Service network device determination method, control device and storage medium

Abstract

The present disclosure relates to a service network device determination method, a control device and a storage medium. The method comprises: on the basis of a key index at the current moment and a key index at a historical moment and by means of an index prediction AI model, predicting a key index of a target moment following the current moment, the key indexes of the current moment and the historical moment being obtained by a terminal measuring downlink reference information of each alternative network device in an alternative network device set; and, on the basis of the key index of the target moment, determining from the alternative network device set a first service network device set of the terminal.

Description

Method for determining serving network device, control device and storage medium Cross-reference to related applications The present disclosure claims priority to Chinese Patent Application No. 2025105986937 entitled “Method for Determining Serving Network Device, Control Device and Storage Medium” filed on May 09, 2025, and Chinese Patent Application No. 2024106337192 entitled “Method for Determining Serving Network Device, Control Device and Storage Medium” filed on May 21, 2024, which are both incorporated by reference in their entirety into the present disclosure. TECHNICAL FIELD The present disclosure relates to the field of communication technology, and in particular, to a method for determining a serving network device, a control device and a storage medium. BACKGROUND In the scenario of cooperative communication, multiple network devices serve a terminal in cooperation. The selection of the network devices that cooperate with each other has a great influence on the communication performance of the served terminal: when the terminal is at rest or moving at a low speed, the selection of the serving network device affects the data rate of the terminal; and when the terminal is moving at a high speed, the selection of the serving network device relates to whether the terminal will encounter a serious problem such as communication interruption. Therefore, the selection of the serving network device when the terminal moves at a high speed is an important problem to be solved. SUMMARY Based on this, it is necessary to provide a method for determining a serving network device, a control device and a storage medium in view of the above technical problems. In a first aspect, the present disclosure provides a method for determining a serving network device, the method comprising: predicting, by an index prediction AI model, a key index of a target moment after a current moment according to the key index of the current moment and a key index of a historical moment, the key index of the current moment and the key index of the historical moment being obtained by a terminal measuring downlink reference information of each candidate network device in a candidate network device set; determining, from the candidate network device set, a first serving network device set of the terminal based on the key index of the target moment. In some embodiments, the predicting, by the index prediction AI model, the key index of the target moment after the current moment according to the key index of the current moment and the key index of the historical moment comprises: inputting the key index of the current moment, the key index of the historical moment and a first time interval into the index prediction AI model; obtaining the key index of the target moment output by the index prediction AI model; The first time interval corresponds to a case where the candidate network device belongs to a second service network device set, wherein the second service network device set is a set of original service network devices that provide services for the terminal. In some embodiments, the first time interval is determined based on the following times: a propagation time of a downlink reference signal for measuring the key indicator sent by the candidate network device; a time for the terminal to measure the key indicator; a time for the terminal to feed back the key indicator to the first control device; a time for the first control device to predict the key indicator; a time for the first control device to determine the first service network device set of the terminal according to the predicted key indicator. In some embodiments, the first time interval is determined based on the following times: a propagation time of a downlink reference signal for measuring the key indicator sent by the candidate network device; a time for the terminal to measure the key indicator; a time for the terminal to predict the key indicator; a time for the terminal to feed back the predicted key indicator to the first control device; a time for the first control device to determine the first service network device set of the terminal according to the predicted key indicator. In some embodiments, the first time interval corresponds to a case where the candidate network device belongs to a second service network device set, wherein the second service network device set is a set of original service network devices that provide services for the terminal. In some embodiments, the first time interval further includes a time spent by the terminal in establishing a connection with a service network device in the first service network device set. In some embodiments, the first time interval corresponds to a case where the candidate network device does not belong to a second service network device set, wherein the second service network device set is a set of original service network devices that provide services for the terminal. In some embodiments, the indicator prediction AI model is obtained by training an initial AI model based on first training samples. The first training sample includes a first training data set and a first label data set; the first training data set includes a key indicator at a third time point and a key indicator at at least one historical time point before the third time point; and the first label data set includes a key indicator at a fourth time point, wherein the fourth time point is equal to the third time point plus the first time interval. In some embodiments, the determining, from the set of candidate network devices, the set of serving network devices of the terminal based on the key indicator at the target time point comprises: The first M candidate network devices in the set of candidate network devices with the largest key indicators are determined as the set of serving network devices of the terminal, where M is greater than or equal to 1. In some embodiments, when the key indicator is a first type of indicator, the larger the value of the key indicator, the better the performance of the system, and the determining, from the set of candidate network devices, the first set of serving network devices of the terminal based on the key indicator at the target time point comprises: The candidate network devices in the set of candidate network devices with a ratio of the key indicator greater than a set threshold are determined as the first set of serving network devices of the terminal. The key indicator ratio is a ratio of the key indicator of the candidate network device to a maximum key indicator, and the maximum key indicator is a maximum value of the key indicators of all candidate network devices in the set of candidate network devices. In some embodiments, when the key indicator is a second type of indicator, the smaller the value of the key indicator, the better the performance of the system, and the determining, from the set of candidate network devices, the first set of serving network devices of the terminal based on the key indicator at the target time point comprises: The candidate network devices in the set of candidate network devices with a ratio of the key indicator less than a set threshold are determined as the first set of serving network devices of the terminal. The key indicator ratio is a ratio of the key indicator of the candidate network device to a minimum key indicator, and the minimum key indicator is a minimum value of the key indicators of all candidate network devices in the set of candidate network devices. In some embodiments, the key indicator comprises at least one of the following: SS-RSRP, SS-RSRQ, SS-SINR, CSI-RSRP, CSI-RSRQ, CSI-SINR, SS-CQI, SS-RSSI, CSI-CQI, and CSI-RSSI. In some embodiments, before the determining, from the set of candidate network devices, the first set of serving network devices of the terminal based on the key indicator at the target time point, the method further comprises: predict, according to the position information of the terminal at the current time and the position information at the historical time, the position information at the target time through a position prediction AI model; From the initial network device set, the first control device determines the candidate network device set based on the position information at the target time. In some embodiments, the base station predicts, according to the position information of the terminal at the current time and the position information at the historical time, the position information at the target time through a position prediction AI model, comprising: inputting the position information at the current time, the position information at the historical time and a second time interval into the position prediction AI model; obtaining the position information at the target time output by the position prediction AI model; Wherein, the second time interval is the interval between the second control device sending downlink reference signal for positioning to the target time. In some embodiments, the second time interval is determined based on the following times: the time when the second control device performs positioning; the time when the second control device predicts the position information of the terminal; the time when the second control device sends the predicted position information of the terminal to the first control device; the time when the first control device determines the candidate network device set based on the predicted position information of the terminal; the time when the first control device informs the candidate network device to send downlink reference signal for measuring key indicators; the first time interval. In some embodiments, before determining, from the candidate network device set, the first service network device set of the terminal based on the key indicators at the target time, the method further comprises: determining the distance between the position information of each network device in the initial network device set and the position information of the terminal at the target time; determining, from the initial network device set, the network device with a distance less than a preset distance as the candidate network device set. In some embodiments, the preset distance is less than or equal to the minimum coverage radius of the network devices in the initial network device set. In some embodiments, the position prediction AI model is obtained by training an initial AI model based on second training samples. The second training sample includes a second training data set and a second label data set; the second training data set includes position information at a fifth time point and position information at at least one historical time point before the fifth time point; and the label data set includes position information at a sixth time point, the sixth time point being equal to the fifth time point plus the second time interval. In a second aspect, the present disclosure provides a control device, including a memory, a transceiver, and a processor: The memory is configured to store a computer program; the transceiver is configured to transceive data under control of the processor; and the processor is configured to read the computer program in the memory and perform the following operations: predicting a key indicator at a target time point after a current time point based on the key indicator at the current time point and a key indicator at a historical time point, the key indicators being obtained by a terminal measuring downlink reference information of each candidate network device in a candidate network device set; determining, from the candidate network device set, a first serving network device set of the terminal based on the key indicator at the target time point. In a third aspect, the present disclosure provides a control device, including: a prediction module configured to predict a key indicator at a target time point after a current time point based on the key indicator at the current time point and a key indicator at a historical time point, the key indicators being obtained by a terminal measuring downlink reference information of each candidate network device in a candidate network device set; a determination module configured to determine, from the candidate network device set, a first serving network device set of the terminal based on the key indicator at the target time point. In a fourth aspect, the present disclosure provides a computer readable storage medium having stored thereon a computer program, the computer program being executed by a processor to implement the method according to the first aspect or any one of the embodiments of the first aspect. In a fifth aspect, the present disclosure provides a computer program product, including a computer program, the computer program being executed by a processor to implement the method according to the first aspect or any one of the embodiments of the first aspect. The method, the control device, the storage medium and the computer program product can determine the first service network device set of the terminal from the candidate network device set based on the key indicator at the target moment, by predicting the key indicator at the target moment after the current moment through the index prediction AI model, wherein the key indicators at the current moment and the historical moment are obtained by the terminal measuring downlink reference information of each candidate network device in the candidate network device set. The scheme considers the influence of terminal movement on network device selection, and performs network device selection by predicting the key indicator, thereby avoiding the influence of terminal movement on network device selection, and making the selection of network devices more accurate in the case of terminal movement. BRIEF DESCRIPTION OF DRAWINGS Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the embodiments. The accompanying drawings are merely illustrative and are not intended to be limiting upon the disclosure. Moreover, the use of the same reference numerals in different drawings designates the same or similar components. In the drawings: FIG. 1 is a flowchart of a method for determining a service network device according to some embodiments; FIG. 2 is a flowchart of another method for determining a service network device according to some embodiments; FIG. 3 is a schematic diagram of a control device according to some embodiments; FIG. 4 is a block diagram of a control device according to some embodiments. DETAILED DESCRIPTION The embodiments of the technical solutions of the present disclosure will be described in detail below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present disclosure, and therefore only serve as examples, and cannot limit the protection scope of the present disclosure. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the present disclosure; the terms "include" and "have" and any variations thereof in the specification and claims of the present disclosure and the above description of drawings are intended to cover non-exclusive inclusion. In the description of the embodiments of the present disclosure, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present disclosure, the meaning of "a plurality of" is two or more, unless otherwise explicitly specified. Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the disclosure. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily mutually exclusive of one another. As will be apparent to those of ordinary skill in the art, embodiments described herein can be combined with other embodiments. In the description of embodiments of the disclosure, the term "and / or" is merely an association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of A alone, A and B together, and B alone. In addition, the character " / " in this paper generally represents that the front and rear associated objects are a "or" relationship. One method of service network device selection is that user equipment (UE) measures the Reference Signal Received Power (RSRP) of the downlink reference signal (such as Synchronization Signal / PBCH-Reference Signal (SSB-RS), Channel State Information-Reference Signal (CSI-RS)) sent by N network devices (N is greater than or equal to 2, that is, the number of network devices is greater than or equal to 2) in a set, and feeds back the measured RSRP to the corresponding network device; each network device feeds back the received RSRP to the control center (or called centralized processing unit), and the control center and each network device are connected through fronthaul and backhaul; the control center selects M network devices from N network devices as the service network devices of the UE. The method of selecting M network devices from N network devices can be to arrange the RSRP of N network devices to the UE in descending order, and take the first M network devices as the service network devices of the UE; it can also be to determine the M network devices with the largest ratio of the RSRP of the current network device to the RSRP of all N network devices, and it can also select the network devices with the RSRP ratio greater than a set threshold value as the service network devices. In addition to selecting a serving network device according to RSRP, a serving network device can also be selected according to a reference signal received quality (RSRQ), a signal to interference and noise ratio (SINR), a channel quality indicator (CQI), a received signal strength indicator (RSSI), and the like. For selection of a serving network device for a high-speed mobile terminal, a single frequency network (SFN) technology is mainly used for a high-speed rail scenario. In the SFN technology, multiple network devices are connected to a same baseband processing unit (BBU) through optical fibers, and share a same cell identity (ID) to avoid frequent selection of a serving network device and cell switching. The above method of selecting a serving network device according to RSRP and RSRQ is only applicable to a case where a terminal is at rest or moves at a low speed. When the terminal moves at a high speed, the above index is no longer accurate for selecting a serving network device, because the above index has changed greatly when the selected network device starts to provide communication services, as compared with when the index is measured. The above SFN technology for a high-speed rail scenario is deployed along a railway track, which is equivalent to a known possible movement track of a terminal, and is not applicable to a case where a terminal moves at a high speed in an urban area. In a cooperative communication scenario, multiple network devices cooperate with each other to serve a terminal (for example, a UE). The network devices can be base stations, transmission and reception points (TRPs), remote radio units (RRUs), and the like. Selection of the network devices that cooperate with each other has a great influence on communication performance of the served terminal. When the terminal is at rest or moves at a low speed, selection of a serving network device affects a data rate of the terminal. When the terminal moves at a high speed, selection of a serving network device relates to whether the terminal will encounter a communication interruption and the like. Therefore, how to solve selection of a serving network device when a terminal moves at a high speed in a terminal-centered distributed multiple input multiple output (MIMO) architecture is an important problem to be solved. In the present disclosure, based on artificial intelligence (AI) prediction of terminal movement, preliminary screening of service network devices is performed according to predicted position information after movement, key indicators for network device selection between the terminal and the preliminary selected network devices are predicted based on AI, the service network devices are secondarily screened based on the key indicators (for example, steps 201 and 202 of the embodiment of FIG. 2 are the first screening, and steps 203 and 204 are the second screening), and a final set of service network devices is obtained. The present disclosure takes into account the impact of terminal movement on network device selection, and through the method of AI prediction of terminal movement trajectory and key indicators, the impact of terminal movement on network device selection is avoided, so that the selection of network devices is more accurate in the case of terminal movement. In some embodiments, as shown in FIG. 1, a flowchart of a method for determining a service network device is provided, which can be applied to a first control device and can include but is not limited to the following steps: 101. Predicting, by an indicator prediction AI model, a key indicator at a target time after a current time according to the key indicator at the current time and the key indicator at a historical time. The key indicator at the current time is obtained by measuring the downlink reference information of each candidate network device in the candidate network device set by the terminal at the current time, and the key indicator at the historical time is obtained by measuring the downlink reference information of each candidate network device in the candidate network device set by the terminal at the historical time. The historical time can refer to a time interval before the current time. The above-mentioned key indicators include at least one of the following: SS-RSRP, SS-RSRQ, SS-SINR, CSI-RSRP, CSI-RSRQ, CSI-SINR, SS-CQI, SS-RSSI, CSI-CQI, and CSI-RSSI. SS-RSRP represents RSRP measured on an SSB-RS signal resource unit; SS-RSRQ represents RSRQ measured on an SSB-RS signal resource unit; SS-SINR represents SINR measured on an SSB-RS signal resource unit; CSI-RSRP represents RSRP measured on a CSI-RS signal resource unit; CSI-RSRQ represents RSRQ measured on a CSI-RS signal resource unit; CSI-SINR represents SINR measured on a CSI-RS signal resource unit; SS-CQI represents CQI measured on an SSB-RS signal resource unit; SS-RSSI represents RSSI measured on an SSB-RS signal resource unit; CSI-CQI represents CQI measured on a CSI-RS signal resource unit; and CSI-RSSI represents RSSI measured on a CSI-RS signal resource unit. The first control device can be the control center. In the process of obtaining the key indicators, the alternative network devices can send downlink reference information (i.e., downlink reference signals) to the terminal, and the corresponding terminal receives the downlink reference information and measures the key indicators, and then transmits the measured key indicators to the first control device through the alternative network devices, or directly transmits the measured key indicators to the first control device. In some embodiments, the prediction of the key indicators at the target time after the current time based on the key indicators at the current time and the key indicators at the historical time can include but is not limited to: inputting the key indicators at the current time, the key indicators at the historical time, and the first time interval into an indicator prediction AI model; obtaining the key indicators at the target time output by the indicator prediction AI model. The historical time can be a time before the current time. The first time interval is the time interval between the first time and the target time, the first time is the time when the alternative network device set sends the downlink reference signal for measuring the key indicators for the first time, and the target time is the time when the first control device determines the first service network device set according to the measurement result, and the measurement result is obtained based on the downlink reference signal sent at the first time. In some embodiments, when each alternative network device in the alternative network device set has undergone synchronization processing, the first time for each alternative network device in the alternative network device set is the same time; when each alternative network device in the alternative network device set has not undergone synchronization processing, the downlink reference signals for measuring the key indicators sent by each alternative network device are slightly different, and the first time is the time when the alternative network device set sends the downlink reference signal for measuring the key indicators for the first time. In some embodiments, the first service network device set can include identification information of at least one service network device determined according to the key indicators at the target time. The identification information of the at least one service network device can be an index of the at least one service network device, which is used to distinguish each service network device. The first time can be the time when each alternative network device in the alternative network device set starts to send the downlink reference signal for measuring the key indicators. In the present disclosure, the control center can control each alternative network device to start sending the downlink reference signal for measuring the key indicators at the same time. In some embodiments, the above-mentioned alternative network device set can be an initial network device set, and can include identification information of a plurality of network devices. The plurality of network devices can be a plurality of network devices in an initial selected target area, and the target area can be an area within a preset distance range from the terminal. In the embodiments of the present disclosure, the preset distance range can be set according to actual needs, and the embodiments of the present disclosure are not limited. In some embodiments, the above-mentioned alternative network device set can also be a set of network devices selected from the initial network device set according to the predicted position information of the terminal at the target moment, and the selected set includes identification information of some network devices within a certain distance range from the position information of the terminal at the target moment. In some embodiments, the determination of the first time interval is different for different cases of the first control device predicting the key indicator and the terminal predicting the key indicator. In the embodiments of the present disclosure, the predicted key indicator refers to a key indicator predicted at a target moment after the current moment. In some embodiments, the determination of the first time interval is also different for different cases of whether the alternative network device belongs to the second service network device set. The second service network device set is the original service network device set that provides services for the terminal. In some embodiments, corresponding to the case that the alternative network device belongs to the second service network device set and the first control device predicts the key indicator, the first time interval is determined based on the following times: The propagation time of the downlink reference signal for measuring the key indicator sent by the alternative network device can be represented as t1; The time for the terminal to measure the key indicator can be represented as t2; The time for the terminal to feed back the key indicator to the first control device can be represented as t3; The time for the first control device to predict the key indicator can be represented as t4; The time for the first control device to determine the first service network device set of the terminal according to the predicted key indicator can be represented as t5. In some embodiments, the first time interval can be represented as ΔT1, ΔT1=t1+t2+t3+t4+t5. In some embodiments, the plurality of times t1, t2, t3, t4, and t5 for determining the first time interval are time lengths. In some embodiments, corresponding to the case that the alternative network device belongs to the second service network device set and the terminal predicts the key indicator, the first time interval is determined based on the following times: a propagation time t1 of a downlink reference signal for measuring the key indicator sent by the alternative network device; a time t2 for the terminal to measure the key indicator; a time t6 for the terminal to predict the key indicator, which can be represented as t6 = t2 + t3 + t4 + t5 + t6; a time t7 for the terminal to feed back the predicted key indicator to the first control device; a time t5 for the first control device to determine the first service network device set of the terminal according to the predicted key indicator. In some embodiments, the first time interval can be represented as ΔT1, ΔT1 = t1 + t2 + t6 + t7 + t5. The multiple times t1, t2, t6, t7, t5 for determining the first time interval are time lengths. If the alternative network device belongs to the second service network device set and the alternative network device is finally determined as a service network device in the first service network device set of the terminal, the alternative network device belongs to the intersection of the second service network device set and the first service network device set. In some embodiments, corresponding to the case that the alternative network device does not belong to the second service network device set and the first control device predicts the key indicator, the first time interval is determined based on the following times: a propagation time t1 of a downlink reference signal for measuring the key indicator sent by the alternative network device; a time t2 for the terminal to measure the key indicator; a time t3 for the terminal to feed back the key indicator to the first control device; a time t3 for the first control device to predict the key indicator; a time t5 for the first control device to determine the first service network device set of the terminal according to the predicted key indicator. a time t8 for the terminal to establish a connection with a service network device in the first service network device set. The second service network device set is an original service network device set that provides services for the terminal. In some embodiments, the first time interval can be represented as ΔT1, ΔT1 = t1 + t2 + t3 + t4 + t5 + t8. The multiple times t1, t2, t3, t4, t5, t8 for determining the first time interval are time lengths. In some embodiments, corresponding to the case that the alternative network device does not belong to the second service network device set and the terminal predicts the key indicator, the first time interval is determined based on the following times: a propagation time of a downlink reference signal for measuring the key indicator sent by the alternative network device; a time for the terminal to measure the key indicator; a time for the terminal to predict the key indicator; a time for the terminal to feed back the predicted key indicator to the first control device; a time for the first control device to determine the first set of serving network devices of the terminal according to the predicted key indicator; a time for the terminal to establish a connection with a serving network device in the first set of serving network devices. In some embodiments, the first time interval can be represented as ΔT1, ΔT1=t1+t2+t6+t7+t5+t8. The plurality of times t1, t2, t6, t7, t5, and t8 for determining the first time interval are time lengths. In some embodiments, before using the indicator prediction AI model, the indicator prediction AI model needs to be trained first, and the indicator prediction AI model is obtained by training an initial AI model based on first training samples. The training process can include but is not limited to: training the initial AI model based on the first training samples until the model converges to obtain the indicator prediction AI model. The input data in the training process includes the key indicator at the third time and the key indicator at at least one historical time before the third time, and the output data includes the predicted key indicator at the fourth time. The first training samples include a first training data set and a first label data set, the first training data set includes the key indicator at the third time and the key indicator at at least one historical time before the third time, and the first label data set includes the key indicator at the fourth time, the time interval between the third time and the fourth time is equal to the first time interval, and the fourth time is equal to the third time plus the first time interval. In the embodiments of the present disclosure, the key indicator at the third time and the key indicator at the fourth time need to be included in the first training samples. For example, the initial AI model can be a neural network such as a Long Short-Term Memory (LSTM) or a Gate Recurrent Unit (GRU). 102. From the set of alternative network devices, determine the first set of serving network devices of the terminal based on the key indicator at the target time. In some embodiments, the first set of serving network devices of the terminal can be the first M alternative network devices with the largest key indicators in the set of alternative network devices, and M is greater than or equal to 1. The larger the value of the key indicator, the better the performance of the system. In some embodiments, when the key indicator is a first type of indicator, the larger the value of the key indicator, the better the system performance. The candidate network devices in the candidate network device set whose ratio of all key indicators is greater than a set threshold can be determined as the first service network device set of the terminal. The key indicator ratio is the ratio of the key indicator of the candidate network device to the maximum key indicator, and the maximum key indicator is the maximum value of the key indicators of all candidate network devices in the candidate network device set. In some embodiments, when the key indicator is a second type of indicator, the smaller the value of the key indicator, the better the system performance. The first set of service network devices for the terminal can be determined by selecting candidate network devices whose ratio of all key indicators is less than a set threshold. The key indicator ratio is the ratio of the key indicators of the candidate network devices to the minimum key indicator, where the minimum key indicator is the minimum value of the key indicators of all candidate network devices in the candidate network device set. In some embodiments, candidate network devices whose ratios of all key metrics are less than a set threshold can be identified as the first set of serving network devices for the terminal. Lower values ​​for key metrics correspond to better system performance. The aforementioned method for determining service network devices can predict key indicators for a target time after the current time based on key indicators at the current moment and key indicators at historical moments. From the set of candidate network devices, the first set of service network devices for the terminal is determined based on the key indicators for the target time. Here, the key indicators are obtained by the terminal measuring the downlink reference information of each candidate network device in the candidate set. This scheme considers the impact of terminal mobility on network device selection and avoids the influence of terminal mobility on network device selection by predicting key indicators, making the selection of network devices more accurate even when the terminal is mobile. In this embodiment of the disclosure, the terminal can communicate by establishing a connection with a service network device in the first set of service network devices. In some embodiments, as shown in FIG2, a flowchart illustrating another method for determining a service network device is provided. This method can be applied to a first control device and may include, but is not limited to, the following steps: 201. Based on the terminal's current location information and historical location information, predict the target location information using a location prediction AI model. In some embodiments, when predicting the position information of the target time according to the position information of the terminal at the current time and the position information at the historical time, the position information at the current time, the position information at the historical time and the second time interval can be input into the position prediction AI model; the position information of the target time output by the position prediction AI model is obtained; and the second time interval is the interval from the second control device sending the downlink reference signal for positioning to the target time. In some embodiments, the second time interval is determined based on the following times: The time when the second control device performs positioning can be represented as t9. The time when the second control device predicts the position information of the terminal can be represented as t10. The time when the second control device sends the predicted position information of the terminal to the first control device can be represented as t11. The time when the first control device determines the set of candidate network devices based on the predicted position information of the terminal can be represented as t12. The time when the first control device instructs the candidate network devices to send the downlink reference signal for measuring the key indicators can be represented as t13. The first time interval can be represented as ΔT1. It should be noted that in the embodiments of the present disclosure, the second control device and the first control device can be the same control device or different control devices. The second control device is configured to send the downlink reference signal for positioning. When the second control device and the first control device are the same control device, t11 can be 0, and when the second control device and the first control device are not the same control device, t11 is the transmission time of the predicted position information of the terminal between the second control device and the first control device. In some embodiments, the second time interval can be represented as ΔT2, ΔT2=t9+t10+t11+t12+t13+ΔT1. In some embodiments, before using the position prediction AI model, the position prediction AI model needs to be trained first, and the position prediction AI model is obtained by training an initial AI model based on a second training sample. The input data in the training process includes the position information at the fifth time and the position information at at least one historical time before the fifth time; and the output data in the training process includes the position information at the sixth time. The second training sample includes: a second training data set and a second label data set; the second training data set includes: position information at the fifth time, position information at at least one historical time before the fifth time; the label data set includes: position information at the sixth time, a time interval between the fifth time and the sixth time is equal to the second time interval, and the sixth time is equal to the fifth time plus the second time interval. In the embodiments of the present disclosure, the position information at the fifth time and the position information at the sixth time are both in the second training sample. For example, the initial AI model can be a neural network such as LSTM or GRU. 202. From the initial network device set, the first control device determines a candidate network device set based on the position information at the target time. In some embodiments, the distance between the position information of each network device in the initial network device set and the position information of the terminal at the target time can be determined first; then the network devices in the initial network device set with a distance less than a preset distance are determined as the candidate network device set. In some embodiments, the preset distance is less than or equal to the minimum coverage radius of the network devices in the initial network device set. For example, assuming that the minimum coverage radius of the network devices in the initial network device set is R, the network devices in the initial network device set with a distance less than or equal to R between the position information of the network devices and the position information of the terminal at the target time are put into the candidate network device set, where R is greater than 0. For example, the initial network device set can be a set of network devices in a target area, the target area can be divided into a plurality of sub-areas, and the target sub-area where the position information of the terminal at the target time is located can be selected from the plurality of sub-areas, and the network devices in the target sub-area with a distance less than or equal to R between the position information of the network devices and the position information of the terminal at the target time are put into the candidate network device set. The target area can be an area within a preset radius from the position information of the terminal at the target time, and the preset radius can be set based on actual needs, which is not limited in the embodiments of the present disclosure. The range of each sub-area in the plurality of sub-areas is greater than the minimum coverage range of a single network device in the initial network device set. 203. According to the key indicators at the current time and the key indicators at the historical time, the key indicators at the target time after the current time are predicted by the index prediction AI model. 204. From the candidate network device set, a first service network device set of the terminal is determined based on the key indicators at the target time. For the description of the above steps 203 and 204, reference can be made to the related description of the above steps 101 and 102, which will not be repeated here. The determination method of the service network device can predict the position information at the target time through the index prediction AI model according to the position information of the terminal at the current time and the position information at the historical time, and perform preliminary selection on the initial network device set based on the predicted position information at the target time to obtain a candidate network device set, and then perform network device selection again based on the predicted key indicators from the candidate network set, and finally determine the first service network device set, thereby avoiding the influence of terminal movement on network device selection, and making the selection of network devices more accurate in the case of terminal movement. In the embodiments of the present disclosure, the network devices in the first service network device set can continue to provide communication services for the terminal or perform a random access process with the terminal. In some embodiments, if the service network device in the first service network device set is directly in a connected and synchronized state with the terminal, that is, belongs to the intersection of the first service network device set and the second service network device set, the selected service network device continues to provide communication services for the terminal. In some embodiments, if the service network device in the first service network device set is not in a connected and synchronized state, a random access process can be started with the terminal to enable the terminal to access the service network device. It should be understood that although each step in the flowchart involved in each of the above embodiments is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise stated herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each of the above embodiments can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with other steps or steps or stages in other steps. Based on the same technical concept, the embodiments of the present disclosure also provide a control device. The control device can implement the functions of the first control device in the above embodiments. For example, FIG. 3 is a structural schematic diagram of a control device provided by an embodiment. The control device includes a memory 301, a transceiver 302, and a processor 303, wherein the memory 301, the transceiver 302, and the processor 303 are connected through a bus interface. The memory 301 is configured to store a computer program; and the transceiver 302 is configured to transceive data under control of the processor 303. The processor 303 is configured to read the computer program in the memory 301 and perform the following operations: predict, according to the key indicator of the current time and the key indicator of the historical time, the key indicator of a target time after the current time by using an indicator prediction AI model, the key indicator of the current time and the key indicator of the historical time being obtained by the terminal measuring downlink reference information of each candidate network device in a candidate network device set; determine, from the candidate network device set, the first service network device set of the terminal based on the key indicator of the target time. In some embodiments, the processor 303 is specifically configured to read the computer program in the memory 301 and perform the following operations: The method further includes: inputting the key indicator of the current time, the key indicator of the historical time and a first time interval into the indicator prediction AI model; obtaining the key indicator of the target time output by the indicator prediction AI model; The first time interval is a time interval between a first time and a target time, the first time is a time at which a downlink reference signal for measuring the key indicator is first transmitted in the candidate network device set, and the target time is a time at which the first control device determines the first service network device set according to a current measurement result, the current measurement result being obtained by measuring the downlink reference signal transmitted at the first time. In some embodiments, the first time interval is determined based on the following times: propagation time of the downlink reference signal for measuring the key indicator transmitted by the candidate network device; time for the terminal to measure the key indicator; time for the terminal to feed back the key indicator to the first control device; time for the first control device to predict the key indicator; time for the first control device to determine the first service network device set of the terminal according to the predicted key indicator. In some embodiments, the first time interval is determined based on the following times: propagation time of the downlink reference signal for measuring the key indicator transmitted by the candidate network device; The terminal measures the key indicator; The terminal predicts the key indicator; The terminal feeds back the predicted key indicator to the first control device; The first control device determines the first service network device set of the terminal according to the predicted key indicator. In some embodiments, the first time interval corresponds to a case where the alternative network device belongs to a second service network device set, wherein the second service network device set is an original service network device set that provides services for the terminal. In some embodiments, the first time interval further includes a time spent by the terminal in establishing a connection with a service network device in the first service network device set. In some embodiments, the first time interval corresponds to a case where the alternative network device does not belong to a second service network device set, wherein the second service network device set is an original service network device set that provides services for the terminal. In some embodiments, the indicator prediction AI model is obtained by training an initial AI model based on first training samples; The first training samples include a first training data set and a first label data set. The first training data set includes a key indicator at a third time and a key indicator at at least one historical time before the third time. The first label data set includes a key indicator at a fourth time, wherein the fourth time is equal to the third time plus the first time interval. In some embodiments, the processor 303 is specifically configured to read the computer program in the memory 301 and perform the following operations: The first service network device set of the terminal is determined from the alternative network device set based on the key indicator at the target time, including: The first service network device set of the terminal is determined from the alternative network device set based on the key indicator at the target time, including: In some embodiments, the processor 303 is specifically configured to read the computer program in the memory 301 and perform the following operations: When the key indicator is a first type of indicator, the greater the value of the key indicator, the better the performance of the system. The first service network device set of the terminal is determined from the alternative network device set based on the key indicator at the target time, including: The first service network device set of the terminal is determined from the alternative network device set based on the key indicator at the target time, including: The key indicator ratio is a ratio of the key indicator of the candidate network device to a maximum key indicator, the maximum key indicator being a maximum value of the key indicators of all candidate network devices in the candidate network device set. Exemplarily, the first type of indicators can include, but are not limited to, at least one of SS-RSRP, SS-RSRQ, SS-SINR, SS-CQI, CSI-RSRP, CSI-RSRQ, CSI-SINR, and CSI-CQI. In some embodiments, when the key indicator is a second type of indicator, the smaller the value of the key indicator, the better the performance of the system, and the processor 303 is specifically configured to read the computer program in the memory 301 and perform the following operations: The first service network device set of the terminal is determined from the candidate network device set based on the key indicator at the target time. The first service network device set of the terminal is determined from the candidate network device set based on the key indicator at the target time. The key indicator ratio is a ratio of the key indicator of the candidate network device to a minimum key indicator, the minimum key indicator being a minimum value of the key indicators of all candidate network devices in the candidate network device set. Exemplarily, the first type of indicators can include, but are not limited to, at least one of SS-RSSI and CSI-RSSI. RSSI can be used to represent the interference level when the network device does not send a signal. In some embodiments, the key indicator includes at least one of the following: SS-RSRP, SS-RSRQ, SS-SINR, CSI-RSRP, CSI-RSRQ, CSI-SINR, SS-CQI, SS-RSSI, CSI-CQI, and CSI-RSSI. In some embodiments, the processor 303 is further configured to read the computer program in the memory 301 and perform the following operations: Before determining the first service network device set of the terminal from the candidate network device set based on the key indicator at the target time, the position information at the target time is predicted by a position prediction AI model according to the position information of the terminal at the current time and the position information at the historical time. The first control device determines the candidate network device set from the initial network device set based on the position information at the target time. In some embodiments, the processor 303 is specifically configured to read a computer program in the memory 301 and perform the following operations: The position information of the target time is predicted by a position prediction AI model according to the position information of the terminal at the current time and the position information at the historical time, including: The position information of the current time, the position information of the historical time and the second time interval are input into the position prediction AI model; The position information of the target time output by the position prediction AI model is obtained; The second time interval is the interval from the second control device sending a downlink reference signal for positioning to the target time. In some embodiments, the second time interval is determined based on the following times: The time when the second control device performs positioning; The time when the second control device predicts the position information of the terminal; The time when the second control device sends the predicted position information of the terminal to the first control device; The time when the first control device determines the set of alternative network devices based on the predicted position information of the terminal; The time when the first control device instructs the alternative network devices to send a downlink reference signal for measuring a key indicator; The first time interval. In some embodiments, the processor 303 is specifically configured to read a computer program in the memory 301 and perform the following operations: The set of alternative network devices is determined from the set of initial network devices based on the position information of the target time, including: The distance between the position information of each network device in the set of initial network devices and the position information of the terminal at the target time is determined; The network devices in the set of initial network devices whose distance is less than a preset distance are determined as the set of alternative network devices. In some embodiments, the preset distance is less than or equal to the minimum coverage radius of the network devices in the set of initial network devices. In some embodiments, the position prediction AI model is obtained by training an initial AI model based on a second training sample. The second training sample includes a second training data set and a second label data set; the second training data set includes position information at a fifth time and position information at at least one historical time before the fifth time; and the label data set includes position information at a sixth time, the sixth time being equal to the fifth time plus the second time interval. In one exemplary embodiment, as shown in FIG. 4, a structural block diagram of a control device is provided, which can be the first control device described above, comprising: A prediction module 401 configured to predict, according to a key indicator at a current time and a key indicator at a historical time, a key indicator at a target time after the current time by an indicator prediction AI model, the key indicators at the current time and the historical time being obtained by the terminal measuring downlink reference information of each candidate network device in a candidate network device set. A determination module 402 configured to determine, from the candidate network device set, a first service network device set of the terminal based on the key indicator at the target time. In some embodiments, the prediction module 401 is specifically configured to: input the key indicator at the current time, the key indicator at the historical time, and a first time interval into the indicator prediction AI model; obtain the key indicator at the target time output by the indicator prediction AI model; The first time interval is a time interval between a first time and a target time, the first time is a time at which the candidate network device set starts to send downlink reference signals for measuring the key indicator, and the target time is a time at which the first control device determines the first service network device set according to a current measurement result, the current measurement result being obtained by measuring the downlink reference signals sent at the first time. In some embodiments, the first time interval is determined based on the following times: a propagation time of downlink reference signals for measuring the key indicator sent by the candidate network device; a time for the terminal to measure the key indicator; a time for the terminal to feed back the key indicator to the first control device; a time for the first control device to predict the key indicator; a time for the first control device to determine the first service network device set of the terminal according to the predicted key indicator. In some embodiments, the first time interval is determined based on the following times: a propagation time of downlink reference signals for measuring the key indicator sent by the candidate network device; a time for the terminal to measure the key indicator; a time for the terminal to predict the key indicator; a time for the terminal to feed back the predicted key indicator to the first control device; a time for the first control device to determine the first service network device set of the terminal according to the predicted key indicator. In some embodiments, the first time interval corresponds to a case that the alternative network device belongs to a second service network device set, wherein the second service network device set is an original service network device set that provides service for the terminal. In some embodiments, the first time interval further includes a time for the terminal to establish a connection with a service network device in the first service network device set. In some embodiments, the first time interval corresponds to a case that the alternative network device does not belong to a second service network device set, wherein the second service network device set is an original service network device set that provides service for the terminal. In some embodiments, the indicator prediction AI model is obtained by training an initial AI model based on first training samples. In some embodiments, the first training samples include a first training data set and a first label data set; the first training data set includes a key indicator at a third time point and a key indicator at at least one historical time point before the third time point; and the first label data set includes a key indicator at a fourth time point, wherein the fourth time point is equal to the third time point plus the first time interval. In some embodiments, the determining module 402 is specifically configured to: determine the first service network device set of the terminal by selecting the M alternative network devices with the largest key indicators from the alternative network device set, where M is greater than or equal to 1. In some embodiments, when the key indicator is a first type of indicator, the larger the value of the key indicator, the better the performance of the system, and the determining module 402 is specifically configured to determine the first service network device set of the terminal by selecting the alternative network devices with a key indicator ratio greater than a set threshold from the alternative network device set. In some embodiments, the key indicator ratio is a ratio of the key indicator of the alternative network device to the maximum key indicator, and the maximum key indicator is the maximum value of the key indicators of all alternative network devices in the alternative network device set. In some embodiments, when the key indicator is a second type of indicator, the smaller the value of the key indicator, the better the performance of the system, and the determining module 402 is specifically configured to: determine, from the set of candidate network devices, a first set of serving network devices of the terminal, based on the key indicator ratio of each candidate network device in the set of candidate network devices being less than a set threshold. The key indicator ratio is a ratio of the key indicator of the candidate network device to a minimum key indicator, where the minimum key indicator is a minimum value of the key indicators of all candidate network devices in the set of candidate network devices. In some embodiments, the key indicator includes at least one of: SS-RSRP, SS-RSRQ, SS-SINR, CSI-RSRP, CSI-RSRQ, CSI-SINR, SS-CQI, SS-RSSI, CSI-CQI, and CSI-RSSI. In some embodiments, the prediction module 401 is further configured to, before the determination module 402 determines the first set of serving network devices of the terminal based on the key indicator of the target time from the set of candidate network devices, predict the location information of the target time based on the location information of the terminal at the current time and the location information of the historical time by using a location prediction AI model. The determination module 402 is further configured to determine the set of candidate network devices from an initial set of network devices based on the location information of the target time. In some embodiments, the prediction module 401 is specifically configured to: input the location information of the current time, the location information of the historical time, and a second time interval into the location prediction AI model; obtain the location information of the target time output by the location prediction AI model; The second time interval is an interval between the time when the second control device sends a downlink reference signal for positioning and the target time. In some embodiments, the second time interval is determined based on the following times: the time when the second control device performs positioning; the time when the second control device predicts the location information of the terminal; the time when the second control device sends the predicted location information of the terminal to the first control device; the time when the first control device determines the set of candidate network devices based on the predicted location information of the terminal; the time when the first control device instructs the candidate network devices to send a downlink reference signal for measuring the key indicator; a first time interval. In some embodiments, the determining module 402 is further configured to: before determining the first set of network devices serving the terminal from the set of candidate network devices based on the key indicator at the target time, determine a distance between the location information of each network device in the initial set of network devices and the location information of the terminal at the target time. The network devices in the initial set of network devices with the distance less than the preset distance are determined as the set of candidate network devices. In some embodiments, the preset distance is less than or equal to the minimum coverage radius of the network devices in the initial set of network devices. In some embodiments, the location prediction AI model is obtained by training an initial AI model based on second training samples. The second training samples include a second training data set and a second label data set. The second training data set includes location information at a fifth time and location information at at least one historical time before the fifth time. The label data set includes location information at a sixth time, and the sixth time is equal to the fifth time plus the second time interval. It should be noted that the division of the modules in the embodiments of the present disclosure is illustrative, and is merely a logical function division. In actual implementation, another division manner can be used. In addition, each functional module in each embodiment of the present disclosure can be integrated in one processing module, or each module can be physically present alone, or two or more modules can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software functional module. When the integrated module is realized in the form of a software functional module and sold or used as an independent product, the module can be stored in a processor-readable storage medium. Based on this understanding, the technical solutions of the present disclosure can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform all or part of the steps of the methods described in the various embodiments of the present disclosure. It should be noted that the above-described apparatus provided by the embodiments of the present disclosure can realize all the method steps realized by the method embodiments and achieve the same technical effects. Therefore, the same parts and beneficial effects of the method embodiments will not be described in detail. In one embodiment, a computer-readable storage medium is provided, which stores a computer program. The computer program is executed by a processor to realize all the method steps realized by the above-described method embodiments. In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements all the method steps of the above method embodiments. Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiments of each method. Any reference to memory, databases, or other media used in the embodiments provided by the present disclosure can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided by the present disclosure can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided by the present disclosure can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto. In the description of the embodiments of the present disclosure, unless explicitly defined and limited otherwise, the technical terms "mounting", "connecting", "connecting", "fixing" and the like should be understood in a broad sense, for example, can be fixedly connected, can be detachably connected, or integrated; can be mechanically connected, or electrically connected; can be directly connected, or indirectly connected through an intermediate medium; can be the internal connection of two elements or the interaction relationship between two elements. For those skilled in the art, the specific meanings of the above terms in the embodiments of the present disclosure can be understood according to the specific circumstances.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than limit them. Although the above embodiments are described with reference to the accompanying drawings, the technical solutions of the present disclosure are not limited to the above examples, and can be implemented in other specific forms. The above embodiments have been described in detail, and those skilled in the art should understand that the technical solutions recorded in the above embodiments can still be modified, or some or all of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present disclosure, and they should be covered in the scope of the claims and the specification of the present disclosure. In particular, as long as there is no structural conflict, each technical feature mentioned in each embodiment can be combined in any way. The present disclosure is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A determination method of a serving network device, the method comprising: predicting, by an index prediction AI model, a key indicator at a target time after a current time according to the key indicator at the current time and a key indicator at a historical time, the key indicators at the current time and the historical time being obtained by a terminal measuring downlink reference information of each candidate network device in a candidate network device set; determining, from the candidate network device set, a first serving network device set of the terminal based on the key indicator at the target time.

2. The method of claim 1, wherein, The method further comprises: inputting the key indicator at the current time, the key indicator at the historical time and a first time interval into the index prediction AI model; obtaining the key indicator at the target time output by the index prediction AI model. The first time interval is a time interval between a first time and the target time, the first time is a time at which a candidate network device in the candidate network device set starts to send a downlink reference signal for measuring the key indicator, and the target time is a time at which a first control device determines the first serving network device set according to a current measurement result, the current measurement result being obtained based on the downlink reference signal sent at the first time.

3. The method of claim 2, wherein: the first time interval is determined based on: a propagation time of a downlink reference signal for measuring the key indicator sent by the candidate network device; a time at which the terminal measures the key indicator; a time at which the terminal feeds back the key indicator to the first control device; a time at which the first control device predicts the key indicator; a time at which the first control device determines the first serving network device set of the terminal according to the predicted key indicator; or the first time interval is determined based on: a propagation time of a downlink reference signal for measuring the key indicator sent by the candidate network device; a time at which the terminal measures the key indicator; a time at which the terminal predicts the key indicator; a time at which the terminal feeds back the predicted key indicator to the first control device; a time at which the first control device determines the first serving network device set of the terminal according to the predicted key indicator. The first time interval corresponds to a case where the candidate network device belongs to a second serving network device set, the second serving network device set being a previous serving network device set of the terminal.

4. The method of claim 3, wherein, The first time interval further includes a time spent by the terminal in establishing a connection with a serving network device in the first serving network device set.

5. The method of claim 4, wherein, The first time interval corresponds to a case where the candidate network device does not belong to a second serving network device set, the second serving network device set being a previous serving network device set of the terminal.

6. The method of claim 3, wherein, The index prediction AI model is obtained by training an initial AI model based on first training samples.

7. The method according to any one of claims 2 to 6, wherein, ​ The first training sample includes: a first training dataset and a first label dataset; the first training dataset includes: key indicators at a third time step, and key indicators at at least one historical time step prior to the third time step; the first label dataset includes: key indicators at a fourth time step, wherein the fourth time step is equal to the third time step plus the first time interval.

8. The method according to any one of claims 1 to 7, wherein, The step of determining the first set of serving network devices for the terminal from the set of candidate network devices based on key indicators at the target time includes: The top M candidate network devices with the highest key indicators from the candidate network device set are selected as the first service network device set for the terminal, where M is greater than or equal to 1.

9. The method according to any one of claims 1 to 7, wherein, When the key indicator is a first-type indicator, the larger the value of the key indicator, the better the system performance. The step of determining the first set of serving network devices for the terminal from the set of candidate network devices based on the key indicator at the target time includes: The candidate network devices in the candidate network device set whose ratio of all key indicators is greater than a set threshold are determined as the first service network device set for the terminal. The key indicator ratio is the ratio of the key indicator of the candidate network device to the maximum key indicator, where the maximum key indicator is the maximum value of the key indicators of all candidate network devices in the candidate network device set.

10. The method according to any one of claims 1 to 7, wherein, When the key indicator is a second type of indicator, the smaller the value of the key indicator, the better the system performance. The step of determining the first set of serving network devices for the terminal from the set of candidate network devices based on the key indicator at the target time includes: The first set of serving network devices for the terminal is determined by selecting the candidate network devices whose ratio of all key indicators is less than a set threshold. The key indicator ratio is the ratio of the key indicator of the candidate network device to the minimum key indicator, where the minimum key indicator is the minimum value of the key indicators of all candidate network devices in the candidate network device set.

11. The method according to any one of claims 1 to 10, wherein, The key indicators include at least one of the following: SS-RSRP, SS-RSRQ, SS-SINR, CSI-RSRP, CSI-RSRQ, CSI-SINR, SS-CQI, SS-RSSI, CSI-CQI, and CSI-RSSI.

12. The method according to any one of claims 1 to 11, wherein, Before determining the first set of serving network devices for the terminal from the set of candidate network devices based on key indicators at the target time, the method further includes: Based on the terminal's location information at the current time and its location information at historical times, the location information at the target time is predicted using a location prediction AI model; The first control device determines the set of candidate network devices from the initial set of network devices based on the location information at the target time.

13. The method of claim 12, wherein, The step of predicting the location information of the target time using a location prediction AI model based on the terminal's location information at the current time and its location information at historical times includes: inputting the position information of the current moment, the position information of the historical moment, and a second time interval to the position prediction AI model; obtaining the position information of the target moment output by the position prediction AI model; The second time interval is determined based on the following times:

14. The method of claim 13, wherein, The time when the second control device performs positioning; The time when the second control device predicts the position information of the terminal; The time when the second control device sends the predicted position information of the terminal to the first control device; The time when the first control device determines the set of alternative network devices based on the predicted position information of the terminal; The time when the first control device instructs the set of alternative network devices to send downlink reference signals for measuring key indicators; The first time interval. The determination of the set of alternative network devices from the initial set of network devices based on the position information of the target moment includes:

15. The method according to any one of claims 12 to 14, wherein, Determining the distance between the position information of each network device in the initial set of network devices and the position information of the terminal at the target moment; Determining the network devices in the initial set of network devices whose distance is less than a preset distance as the set of alternative network devices. The preset distance is less than or equal to the minimum coverage radius of the network devices in the initial set of network devices.

16. The method of claim 15, wherein, The position prediction AI model is obtained by training an initial AI model based on a second training sample; 17. The method of claim 13 or 14, wherein, The second training sample includes a second training data set and a second label data set; the second training data set includes position information at a fifth moment and position information at at least one historical moment before the fifth moment; the label data set includes position information at a sixth moment, and the sixth moment is equal to the fifth moment plus the second time interval. Memory, transceiver, processor:

18. A control device comprising: The memory is used to store computer programs; the transceiver is used to transceive data under the control of the processor; and the processor is used to read the computer programs in the memory and perform the following operations: According to the key indicators at the current moment and the historical moment, the key indicators at the target moment after the current moment are predicted by an indicator prediction AI model, and the key indicators at the current moment and the historical moment are obtained by the terminal measuring the downlink reference information of each alternative network device in the set of alternative network devices; From the set of alternative network devices, the first service network device set of the terminal is determined based on the key indicators at the target moment. The processor is specifically configured to read the computer programs in the memory and perform the following operations:

19. The control device of claim 18, wherein, Input the key indicators at the current moment, the key indicators at the historical moment, and the first time interval to the indicator prediction AI model; Obtain the key indicators at the target moment output by the indicator prediction AI model; ​ The first time interval is a time interval between a first time and a target time, the first time is a time at which a downlink reference signal for performing the key indicator measurement is earliest transmitted in the set of candidate network devices, and the target time is a time at which the first control device determines the set of first service network devices according to a current measurement result, the current measurement result being obtained by measuring the downlink reference signal transmitted at the first time.

20. The control device according to claim 18 or 19, wherein The processor is specifically configured to read a computer program in the memory and perform the following operations: The first M candidate network devices in the set of candidate network devices with the largest key indicators are determined as the set of first service network devices of the terminal, and M is greater than or equal to 1.

21. The control device of claim 19, wherein, When the key indicator is a first type of indicator, the larger the value of the key indicator, the better the performance of the system, and the processor is specifically configured to read a computer program in the memory and perform the following operations: The candidate network device with a key indicator ratio greater than a set threshold in the set of candidate network devices is determined as the set of first service network devices of the terminal. The key indicator ratio is a ratio of the key indicator of the candidate network device to the largest key indicator.

22. The control device according to any one of claims 18 to 21, wherein When the key indicator is a second type of indicator, the smaller the value of the key indicator, the better the performance of the system, and the processor is specifically configured to read a computer program in the memory and perform the following operations: The candidate network device with a key indicator ratio less than a set threshold in the set of candidate network devices is determined as the set of first service network devices of the terminal. The key indicator ratio is a ratio of the key indicator of the candidate network device to the smallest key indicator, and the smallest key indicator is a minimum value of the key indicators of all candidate network devices in the set of candidate network devices.

23. A control device, comprising: a prediction module configured to predict, by an indicator prediction AI model, a key indicator at a target time after a current time according to the key indicator at the current time and a key indicator at a historical time, the key indicators at the current time and the historical time being obtained by a terminal measuring downlink reference information of each candidate network device in a set of candidate network devices; a determination module configured to determine, from the set of candidate network devices, a set of first service network devices of the terminal based on the key indicator at the target time.

24. A computer readable storage medium, wherein, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the method in any one of claims 1 to 17.

Images