the monitoring results and / or the overhead saving information related to the model, or it can send the monitoring results to the NW so that the NW can perform model management based on the monitoring results and / or the overhead saving information related to the model.

[0247] The overhead saving information refers to the overhead saving when the model is used to derive the RRM of the UE on the desired frequency or frequency combination, compared to when the model is not used to derive the RRM of the UE on the desired frequency or frequency combination. In one embodiment, the overhead saving information includes at least one of the following: interruption time saving; measurement gap usage time; measurement gap time saving; and measurement required UE power saving.

[0248] The interruption time refers to the duration of UE service interruption due to measuring the RRM of the UE on the desired frequency or frequency combination. Naturally, the interruption time saving refers to the interruption time saved by using the model to deduce the RRM of the UE on the desired frequency or frequency combination compared to not using the model deduction. For example, when using the model to deduce the RRM of the UE on the desired frequency or frequency combination, the UE's interruption time is T1, and when the model is not used for deduction, the UE's interruption time is T2, then the interruption time saving is T1-T2. Of course, because the UE may not have service interruption when the model deduction is used in the embodiment of the present application, T1 can be 0 in actual applications.

[0249] The measurement GAP is the gap between the RRM performance of the UE at the desired frequency or frequency combination measured using the measurement GAP in a conventional manner. Because the embodiments of the present application use a model to derive the RRM performance of the UE at the desired frequency or frequency combination, the time saved from the conventional measurement GAP is also considered time saved. Similarly, the saved measurement GAP time can be the entire measurement GAP usage time, or the difference between the time T3 spent using the model derivation method and the measurement GAP usage time T4.

[0250] In practical applications, performing UE RRM measurements also consumes a certain amount of UE power, so UE power also belongs to measurement overhead. Therefore, the overhead time saved in the embodiments of the present application also includes the UE power savings required for measurement. Following the same logic as above, the difference between the UE power consumed when using the model to derive the UE RRM on the desired frequency or frequency combination and the power consumed when not using the model derivation is the UE power savings required for measurement.

[0251] The above-mentioned overhead saving information not only involves the saving information related to measuring GAP, but also involves the UE's service interruption saving time and UE's power saving. It comprehensively covers the multiple types of overhead involved in UE frequency domain measurement, ensuring the integrity and accuracy of the overhead saving information, so that the overhead saving information can provide more accurate basis and guidance for the management of the model.

[0252] When managing a model, the communication device can perform a variety of management operations. That is, different management operations are determined by monitoring results of different models and overhead saving information related to different models.

[0253] The following is an introduction to some management operations performed on the model: In one embodiment, the management operation includes performing at least one of the following operations: activating the model or model-related functions; deactivating the model or model-related functions; converting the model into other models with better performance indicators and / or RRM measurement results; converting model-related functions into other model-related functions with better performance indicators and / or RRM measurement results; rolling back the model or model-related functions; updating the model or model-related functions; retraining the model; and fine-tuning the model.

[0254] In the embodiments of the present application, the management of the model can be the management of the entire model itself or the management of some functions within the model. For example, the model or model-related functions can be activated, deactivated, converted, rolled back, updated, retrained, fine-tuned, etc.

[0255] Activating and deactivating a model or model-related functions are two relative operations. An activated function is now usable, while deactivating it stops the normal use of the corresponding function. Of course, activation and deactivation of a model or model-related functions can be performed for different functions, for example, activating function A of a model and deactivating function B of the model.

[0256] Converting a model to another model with improved performance metrics and / or RRM measurement results; and converting model-related functions to other model-related functions with improved performance metrics and / or RRM measurement results. The difference between these two management operations is whether the model itself is converted or only part of the model's functions. The performance metrics and / or inferred RRM measurement results of the converted model or model-related functions are improved compared to those before the conversion, effectively upgrading the model's inference performance.

[0257] Rolling back a model or model-related functionality means returning it to the previous version. Updating a model or model-related functionality not only includes performance upgrades but can also include updating old parameters, old information, and even storage locations, sizes, and names. Retraining and fine-tuning a model can also involve retraining or fine-tuning all or part of a model's functionality.

[0258] Fine-tuning a model can make its parameters more suitable for a specific task. For example, full fine-tuning and partial fine-tuning can be performed. Full fine-tuning involves fine-tuning the entire pre-trained model, including all model parameters. All layers and parameters of the pre-trained model are updated and optimized to meet the requirements of the target task. Partial fine-tuning involves updating only the top layer or a few layers of the model during fine-tuning, while keeping the underlying parameters of the pre-trained model unchanged. The goal is to preserve the general knowledge of the pre-trained model while fine-tuning the top layers to adapt it to a specific task.

[0259] The management operations in the embodiments of the present application are rich and varied, and different management operations are performed based on different needs, making the management of the model more flexible. At the same time, targeted management can also improve the reasoning performance of the model.

[0260] In an embodiment of the present application, the model is monitored, and management operations are performed on the model based on the monitoring results of the model and / or the cost saving information related to the model. In this method, when performing management operations on the model, not only the monitoring results of the model are considered, but also the cost saving information related to the model is considered. Both of these are some information that can reflect the reasoning performance of the model. In this way, multi-dimensional information comprehensively considers the current reasoning performance level of the model and performs corresponding management, which will make the management of the model more reasonable and more accurate.

[0261] The aforementioned embodiments have been based on the premise that the model is deployed on the communication device, and the model usage process and the model supervision process have been described. The following will continue to explain the model training process in detail through some embodiments.

[0262] In one embodiment, the frequency domain measurement prediction method provided in the present application further includes: training a model based on training data; and / or receiving a model sent by a training node, where the model is trained by the training node based on the training data.

[0263] The model training process is completed before the model is used, so the model training in the embodiment of the present application can be performed at any time before the above-mentioned derivation using the first measured RRM measurement result as the input of the model.

[0264] In the embodiment of the present application, although the model is deployed on the communication device, the training of the model is not necessarily completed by the communication device itself. It can be trained by the communication device itself, or by other training nodes, or jointly trained by the communication device and other training nodes.

[0265] In one embodiment, the training node includes at least one of the following: a third-party server; a base station node; an operator node; a UE; or a core network node.

[0266] Among them, the third-party server can generally refer to the server of the model supplier (third-party service provider), for example, an OTT (Over-The-Top) server.

[0267] The training node can be any one of the base station nodes, operator nodes, UE or core network nodes, or multiple nodes that perform training together. For example, one or more of the servers that train the model in the UE manufacturer, chip manufacturer, network manufacturer, operator, base station manufacturer, or core network manufacturer.

[0268] Taking the communication device's own training model as an example, the communication device trains the model based on the training data, and after the model is trained, the model can be directly installed to use the model for reasoning.

[0269] The communication device training model must first obtain training data. For example, if the communication device is a UE, the UE can collect some training data itself and receive some training data from the NW. If the communication device is a NW, the NW can also collect some training data itself and receive some training data from the UE.

[0270] Taking other training node training models as an example, in actual applications, the training node first obtains training data, trains the model based on the obtained training data, and then passes the model to the communication device after the model training is completed, so that the communication device deploys the model and uses the model for reasoning.

[0271] In the embodiment of the present application, the nodes of the training model are diverse, and the model is not limited to one device. Instead, it is completed by different training nodes based on actual conditions and finally deployed to the communication device that needs to perform inference. This makes the training and deployment methods of the model more flexible, and improves the applicability of the scenarios in which the model is used for inference and prediction in the embodiment of the present application.

[0272] The training model requires a large amount of sample data, so some actual RRM measurement results of the UE at different frequencies or frequency combinations, as well as actual related information of the UE, can be collected as training data. Of course, the UE-related information in the training data may or may not exist, depending on actual needs. Based on this, in one embodiment, the training data includes at least one of the following: the actual RRM measurement results of the UE at the third frequency / frequency combination; the actual RRM measurement results of the UE at the fourth frequency / frequency combination; UE-related information collected by the UE; and UE-related information collected by network-side equipment.

[0273] In actual applications, actual RRM measurement results of the UE at different frequencies or frequency combinations are collected. Since real data is collected, the collection time of these actual RRM measurement results in the training data is historical time compared to the current time.

[0274] Based on this, in the embodiment of the present application, the third frequency point or frequency point combination and the fourth frequency point or frequency point combination, and the aforementioned first frequency point or frequency point combination and the second frequency point or frequency point combination, these frequency points can be the same frequency point or different frequency points. The difference is that the measured RRM measurement results of the same frequency point among these frequency points correspond to the RRM at different times. For example, taking frequency point S and frequency point F as an example, if the first frequency point is frequency point S and the second frequency point is frequency point F, then: the communication device collects the measured RRM of frequency point S at the current time, and uses it as the input of the model for derivation to obtain the predicted RRM of frequency point F. However, if the third frequency point is frequency point S and the fourth frequency point is frequency point F, then: the measured RRM of frequency point S at historical time and the measured RRM of frequency point F at historical time are collected, and these are all used as training data to train the model.

[0275] In one embodiment, the third frequency point or frequency point combination includes at least one of the following: one or more same-frequency points; one or more different-frequency points; a combination of same-frequency points and different-frequency points. The fourth frequency point or frequency point combination includes at least one of the following: one or more same-frequency points; one or more different-frequency points; one or more different-technology points; a combination of at least two of same-frequency points, different-frequency points, and different-technology points.

[0276] Aside from the timing differences, the specific description of the third frequency or frequency combination can be found in the description of the first frequency or frequency combination in the aforementioned embodiment. Similarly, apart from the timing differences, the specific description of the fourth frequency or frequency combination can be found in the description of the second frequency or frequency combination in the aforementioned embodiment. These details will not be repeated here.

[0277] When training the model, only the UE's measured RRM measurement results on the third frequency / frequency combination and the UE's measured RRM measurement results on the fourth frequency / frequency combination can be used as training data for training. However, there is no case where only UE-related information is used as training data. That is, the UE-related information in the training data is optional. The UE-related information can be collected by the UE itself or by the NW. The specific description of the UE-related information can also be found in the description of the aforementioned embodiment and will not be repeated here.

[0278] In an embodiment of the present application, the training data of the training model includes multiple types of real information, and this real information can be RRM measurement results of multiple frequencies or frequency combinations or UE-related information collected by different devices, which makes the data type of the training data very rich and diverse, greatly improves the diversity of the training data, and ensures the reasoning performance and robustness of the trained model.

[0279] In the embodiment of the present application, the communication device uses a model to derive / predict the RRM measurement results on the second frequency point or frequency combination, which can save the use of measurement GAP, reduce data interruption time, and thus improve throughput. In addition, the embodiment of the present application also involves a process of measurement GAP recommendation / preference and network GAP configuration / indication issuance, which can be used in conjunction with the model in the embodiment of the present application. For example, the process is for the measurement GAP related information recommended or preferred by the model used, as well as the GAP parameters configured by the network in the actual process or the saved measurement GAP or related instructions.

[0280] First, the process of configuring the GAP parameters or saving the measured GAP or related indications in the actual process of the network is described. In one embodiment, the embodiment of the present application further includes: the network side device sends the model-related GAP configuration or related indications to the UE.

[0281] The NW may send the model-related GAP configuration or related instructions to the UE while the communication device is using the model.

[0282] The NW sends the model-related GAP configuration or related indication to the UE to inform the UE whether the RRM measurement on some frequency points or frequency combinations is performed using model prediction or actual configured GAP measurement, or some related GAP indications when the model prediction is used, thereby ensuring that the UE knows the actual configured GAP parameters and related indications when the model is used for RRM prediction on the second frequency point or frequency combination in a timely manner.

[0283] In one embodiment, the network-side device sends the model-related GAP configuration or related indication to the UE via any of the following signaling: radio resource control signaling; media access control layer control element; physical layer instruction. That is, the NW may send the model-related GAP configuration or indication to the UE via radio resource control signaling (RRC) signaling, media access control layer control element (MAC CE) or physical layer instruction. By using some existing signaling to transmit the model-related GAP configuration or related indication, while meeting the UE's own quality of service requirements, the limited wireless network resources are fully utilized as much as possible to ensure that the information can effectively reach the UE.

[0284] In one embodiment, the GAP configuration or related indication in the embodiment of the present application includes at least one of the following: whether the GAP configuration recommended or preferred by the UE is adopted; the adoption status of the GAP configuration recommended or preferred by the UE; model-related use or non-use of GAP for the same frequency point; model-related use or non-use of GAP for different frequency points; model-related use or non-use of GAP for different technology frequency points; model-related GAP relaxation on a single or multiple frequency points; model-related GAP configuration.

[0285] The content of the GAP configuration or related instructions sent by the NW to the UE may include whether the GAP configuration recommended or preferred by the UE is adopted and / or the adoption situation. The adoption situation may be whether all or part of the GAP configuration recommended or preferred by the UE is adopted. For partial adoption, the UE is informed of which parts are adopted, for example, the frequency point identifier is used to indicate the frequency points for GAP use / non-use / relaxation. The frequency point identifier may be composed of numbers and / or letters, which is not limited in the embodiments of the present application. Of course, the GAP recommended or preferred by the UE may be sent to the NW in advance by the UE, or it may be determined by the NW based on historical data instead of being sent to the NW in advance by the UE.

[0286] The content of the GAP configuration or related instructions sent by the NW to the UE may also include model-related GAP conditions specific to some different frequency points, such as the use or non-use of GAP on single / multiple same-frequency frequency points, different-frequency frequency points, different-technology frequency points, etc.; it may also include which GAP relaxations are used on single / multiple frequency points; and the frequency identifiers of these frequency points indicating GAP use, GAP non-use, or GAP relaxation. Among them, GAP relaxation can be understood as GAP saving. For example, GAP relaxation includes but is not limited to configuring a GAP repetition period longer than the current GAP repetition period, configuring a GAP length shorter than the current GAP repetition period, and puncturing / omitting some GAP positions, etc. For example, if the NW configures that GAP relaxation is not used on frequency point 3, it means that the RRM measurement of frequency point 3 will not be actually measured, but will be derived and predicted using the model.

[0287] Among them, the above-mentioned different frequency points may refer to the derived frequency points of the application model or the input frequency points. For example, these frequency points may be the second frequency point or frequency point combination in the aforementioned embodiment, or the first frequency point or frequency point combination.

[0288] When these frequency points are derived frequency points of the model, then, taking the second frequency point as frequency point 1 as an example:

[0289] If GAP is not used on frequency 1, the RRM measurement results on frequency 1 are derived through model inference. Alternatively, if GAP is used on frequency 1, the RRM measurement results on frequency 1 are derived through traditional non-model inference methods. Alternatively, if GAP is relaxed on frequency 1, the RRM measurement results on frequency 1 are derived through model inference, but the RRM measurement results obtained after GAP relaxation can be compared and monitored with the predicted RRM measurement results, and subsequent management can be performed based on the monitoring results. For example, when the difference between the measured RRM result and the predicted RRM result exceeds a threshold, management operations such as rollback or model retraining / update can be performed.

[0290] But if these frequency points are the input frequency points of the model, take the first frequency point as an example:

[0291] If GAP relaxation is adopted, the frequency point 2 is used as the derivation input frequency point, and the model can be applied to derive the RRM measurement result at the frequency point 2 when GAP relaxation is not adopted.

[0292] Of course, the GAP configuration or related instructions sent by the NW to the UE may also include model-related specific GAP configurations, such as GAP measurement ID, GAP type, GAP offset, GAP length, GAP repetition period and GAP time advance, as well as GAP start frame / subframe, etc.

[0293] Among them, the GAP measurement ID represents the index information used to identify the current set of GAP configurations, the GAP type refers to the GAP configuration, which can be per-UE, per-FR1, per-FR2, etc., or it can be a configuration based on a single frequency point, such as using GAP relaxation for a certain frequency point; the GAP offset refers to the position offset of the start of the GAP within the time length of the GAP repetition period; the GAP length refers to the duration of the GAP after the start of the GAP; the GAP repetition period refers to the time interval between two adjacent GAPs; the GAP time advance refers to the time advance of the UE to perform measurements before the start of the GAP subframe, which is to avoid the overlap of the RF tuning time and the position of the SSB in the SSB measurement time configuration window in the time domain.

[0294] When these frequency points are derived frequency points of the model, let's continue with the example where the second frequency point is frequency point 1:

[0295] The model-related GAP configuration sent by the NW to the UE is to inform the UE that the RRM measurement result on a certain frequency point is obtained through model derivation, but the RRM measurement result obtained by applying the model-related GAP can be used to perform monitoring and other operations on the model.

[0296] When these frequency points are the input frequency points of the model, let's continue with the example where the first frequency point is frequency 2:

[0297] If the network is configured with a model-dependent GAP, frequency 2 is used as the derivation input frequency. The model can be used to derive the RRM measurement results for frequency 2 when the network is configured with a non-model-dependent GAP.

[0298] In the embodiment of the present application, the GAP configuration or related indication sent by the NW to the UE not only involves the model-related GAP configuration, but also includes the GAP usage at multiple different frequency points and some GAP usage preferred by the UE, etc. These enable the NW to comprehensively send the GAP parameters configured by the network in the actual process or the saved measurement GAP or related indications to the UE, thereby improving the information integrity of the GAP configuration or related indications sent by the NW to the UE.

[0299] In one embodiment, the network side device sends a model-related GAP configuration or related indication to the UE, including: the network side device sends a model-related GAP configuration or related indication to the UE based on at least one of the following information: recommended or preferred GAP configuration related information sent by the UE; monitoring results of the UE or itself for the model.

[0300] The information related to the recommended or preferred GAP configuration sent by the UE refers to the situation where the NW receives the information related to the recommended / preferred GAP configuration sent by the UE. The monitoring result of the UE or itself for the model refers to the situation where the NW completes the monitoring of the model and obtains the monitoring result of the model, or the UE completes the monitoring of the model and obtains the monitoring result of the model.

[0301] In an embodiment of the present application, the NW does not send the model-related GAP configuration or related indications to the UE without restriction, but sends them based on some specific information. In this way, sending based on specific information avoids the waste of resources when the NW sends the model-related GAP configuration or related indications to the UE, and also increases the accuracy of the model-related GAP configuration or related indications sent by the NW to the UE.

[0302] The following describes the process in which the UE sends the recommended or preferred GAP configuration related information to the NW.

[0303] In one embodiment, the embodiment of the present application also includes: the UE sends model-related recommended or preferred GAP configuration related information to the network side device.

[0304] In the embodiment of the present application, before the AI / ML model is enabled, the UE sends the model-related recommended or preferred GAP configuration related information to the NW. This process is optional and may or may not exist. However, during the process of the NW sending the model-related GAP configuration or related indication to the UE, even if the UE does not send the model-related recommended or preferred GAP configuration related information to the NW, the NW will send the model-related GAP configuration or related indication to the UE.

[0305] It should also be noted that which specific communication device uses the model for inference and prediction may not be necessarily related to the UE sending the model-related recommended or preferred GAP configuration related information, that is, the communication device is any node that performs model inference and prediction, but the model-related recommended or preferred GAP configuration related information must be sent by the UE to the NW.

[0306] In an embodiment of the present application, the model-related recommended or preferred GAP configuration-related information sent by the UE to the NW may be the input frequency of the model, that is, the first frequency or frequency combination, or the derived frequency of the model, that is, the second frequency or frequency combination. The purpose of the model-related recommended or preferred GAP configuration-related information sent by the UE to the NW is to inform the NW in advance of some GAP configurations that are more inclined to some frequencies, so that the NW can use this as a basis for reasonable configuration of each frequency.

[0307] In one embodiment, the GAP configuration related information includes at least one of the following: whether the GAP is used or not for a single or multiple frequency points; the frequency points include at least one of the same frequency points, different frequency points and different technology frequency points; the GAP is relaxed on a single or multiple frequency points; the frequency point identifier for the GAP used for a single or multiple frequency points; the frequency point identifier for the GAP not used for a single or multiple frequency points; the frequency point identifier for the GAP relaxation for a single or multiple frequency points.

[0308] That is, the UE sends model-related recommended or preferred GAP configuration related information to the network-side device, including the use or non-use of GAP for one or more frequencies, which can be any one or more of the same frequency, different frequency, and different technology frequencies. Furthermore, the indication of GAP use or GAP non-use or GAP relaxation for these frequencies can be indicated by a frequency identifier. The frequency identifier can be composed of numbers and / or letters, which is not limited in this embodiment of the present application.

[0309] Optionally, the GAP relaxation includes at least one of the following: configuring a GAP repetition period that is longer than the current GAP repetition period; configuring a GAP length that is shorter than the current GAP repetition period; puncturing or omitting some GAP positions.

[0310] The GAP configuration related information sent by the UE to the NW also includes GAP usage of some frequencies, such as GAP usage or non-usage on the same frequency, different frequency, and different technology frequencies, or GAP relaxation on a single or multiple frequencies, etc. For example, the UE recommends that the NW configure GAP not to be used on frequency 1, or that the UE recommends that the NW configure GAP relaxation on frequency 2, etc.

[0311] Take the two cases where the targeted frequency points are the input frequency points of the model and the derived frequency points of the model as examples:

[0312] When the GAP configuration-related information recommended / preferred by the UE is a derived frequency (second frequency / frequency combination) of the model, if the second frequency is frequency 1, then:

[0313] If GAP is not used on frequency point 1, it means that the RRM measurement result on frequency point 1 is obtained through model derivation; alternatively, if GAP is used on frequency point 1, it means that the RRM measurement result on frequency point 1 is obtained through traditional non-model derivation measurement method; alternatively, if GAP relaxation is used, the RRM measurement result on frequency point 1 is obtained through model derivation, but the RRM measurement result obtained after GAP relaxation can be used to perform monitoring and other operations.

[0314] When the GAP configuration-related information recommended / preferred by the UE is the input frequency (first frequency / frequency combination) of the model, if the first frequency is frequency 2, then:

[0315] If GAP relaxation is adopted, the frequency point 2 is used as the input frequency point of the model, and the RRM measurement result at the frequency point 2 when GAP relaxation is not adopted can be derived by applying the model.

[0316] In an embodiment of the present application, the UE sends some information related to the GAP configuration that it recommends or prefers to the NW, which is used to inform the NW whether to use or not use multiple GAPs preferred by different frequency points, and whether these frequency points prefer to adopt GAP relaxation and other configuration information, so that the NW can refer to this information to reasonably perform relevant GAP configurations of the model.

[0317] In one embodiment, the UE may transmit model-related recommended or preferred GAP configuration-related information to the NW via any of radio resource control signaling, media access control layer control elements, or physical layer instructions. Specifically, this may be transmitted via RRC signaling, MAC CE, or physical layer instructions. In this way, by utilizing existing signaling to transmit information, limited wireless network resources can be fully utilized while meeting the UE's own quality of service requirements, ensuring that information can be effectively transmitted to the NW.

[0318] Based on any of the foregoing embodiments, the following provides two embodiments in which the communication device is a UE and the communication device is a network-side device (illustrated as NW). In these embodiments, the model is an AI / ML model as an example. Detailed information is as follows:

[0319] Example 1:

[0320] In this embodiment, the AI / ML model is deployed on the UE side. The AI / ML model is trained by a third-party server as a training node, and the UE performs monitoring. As shown in Figure 6, this embodiment includes the following steps:

[0321] S11: The UE sends training data to a third-party server.

[0322] The third-party server may be a third-party server outside the 3GPP system.

[0323] The training data sent by the UE includes: RRM measurement results on the third frequency point / frequency point combination, RRM measurement results on the fourth frequency point / frequency point combination, and / or UE related information.

[0324] (1) The third frequency point / frequency point combination may be one or more co-frequency points, one or more heterogeneous frequency points, or a combination of co-frequency points and heterogeneous frequency points;

[0325] (2) The fourth frequency / frequency combination may be one or more co-frequency frequencies, one or more heterogeneous frequencies, one or more heterogeneous technology frequencies, or a combination of co-frequency frequencies / heterogeneous frequencies / heterogeneous technology frequencies, which are different from the first frequency / frequency combination;

[0326] (3) The third frequency point / frequency point combination may be completely different from the fourth frequency point / frequency point combination; or there may be overlap between the frequencies of the third frequency point / frequency point combination and the fourth frequency point / frequency point combination. For example, the GAP relaxation mechanism is adopted on the third frequency point / frequency point combination or the measurement is performed based on the GAP configuration related to the AI / ML model, and the RRM measurement result is derived by the AI / ML model when the GAP relaxation mechanism is not adopted on the fourth frequency point / frequency point combination or the network configuration is a GAP not related to the AI / ML model.

[0327] (4) UE related information may include but is not limited to: UE geographical location information, UE speed information, etc.

[0328] The geographic location information of the UE may be the horizontal / vertical coordinates of the UE, or location-related information obtained based on various positioning methods defined within / outside the 3GPP system (eg, positioning methods based on protocols such as TS37.355).

[0329] The UE speed information may be actual UE speed information, or various parameters related to the UE speed, for example, a UE high / medium / low speed classification result obtained based on the calculation method in TS38.304.

[0330] (5) RRM measurement results for each frequency / frequency combination, which can be the serving cell measurement result, the neighboring cell measurement result, or a combination of the serving cell and neighboring cell measurement results, or the UE's measurement results for the beam in the cell;

[0331] (6) The type of the RRM measurement result at each frequency point / frequency point combination can be one or more of RSRP, RSRQ, and SINR measurement results.

[0332] S11a: The third-party server may also receive part of the training data from the NW.

[0333] The training data sent by the NW to the third-party server includes UE-related information. This step is optional and does not need to be performed if the UE-related information collected by the NW is not required.

[0334] For example, some UE-related information received from the NW includes:

[0335] (1) UE geographic location information. This can be the UE's horizontal / vertical coordinates, or location-related information obtained by the NW based on various positioning methods defined within or outside the 3GPP system (for example, positioning methods based on protocols such as TS37.355).

[0336] (2) The UE speed information may be actual UE speed information, or various parameters related to the UE speed, for example, the UE high / medium / low speed classification result calculated based on the UE mobility history information.

[0337] S12: The third-party server performs model training to obtain a trained model.

[0338] S13: The third-party server transmits the model to the UE.

[0339] It can be sent to the UE through the operator node, base station and other nodes within the 3GPP system; or sent to the UE through other methods outside the 3GPP system, for example, the UE downloads it from an APP.

[0340] S14 / S14a: The UE obtains the input data required for derivation from its own measurement results or from the NW.

[0341] The derivation input data may include: an RRM measurement result on the first frequency point / frequency point combination, and / or UE related information.

[0342] (1) The first frequency point / frequency point combination may be one or more co-frequency points, one or more heterogeneous frequency points, or a combination of co-frequency points and heterogeneous frequency points. For example, the measurement is performed on the first frequency point / frequency point combination using a GAP relaxation mechanism or a GAP configuration related to an AI / ML model.

[0343] (2) UE related information may include but is not limited to: UE geographical location information, UE speed information, etc. See the description in S11 above for details.

[0344] S15: The UE side performs derivation based on the model.

[0345] The result derived by the UE may be a prediction of the cell measurement result on the second frequency point / frequency point combination.

[0346] That is, the RRM measurement results are derived through the AI / ML model when GAP relaxation is not adopted on the second frequency point / frequency point combination, or when the network configuration has a GAP that is not related to the AI / ML model.

[0347] The second frequency / frequency combination may be one or more same-frequency frequencies, one or more different-frequency frequencies, one or more different-technology frequencies, or a combination of same-frequency frequencies / different-frequency frequencies / different-technology frequencies, which are different from the first frequency / frequency combination;

[0348] The first frequency point / frequency point combination may be completely different from the second frequency point / frequency point combination; or there may be overlap between the frequencies of the first frequency point / frequency point combination and the second frequency point / frequency point combination.

[0349] S16 / S16a: The result derived by the UE can be used by the UE itself or sent to the NW for subsequent use.

[0350] The UE side uses it on its own, or the NW uses it later, to make a handover decision based on the derivation result.

[0351] S17: The UE may perform model monitoring based on the KPI and / or the actually measured RRM measurement results.

[0352] The UE side can monitor KPIs, for example:

[0353] (1) Compared with the case where the AI / ML model is not used to predict the cell measurement results on the second frequency point / frequency point combination, the AI / ML model in the embodiment of the present application is used to predict the cell measurement results on the second frequency point / frequency point combination, and whether the increase in the radio link failure rate exceeds the threshold, whether the comparison result of the occurrence rate of the ping-pong handover probability exceeds the threshold, or whether the comparison result of the interruption time exceeds the threshold, etc.

[0354] (2) Compared with using other AI / ML models to predict the measurement results of the cell on the second frequency point / frequency combination, using the AI / ML model in the embodiment of the present application to predict the measurement results of the cell on the second frequency point / frequency combination, whether the increase in the occurrence rate of the ping-pong handover probability exceeds the threshold, whether the increase in the occurrence rate of the line link failure exceeds the threshold, or whether the comparison result of the interruption time exceeds the threshold, etc.;

[0355] (3) Compared with using other AI / ML models to predict the cell measurement results on the second frequency point / frequency point combination, using the AI / ML model in the embodiment of the present application to predict the cell measurement results on the second frequency point / frequency point combination, whether the comparison result of the interruption time exceeds the threshold, whether the increase in the occurrence rate of wireless link failure exceeds the threshold, whether the comparison result of the occurrence rate of ping-pong handover probability exceeds the threshold, etc.

[0356] The UE also monitors the RRM measurement results, for example:

[0357] (1) A comparison result between the RRM measurement result at the second frequency point / frequency point combination predicted by the AI / ML model in the embodiment of the present application and the RRM measurement result actually measured at the second frequency point / frequency point combination, for example, whether the difference between the two exceeds a threshold. The RRM measurement result actually measured at the second frequency point / frequency point combination may be an RRM measurement result based on a relaxed measurement GAP or an AI / ML-related GAP configuration.

[0358] (2) Comparison results between using other AI / ML models to predict the RRM measurement results at the second frequency point / frequency point combination and using the AI / ML model in the embodiment of the present application to predict the RRM measurement results at the second frequency point / frequency point combination.

[0359] S18: The UE may send the monitoring result to the NW side.

[0360] This step is performed if the model / function management node is in the NW, that is, if the model / function management node is in the UE, this step may not be performed.

[0361] S18a / S19: The model / function monitoring node (eg, UE or NW side) may decide whether to initiate a model / function management-related process based on the monitoring results and / or overhead savings.

[0362] If the model / function monitoring node is the UE, the UE will decide whether to initiate the model / function management-related process based on the monitoring results and / or cost savings. If the model / function monitoring node is the NW side, the NW side will decide whether to initiate the model / function management-related process based on the monitoring results and / or cost savings.

[0363] (1) Overhead savings may include: interruption time savings, and / or measurement of GAP usage / savings, etc.;

[0364] (2) Model / function management can include activation, deactivation, conversion, rollback, update, retraining, fine-tuning and other behaviors of the model / function.

[0365] The implementation principles, processes and technical effects of each step in this embodiment are the same as those in the above embodiment. For some contents that are not fully described, please refer to the above embodiment and will not be repeated here.

[0366] Example 2:

[0367] In this embodiment, the AI / ML model is deployed on the NW side. The AI / ML model is trained by the operator node as a training node, and the NW performs monitoring. As shown in Figure 7, this embodiment includes the following steps:

[0368] S21: The UE sends training data to a model training node, which may be an operator node, a core network node, or a base station node in a 3GPP system. The training data is consistent with the training data in step S11 in Example 1.

[0369] S21a: Optionally, the model training node may also receive some UE-related information from the NW.

[0370] The UE related information is consistent with the UE related information in step S11a in embodiment 1.

[0371] S22: The model training node performs model training to obtain a trained model.

[0372] S23: The model training node sends the model to the NW side (eg, base station).

[0373] Likewise, it can be sent to the NW via the operator node, base station and other nodes within the 3GPP system.

[0374] S24 / S24a: The NW obtains the derivation input data required for derivation from itself or from the UE:

[0375] (1) The derivation input data may include: RRM measurement results on the first frequency / frequency combination, and / or UE-related information. The specific content is the same as step S14 / step S14a in embodiment 1;

[0376] (2) The NW obtains the input data required for derivation from the UE, which can be obtained through UE measurement reporting, for example, by measuring and periodically reporting the RRM measurement results on the first frequency point / frequency point combination.

[0377] S25: The NW side performs derivation based on the model.

[0378] The result derived by NW may be a prediction of the RRM measurement result at the second frequency point / frequency point combination.

[0379] The specific content of the second frequency point / frequency point combination is the same as step S15 in embodiment 1.

[0380] S26: The results derived by NW can be used by the NW side.

[0381] The NW side uses the derivation result to make a handover decision.

[0382] S27: The NW may perform model monitoring based on the KPI and / or actual measured RRM measurement results:

[0383] (1) The NW side may monitor the KPI, and the monitored KPI may be consistent with the content in step S17 in embodiment 1.

[0384] (2) The NW side may monitor the RRM measurement results, which may be consistent with the content of step S17 in embodiment 1.

[0385] S28: The model / function monitoring node (eg, base station, etc.) may decide whether to initiate a model / function management-related process based on the monitoring results and / or the cost savings.

[0386] The cost saving and model / function management content may be consistent with the content in step S18a / S19 in embodiment 1.

[0387] The implementation principles, processes and technical effects of each step in this embodiment are the same as those in the above embodiment. For some contents that are not fully described, please refer to the above embodiment and will not be repeated here.

[0388] The above Figures 6 and 7 provide a simple schematic diagram of each step in Examples 1-2 for easy reference. In Figures 6 and 7, each step is distinguished by solid and dashed lines, with solid-line boxes representing required steps and dashed-line boxes representing optional steps.

[0389] In addition, based on any of the above embodiments, the present application also provides an embodiment in which the UE sends recommended / preferred GAP configuration related information to the NW, and the NW sends relevant GAP configuration or related instructions to the UE. Of course, as mentioned in the previous embodiment, the UE sending recommended / preferred GAP configuration related information to the NW is an optional process, that is, this process may or may not exist. See the following for details:

[0390] Example 3:

[0391] This embodiment measures GAP recommendations / preferences and network GAP configuration / instructions. Taking the network-side device as an NW and the model as an AI / ML model as an example, this embodiment includes the following:

[0392] (1) Optionally, the UE may send GAP configuration-related information related to the AI / ML model recommendation / preferred information to the NW. The specific GAP configuration-related information includes:

[0393] Recommend / preference the use / non-use of GAP for different same-frequency points / different-frequency points / different-technology points;

[0394] Recommendation / preference is given to GAP relaxation on a single / multiple frequency points, where GAP relaxation includes, for example: configuring a GAP repetition period longer than the current GAP repetition period; configuring a GAP length shorter than the current GAP repetition period; puncturing / omitting some GAP positions, etc.

[0395] (2) The NW can send AI / ML model-related GAP configuration or related instructions to the UE, including:

[0396] Whether the GAP configuration recommended / preferred by the UE is adopted and / or the adoption status;

[0397] AI / ML model-related specific GAP use / non-use for different same-frequency / different-frequency / different-technology frequency points;

[0398] AI / ML models are specifically related to GAP relaxation on single / multiple frequency points. GAP relaxation includes: configuring a GAP repetition period that is longer than the current GAP repetition period; configuring a GAP length that is shorter than the current GAP repetition period; puncturing / omitting some GAP positions, etc.

[0399] Specific GAP configurations related to AI / ML models, such as GAP start frame / subframe, GAP repetition period, GAP length, GAP offset, GAP time advance, and other parameters.

[0400] (3) The NW sends AI / ML model-related GAP configuration or related instructions to the UE, which may be based on but not limited to the following information:

[0401] Recommended / preferred GAP configuration information sent by the UE to the NW;

[0402] UE / NW monitoring results for AI / ML.

[0403] (4) The UE sends AI / ML model-related recommended / preferred GAP configuration-related information to the NW, or the NW sends AI / ML model GAP-related configuration / indication to the UE via RRC signaling, MAC CE, or physical layer instructions.

[0404] (5) When the frequency of the GAP configuration-related information recommended / preferred by the UE is the derived frequency (second frequency / frequency combination) of the AI / ML model, and / or the frequency of the GAP parameter / related indication related to the AI / ML model is the derived frequency (second frequency / frequency combination) of the AI / ML model:

[0405] If GAP is not used on a certain frequency point, the RRM measurement results on this frequency point are derived through the AI / ML model; or, if GAP is used on a certain frequency point, the RRM measurement results on this frequency point are obtained through traditional non-AI derivation measurement methods; or, if GAP relaxation is used on a certain frequency point, the RRM measurement results on this frequency point are derived through the AI / ML model, but the RRM measurement results obtained after GAP relaxation can be used to perform monitoring and other operations.

[0406] For the GAP configuration of a certain frequency point related to the AI / ML model sent by the NW to the UE, the RRM measurement result on the frequency point is obtained by inferring the AI / ML model, but the RRM measurement result obtained by applying the GAP related to the AI / ML model can be used to perform monitoring and other operations.

[0407] (6) When the frequency of the GAP configuration-related information recommended / preferred by the UE is the derivation input frequency (first frequency / frequency combination) of the AI / ML model, and / or the frequency of the GAP parameter / related indication related to the application of the AI / ML model is the derivation input frequency (first frequency / frequency combination) of the AI / ML model:

[0408] If GAP relaxation is used for a frequency point, the AI / ML model can be applied to the frequency point to derive the RRM measurement results when GAP relaxation is not used. If the network is configured with a GAP configuration for a frequency point related to the AI / ML model, the RRM measurement results when the network is configured with a non-AI / ML related GAP can be applied to the frequency point as the derivation input frequency.

[0409] The implementation principles, processes and technical effects of each step in this embodiment are the same as those in the above embodiment. For some contents that are not fully described, please refer to the above embodiment and will not be repeated here.

[0410] In addition, each step in each of the above embodiments 1-3 is not mandatory, and other steps may be included between the steps. The steps in each embodiment of the present application are not limited to be performed in the order described. Some steps in each embodiment of the present application can be performed in parallel.

[0411] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0412] Based on the same inventive concept, an embodiment of the present application further provides a frequency domain measurement prediction device for implementing all the frequency domain measurement prediction methods involved above.

[0413] As shown in FIG8 , an embodiment of the present application provides a frequency domain measurement prediction device 80 , which includes:

[0414] An acquisition module 801 is configured to obtain a first measured RRM measurement result of a UE on a first frequency or frequency combination;

[0415] The prediction module 802 is configured to perform derivation using the first measured RRM measurement result as an input of a model to obtain a predicted RRM measurement result of the UE at a second frequency point or frequency point combination.

[0416] In one embodiment, the first frequency point or frequency point combination includes at least one of the following: one or more same-frequency frequency points; one or more different-frequency frequency points; a combination of same-frequency frequency points and different-frequency frequency points.

[0417] In one embodiment, the second frequency point or frequency point combination includes at least one of the following: one or more same-frequency points; one or more different-frequency points; one or more different-technology points; a combination of at least two of the same-frequency points, different-frequency points and different-technology points.

[0418] In one embodiment, the first measured RRM measurement result or the predicted RRM measurement result includes at least one of the following: a serving cell measurement result; a neighboring cell measurement result; a combination of serving cell and neighboring cell measurement results; or a UE measurement result of a beam in the cell.

[0419] In one embodiment, the type of the first measured RRM measurement result or the predicted RRM measurement result includes at least one of the following: reference signal received power; reference signal received quality; and signal-to-noise ratio.

[0420] In one embodiment, the communication device is a network-side device, and the acquisition module 801 includes:

[0421] The measurement result acquiring unit is configured to receive a first measured RRM measurement result periodically reported by the UE according to the measurement configuration.

[0422] In one embodiment, the prediction module 802 includes:

[0423] UE information acquisition unit, used to obtain UE related information;

[0424] The measurement result prediction unit is configured to derive the first measured RRM measurement result and the UE related information as inputs of a model to obtain a predicted RRM measurement result.

[0425] In one embodiment, the UE information acquiring unit includes:

[0426] The first UE information acquisition subunit is used to acquire UE-related information collected by the UE, and / or the second UE information acquisition subunit is used to acquire UE-related information collected by the network side device.

[0427] In one embodiment, the first UE information acquisition subunit is further configured to receive UE related information periodically reported by the UE according to the measurement configuration.

[0428] In one embodiment, the UE related information includes at least one of the following: UE location information; UE speed information.

[0429] In one embodiment, the location information of the UE includes at least one of the following: the geographic location coordinates of the UE; location related information obtained by the UE based on a positioning method; location related information obtained by a network side device based on a positioning method.

[0430] In one embodiment, the UE speed information includes: UE actual speed information, and / or various parameter information related to the UE speed.

[0431] In one embodiment, the apparatus further comprises:

[0432] The handover decision module is used to make a handover decision for the UE based on the predicted RRM measurement result.

[0433] In one embodiment, the communication device is a UE, and the apparatus further includes:

[0434] The measurement result sending module is used to send the predicted RRM measurement result to the network side device.

[0435] In one embodiment, the apparatus further comprises:

[0436] Model monitoring module, used to monitor model execution;

[0437] The management module is used to perform management operations on the model based on the monitoring results of the model and / or the overhead saving information related to the model.

[0438] In one embodiment, monitoring the model includes at least one of the following:

[0439] Compare the performance indicator values ​​predicted by the application model with the preset / network configured performance indicator values;

[0440] Compare the performance index values ​​predicted by the application model with the performance index values ​​obtained by actual measurement;

[0441] Compare the performance index values ​​predicted by the application model with those predicted by other different models;

[0442] comparing the predicted RRM measurement result with a second measured RRM measurement result at the second frequency point / frequency point combination;

[0443] The predicted RRM measurement result is compared with the RRM measurement result at the second frequency point / frequency point combination predicted by using other different models.

[0444] In one embodiment, the performance indicator value includes at least one of the following: radio link failure occurrence rate; handover failure occurrence rate; ping-pong handover probability; and interruption duration.

[0445] In one embodiment, the overhead saving information includes at least one of the following: interruption time saving time; measurement of GAP usage time; measurement of GAP saving time; measurement of required UE power saving.

[0446] In one embodiment, the management operation includes performing at least one of the following operations:

[0447] Activate the model or model-related functions;

[0448] Deactivate the model or model-related functions;

[0449] Transforming the model into another model with better performance indicators and / or RRM measurements;

[0450] Convert model-related features into other model-related features that provide better performance indicators and / or RRM measurements;

[0451] Roll back the model or model-related functions;

[0452] Update the model or model-related functions;

[0453] Retrain the model;

[0454] Fine-tune the model.

[0455] In one embodiment, the device further includes: a model acquisition module, configured to train a model based on training data; and / or receive a model sent by a training node, wherein the model is trained by the training node based on the training data.

[0456] In one embodiment, the training node includes at least one of the following: a third-party server; a base station node; an operator node; a UE; or a core network node.

[0457] In one embodiment, the training data includes at least one of the following: the measured RRM measurement result of the UE on the third frequency point / frequency point combination; the measured RRM measurement result of the UE on the fourth frequency point / frequency point combination; UE-related information collected by the UE; and UE-related information collected by the network side device.

[0458] In one embodiment, the apparatus further comprises:

[0459] The configuration information recommendation module is used for the UE to send model-related recommendations or preferred GAP configuration related information to the network side device.

[0460] In one embodiment, the GAP configuration related information includes at least one of the following:

[0461] GAP use or non-use for a single or multiple frequency points; the frequency point includes at least one of the same frequency point, different frequency point, and different technology frequency point;

[0462] GAP relaxation on single or multiple frequency points;

[0463] Frequency identifiers used for single or multiple frequency GAPs;

[0464] Frequency identifiers not used for GAP of single or multiple frequencies;

[0465] Frequency identifier for GAP relaxation for single or multiple frequencies.

[0466] In one embodiment, GAP relaxation includes at least one of the following:

[0467] Configure a longer GAP repetition period than the current GAP repetition period;

[0468] Configure a GAP length that is shorter than the current GAP repetition period;

[0469] Punch holes or omit some GAP positions.

[0470] In one embodiment, the configuration information recommendation module includes: a configuration information recommendation unit, configured for the UE to send model-related recommendations or preferred GAP configuration-related information to the NW via any of the following signaling: radio resource control signaling; media access control layer control elements; physical layer instructions.

[0471] In one embodiment, the apparatus further comprises:

[0472] The configuration indication module is used for the network side device to send model-related GAP configuration or related indications to the UE.

[0473] In one embodiment, the GAP configuration or related indication includes at least one of the following:

[0474] Whether the GAP configuration recommended or preferred by the UE is adopted;

[0475] Adoption of GAP configurations recommended or preferred by the UE;

[0476] Model-dependent GAP use or non-use for the same frequency points;

[0477] Model-dependent GAP use or non-use for different frequency points;

[0478] Model-dependent use or non-use of GAP for different technology frequencies;

[0479] Model-dependent GAP relaxation at single or multiple frequency points;

[0480] Model-related GAP configuration.

[0481] In one embodiment, the configuration indication module includes: a first configuration indication sending unit, configured for the network side device to send a model-related GAP configuration or related indication to the UE based on at least one of the following information:

[0482] Recommended or preferred GAP configuration information sent by the UE;

[0483] UE or its own monitoring results for the model.

[0484] In one embodiment, the above-mentioned configuration indication module includes: a second configuration indication sending unit, which is used for the network side device to send model-related GAP configuration or related indications to the UE through any of the following signaling: radio resource control signaling; media access control layer control element; physical layer instruction.

[0485] It should be noted here that the implementation principles, processes and technical effects of the various frequency domain measurement prediction devices provided in the embodiments of the present application are similar to those of the aforementioned frequency domain measurement prediction method. The parts and beneficial effects of the frequency domain measurement prediction device embodiments that are the same as those in the frequency domain measurement prediction method embodiments will not be described in detail here.

[0486] It should be noted that the division of modules, units, and sub-units in the embodiments of the present application is schematic and is only a logical functional division. In actual implementation, there may be other division methods. In addition, the functional units in the various embodiments of the present application may be integrated into a processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The above-mentioned integrated units may be implemented in the form of hardware or in the form of software functional units.

[0487] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a processor-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application.

[0488] It should be noted here that the above-mentioned device provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned method embodiment and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as those in the method embodiment will not be described in detail here.

[0489] In one embodiment, a communication device is also provided, wherein the communication device includes a memory, a transceiver, and a processor; the memory is used to store computer programs; the transceiver is used to send and receive data under the control of the processor; the processor is used to read the computer program in the memory and perform the steps in any one of the above-mentioned frequency domain measurement prediction method embodiments.

[0490] A schematic structural diagram of a communication device provided in an embodiment of the present application can be shown with reference to FIG1 above. In FIG1 , the bus architecture of the communication device may include any number of interconnected buses and bridges, specifically various circuits of one or more processors represented by a processor and a memory represented by a memory linked together. The bus architecture may also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface is used to provide an interface. The transceiver may be a plurality of components, i.e., a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium, such as a wireless channel, a wired channel, an optical cable, etc. The processor is responsible for managing the bus architecture and general processing, and the memory may store data used by the processor when performing operations. The processor may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a complex programmable logic device (CPLD), and the processor may also adopt a multi-core architecture. The processor calls the program stored in the memory to execute the method steps in any of the above embodiments of the present application according to the obtained executable instructions.

[0491] In addition, as shown in Figure 9, an embodiment of the present application also provides a structural diagram of another communication device. In Figure 9, in addition to including a processor 900, a transceiver 910 and a memory 920, the communication device also includes a user interface 930;

[0492] Among them, the functions and principles of the processor 900, transceiver 910 and memory 920 are the same as those in Figure 1 above, and will not be repeated here. In the communication device architecture shown in Figure 9, for different user devices, the user interface can be an interface that can connect to external or internal devices. The connected devices include but are not limited to keypads, displays, speakers, microphones, joysticks, etc. The processor 900 is responsible for managing the bus architecture and general processing, and the memory 920 can store data used by the processor 900 when performing operations. Similarly, the processor 900 in Figure 9 can be a CPU (central processing unit), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) or CPLD (Complex Programmable Logic Device), and the processor 900 can also adopt a multi-core architecture. Among them, the processor 900 calls the program stored in the memory to perform the method steps in any of the above embodiments of the present application according to the obtained executable instructions.

[0493] In one embodiment, a processor-readable storage medium is further provided, the processor-readable storage medium storing a program for causing a processor to execute any of the above-mentioned frequency domain measurement prediction methods. The processor-readable storage medium can be any available medium or data storage device that can be accessed by the processor, including but not limited to magnetic storage (such as a floppy disk, hard disk, magnetic tape, magneto-optical disk (MO), etc.), optical storage (such as CD, DVD, BD, HVD, etc.), and semiconductor storage (such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)), etc.

[0494] In one embodiment, a computer program product is also provided. When executed by a processor, the computer program can implement any of the above-described frequency domain measurement prediction methods. The computer program product includes one or more computer instructions. When these computer instructions are loaded and executed on a computer, part or all of the above-described methods can be implemented in whole or in part according to the processes or functions described in the embodiments of this application.

[0495] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage, etc.) that contain computer-usable program code.

[0496] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. A frequency domain measurement prediction method, applied to a communication device, wherein: The method comprises: Obtaining a first actually measured radio resource management (RRM) measurement result of a user equipment (UE) on a first frequency or frequency combination; and The first measured RRM measurement result is used as an input of a model to perform derivation to obtain a predicted RRM measurement result of the UE at a second frequency point or a frequency point combination.

2. The method according to claim 1, wherein The first frequency or frequency combination includes at least one of the following: One or more co-frequency points; One or more off-frequency points; Combination of same-frequency points and different-frequency points; The same-frequency point is the same frequency relative to the current working frequency of the UE, and the different-frequency point is the different frequency relative to the current working frequency of the UE.

3. The method according to claim 1 or 2, wherein The second frequency or frequency combination includes at least one of the following: One or more co-frequency points; One or more off-frequency points; One or more different technology frequencies; A combination of at least two of the same-frequency frequency points, different-frequency frequency points, and different-technology frequency points.

4. The method according to any one of claims 1 to 3, wherein: The first measured RRM measurement result or the predicted RRM measurement result includes at least one of the following: Serving cell measurement results; Neighboring cell measurement results; A combination of serving cell and neighbor cell measurement results; UE measurement results for beams in the cell.

5. The method according to any one of claims 1 to 4, wherein: The type of the first measured RRM measurement result or the predicted RRM measurement result includes at least one of the following: Reference signal received power; Reference signal reception quality; Signal-to-noise ratio.

6. The method according to any one of claims 1 to 5, wherein: The communication device is a network-side device, and obtaining a first actually measured RRM measurement result of the UE at the first frequency or frequency combination includes: receiving the first measured RRM measurement result periodically reported by the UE according to the measurement configuration.

7. The method according to any one of claims 1 to 6, wherein: The derivation using the first measured RRM measurement result as an input of a model to obtain a predicted RRM measurement result of the UE at the second frequency point or frequency point combination includes: Obtaining UE related information; and The first measured RRM measurement result and the UE-related information are used as inputs of the model to perform deduction to obtain the predicted RRM measurement result.

8. The method according to claim 7, wherein: The acquiring UE related information includes: acquiring the UE related information collected by the UE, and / or acquiring the UE related information collected by a network side device.

9. The method according to claim 8, wherein The acquiring the UE-related information collected by the UE includes: receiving the UE-related information periodically reported by the UE according to the measurement configuration.

10. The method according to any one of claims 7 to 9, wherein: The UE-related information includes at least one of the following: UE location information; UE speed information.

11. The method according to claim 10, wherein: The UE location information includes at least one of the following: The geographic location coordinates of the UE; Location-related information obtained by the UE based on the positioning method; The network-side device obtains location-related information based on the positioning method.

12. The method according to claim 10, wherein: The UE speed information includes: UE actual speed information, and / or various parameter information related to the UE speed.

13. The method according to any one of claims 1 to 12, wherein: The method further comprises: A handover decision of the UE is made according to the predicted RRM measurement result.

14. The method according to any one of claims 1 to 5, wherein: The communication device is a UE, and the method further includes: Send the predicted RRM measurement result to the network side device.

15. The method according to any one of claims 1 to 14, wherein: The method further comprises: performing monitoring on the model; A management operation is performed on the model according to the monitoring result of the model and / or the overhead saving information related to the model.

16. The method according to claim 15, wherein The monitoring of the model includes at least one of the following: Comparing the performance indicator value predicted by applying the model with the preset / network configured performance indicator value; Comparing the performance index values ​​predicted by applying the model with the performance index values ​​actually measured; Comparing the performance index values ​​predicted by the model with those predicted by other different models; comparing the predicted RRM measurement result with a second measured RRM measurement result at the second frequency point / frequency point combination; The predicted RRM measurement result is compared with an RRM measurement result at the second frequency point / frequency point combination predicted by using another different model.

17. The method according to claim 16, wherein The performance indicator value includes at least one of the following: Radio link failure rate; Switching failure rate; Ping-pong switching probability; Length of the outage.

18. The method according to any one of claims 15 to 17, wherein: The cost saving information includes at least one of the following: Interruptions save time; Measuring the usage time of the gap GAP; Time saved in measuring GAP; Measure the required UE power saving.

19. The method according to any one of claims 15 to 17, wherein: The management operation includes performing at least one of the following operations: activating the model or a function related to the model; Deactivating the model or a function related to the model; Converting the model to another model with better performance indicators and / or RRM measurements; Converting the model-related functions into other model-related functions with better performance indicators and / or RRM measurement results; Rolling back the model or functions related to the model; updating the model or functions related to the model; retraining the model; Fine-tune the model.

20. The method according to any one of claims 1 to 19, wherein: Before performing derivation using the first measured RRM measurement result as an input to the model, the method further includes: Training the model based on training data; and / or, Receive the model sent by the training node, where the model is trained by the training node based on the training data.

21. The method according to claim 20, wherein The training node includes at least one of the following: Third-party servers; Base station node; Operator node; UE; Core network node.

22. The method according to claim 20, wherein The training data includes at least one of the following: An actual RRM measurement result of the UE on a third frequency or frequency combination; An actual RRM measurement result of the UE on a fourth frequency point / frequency point combination; UE-related information collected by the UE; UE-related information collected by network-side devices.

23. The method according to any one of claims 1 to 22, wherein: The method further comprises: The UE sends the recommended or preferred GAP configuration related information related to the model to the network side device.

24. The method according to claim 23, wherein The GAP configuration related information includes at least one of the following: GAP use or non-use for a single or multiple frequency points, where the frequency points include at least one of a same-frequency frequency point, a different-frequency frequency point, and a different-technology frequency point; GAP relaxation on single or multiple frequency points; Frequency identifiers used for single or multiple frequency GAPs; Frequency identifiers not used for GAP of single or multiple frequencies; Frequency identifier for GAP relaxation for single or multiple frequencies.

25. The method according to claim 24, wherein The GAP relaxation includes at least one of the following: Configure a longer GAP repetition period than the current GAP repetition period; Configure a GAP length that is shorter than the current GAP repetition period; Punch holes or omit some GAP positions.

26. The method according to any one of claims 23 to 25, wherein: The UE sends the model-related recommended or preferred GAP configuration related information to the NW, including: The UE sends the model-related recommendation or preferred GAP configuration related information to the NW through any of the following signaling: Radio resource control signaling; Media access control layer control elements; Physical layer instructions.

27. The method according to any one of claims 1 to 26, wherein: The method further comprises: The network side device sends the GAP configuration or related instructions related to the model to the UE.

28. The method according to claim 27, wherein The GAP configuration or related indication includes at least one of the following: Whether the GAP configuration recommended or preferred by the UE is adopted; Adoption of the GAP configuration recommended or preferred by the UE; The model-related GAP for the same frequency point is used or not; The model-related GAP for different frequency points is used or not; The use or non-use of the GAP for different technology frequencies related to the model; The model is related to GAP relaxation at single or multiple frequency points; The GAP configuration associated with the model.

29. The method according to claim 27 or 28, wherein The network side device sends the model-related GAP configuration or related indication to the UE, including: The network-side device sends the model-related GAP configuration or related indication to the UE based on at least one of the following information: Recommended or preferred GAP configuration related information sent by the UE; The monitoring result of the UE or itself for the model.

30. The method according to any one of claims 27 to 29, wherein: The network side device sends the model-related GAP configuration or related indication to the UE, including: The network side device sends the model-related GAP configuration or related indication to the UE through any of the following signaling: Radio resource control signaling; Media access control layer control elements; Physical layer instructions.

31. A frequency domain measurement prediction device, wherein: The device comprises: An acquisition module, configured to obtain a first measured RRM measurement result of the UE on a first frequency or frequency combination; The prediction module is configured to derive the first measured RRM measurement result as an input of a model to obtain a predicted RRM measurement result of the UE at a second frequency point or a frequency point combination.

32. A communication device, wherein: The communication device includes a memory, a transceiver, and a processor; memory for storing computer programs; a transceiver, configured to transmit and receive data under the control of the processor; A processor, configured to read the computer program in the memory and execute the operation steps in the method according to any one of claims 1 to 30.

33. A processor-readable storage medium, wherein: The processor-readable storage medium stores a program, and the program is used to enable the processor to execute the method according to any one of claims 1 to 30.

Images