Reconstruction network selection method and device, electronic equipment, chip and medium

Abstract

The invention relates to a network selection reconstruction method and device, electronic equipment, a chip and a medium, and the method comprises the steps: obtaining a plurality of historical frequency points and network selection related information of terminal equipment in a network selection reconstruction process of the terminal equipment; the network selection related information comprises a current service cell identifier; inputting the plurality of historical frequency points and the network selection related information into a network selection recommendation model, and obtaining at least one candidate frequency point selected from the plurality of historical frequency points by the network selection recommendation model in combination with the network selection related information; performing cell measurement and cell reselection processing in combination with the at least one candidate frequency point to obtain a target cell having an adjacent cell relationship with a current service cell corresponding to the current service cell identifier; performing reconstruction processing based on the target cell; wherein the selected candidate frequency points are the frequency points capable of measuring the cell, the number of the frequency points needing to be processed is reduced, the reconstruction network selection time delay is shortened, and therefore the reconstruction network selection efficiency of the terminal equipment is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technology, and in particular to a method, device, electronic device, chip and medium for reconstructing network selection. Background Art

[0002] The current reconstruction network selection mechanism mainly includes: the terminal device obtains multiple historical frequency points; performs cell measurement processing on each of the multiple historical frequency points; and selects a cell from the multiple cells obtained by the measurement processing for reconstruction processing.

[0003] In the above solution, it is necessary to perform cell measurement processing on each of the multiple historical frequency points, which results in a long network selection reconstruction delay, thereby reducing the network selection reconstruction efficiency of the terminal device. Summary of the invention

[0004] The present disclosure provides a network selection reconstruction method, device, electronic device, chip and medium.

[0005] According to a first aspect of an embodiment of the present disclosure, a method for reconstructing network selection is provided, the method comprising: in a process of reconstructing network selection of a terminal device, obtaining a plurality of historical frequency points and network selection related information of the terminal device; the network selection related information comprises a current serving cell identifier; inputting the plurality of historical frequency points and the network selection related information into a network selection recommendation model, obtaining at least one candidate frequency point selected from the plurality of historical frequency points by the network selection recommendation model in combination with the network selection related information; performing cell measurement and cell reselection processing in combination with at least one of the candidate frequency points, and obtaining a target cell having a neighboring relationship with the current serving cell corresponding to the current serving cell identifier; and performing reconstruction processing based on the target cell.

[0006] In one embodiment of the present disclosure, the cell measurement and cell reselection processing are performed in combination with at least one of the candidate frequencies to obtain a target cell that has a neighboring relationship with the current service cell corresponding to the current service cell identifier, including: performing cell measurement processing in combination with at least one of the candidate frequencies to obtain multiple candidate cells; inputting the multiple candidate cells and the network selection related information into the network selection recommendation model, and obtaining the target cell selected by the network selection recommendation model in combination with the network selection related information from the multiple candidate cells.

[0007] In one embodiment of the present disclosure, before inputting the multiple historical frequency points and the network selection related information into a network selection recommendation model, and obtaining at least one candidate frequency point selected by the network selection recommendation model in combination with the network selection related information from the multiple historical frequency points, the method also includes: determining whether to trigger a reconstruction network selection mechanism based on the network selection recommendation model, and determining whether the network selection recommendation model has a network selection recommendation capability; in a case where the reconstruction network selection mechanism based on the network selection recommendation model is not triggered, or in a case where the network selection recommendation model does not have a network selection recommendation capability, performing cell measurement and cell reselection processing in combination with the multiple historical frequency points, and performing reconstruction processing based on the cell obtained by the selection processing.

[0008] In one embodiment of the present disclosure, determining whether the network selection recommendation model has the network selection recommendation capability includes: determining whether the network selection recommendation model has been trained, and determining whether the accuracy of the network selection recommendation model meets a preset accuracy condition; if the network selection recommendation model has been trained, and the accuracy of the network selection recommendation model meets the preset accuracy condition, determining that the network selection recommendation model has the network selection recommendation capability; if the network selection recommendation model has not been trained, or the accuracy of the network selection recommendation model does not meet the preset accuracy condition, determining that the network selection recommendation model does not have the network selection recommendation capability.

[0009] In one embodiment of the present disclosure, the method also includes: obtaining historical reconstructed network selection data of the terminal device; sending the historical reconstructed network selection data to the cloud, and receiving network selection recommendation model parameters returned by the cloud; the network selection recommendation model parameters are parameters of the network selection recommendation model obtained after the cloud combines the historical reconstructed network selection data of multiple terminal devices to train the initial network selection recommendation model; according to the network selection recommendation model parameters, the network selection recommendation model of the terminal device is updated.

[0010] In one embodiment of the present disclosure, the reconstruction processing based on the target cell includes: obtaining the terminal context of the terminal device in the network device corresponding to the target cell; the terminal context is obtained by the interaction between the network device corresponding to the target cell and the network device corresponding to the current service cell; and reconstruction processing is performed based on the target cell and the terminal context.

[0011] In one embodiment of the present disclosure, the network selection related information also includes at least one of the following: the public land mobile network PLMN identifier of the current service cell, the frequency of the current service cell, cell resource information, the capabilities of the terminal device, the location of the terminal device, and the moving speed of the terminal device.

[0012] According to the second aspect of the embodiment of the present disclosure, a training method for reconstructing a network selection model is also provided, the method comprising: obtaining historical reconstructed network selection data of multiple terminal devices; obtaining an initial network selection recommendation model; training the initial network selection recommendation model in combination with the historical reconstructed network selection data to obtain a trained network selection recommendation model.

[0013] In one embodiment of the present disclosure, the historical reconstruction network selection data includes at least one of the following parameters in the historical reconstruction network selection process: multiple historical frequency points, network selection related information, at least one historical candidate frequency point selected in combination with the historical frequency points, multiple historical candidate cells obtained by performing cell measurements in combination with at least one of the historical candidate frequencies, and historical target cells selected in combination with the multiple historical candidate cells; the reconstruction processing result of the historical reconstruction network selection process is successful reconstruction of the network selection.

[0014] In one embodiment of the present disclosure, the initial network selection recommendation model is trained in combination with the historical reconstructed network selection data to obtain a trained network selection recommendation model, including: inputting the multiple historical frequency points and the network selection related information into the initial network selection recommendation model to obtain at least one predicted frequency point output by the network selection recommendation model; inputting the multiple historical candidate cells and the network selection related information into the network selection recommendation model to obtain a predicted target cell output by the network selection recommendation model; combining at least one of the predicted frequency points, at least one of the historical candidate frequency points and a first loss function to determine the value of the first loss function; combining the predicted target cell, the historical target cell and the second loss function to determine the value of the second loss function; and adjusting parameters of the network selection recommendation model according to the value of the first loss function and the value of the second loss function to achieve training.

[0015] According to the third aspect of the embodiment of the present disclosure, a reconstruction network selection device is also provided, and the device includes: a first acquisition module, used to obtain multiple historical frequency points and network selection related information of a terminal device during the reconstruction network selection process of the terminal device; the network selection related information includes a current service cell identifier; a second acquisition module, used to input the multiple historical frequency points and the network selection related information into a network selection recommendation model, and obtain at least one candidate frequency point selected by the network selection recommendation model in combination with the network selection related information from the multiple historical frequency points; a measurement selection module, used to perform cell measurement and cell reselection processing in combination with at least one of the candidate frequencies, and obtain a target cell that has a neighboring relationship with the current service cell corresponding to the current service cell identifier; a reconstruction processing module, used to perform reconstruction processing based on the target cell.

[0016] In one embodiment of the present disclosure, the measurement selection module is specifically used to perform cell measurement processing in combination with at least one of the candidate frequency points to obtain multiple candidate cells; input the multiple candidate cells and the network selection related information into the network selection recommendation model, and obtain the target cell selected by the network selection recommendation model in combination with the network selection related information from the multiple candidate cells.

[0017] In one embodiment of the present disclosure, the device also includes: a determination module, used to determine whether to trigger a reconstruction network selection mechanism based on a network selection recommendation model, and determine whether the network selection recommendation model has a network selection recommendation capability; the measurement selection module is also used to perform cell measurement and cell reselection processing in combination with the multiple historical frequency points when the reconstruction network selection mechanism based on the network selection recommendation model is not triggered, or when the network selection recommendation model does not have a network selection recommendation capability, and perform reconstruction processing based on the cell obtained by the selection processing.

[0018] In one embodiment of the present disclosure, the determination module is specifically used to determine whether the network selection recommendation model has been trained and whether the accuracy of the network selection recommendation model meets the preset accuracy condition; when the network selection recommendation model has been trained and the accuracy of the network selection recommendation model meets the preset accuracy condition, determine that the network selection recommendation model has the network selection recommendation capability; when the network selection recommendation model has not been trained or the accuracy of the network selection recommendation model does not meet the preset accuracy condition, determine that the network selection recommendation model does not have the network selection recommendation capability.

[0019] In one embodiment of the present disclosure, the device also includes: a third acquisition module, a sending and receiving module and a parameter updating module; the third acquisition module is used to obtain the historical reconstructed network selection data of the terminal device; the sending and receiving module is used to send the historical reconstructed network selection data to the cloud, and receive the network selection recommendation model parameters returned by the cloud; the network selection recommendation model parameters are the parameters of the network selection recommendation model obtained after the cloud combines the historical reconstructed network selection data of multiple terminal devices to train the initial network selection recommendation model; the parameter updating module is used to perform parameter update processing on the network selection recommendation model of the terminal device according to the network selection recommendation model parameters.

[0020] In one embodiment of the present disclosure, the reconstruction processing module is specifically used to obtain the terminal context of the terminal device in the network device corresponding to the target cell; the terminal context is obtained by the interaction between the network device corresponding to the target cell and the network device corresponding to the current service cell; and reconstruction processing is performed based on the target cell and the terminal context.

[0021] In one embodiment of the present disclosure, the network selection related information also includes at least one of the following: the public land mobile network PLMN identifier of the current service cell, the frequency of the current service cell, cell resource information, the capabilities of the terminal device, the location of the terminal device, and the moving speed of the terminal device.

[0022] According to the fourth aspect of the embodiment of the present disclosure, a training device for reconstructing a network selection model is also provided, the device comprising: a first acquisition module, used to obtain historical reconstructed network selection data of multiple terminal devices; a second acquisition module, used to obtain an initial network selection recommendation model; a training processing module, used to train the initial network selection recommendation model in combination with the historical reconstructed network selection data to obtain a trained network selection recommendation model.

[0023] In one embodiment of the present disclosure, the historical reconstruction network selection data includes at least one of the following parameters in the historical reconstruction network selection process: multiple historical frequency points, network selection related information, at least one historical candidate frequency point selected in combination with the historical frequency points, multiple historical candidate cells obtained by performing cell measurements in combination with at least one of the historical candidate frequencies, and historical target cells selected in combination with the multiple historical candidate cells; the reconstruction processing result of the historical reconstruction network selection process is successful reconstruction of the network selection.

[0024] In one embodiment of the present disclosure, the training processing module is specifically used to input the multiple historical frequency points and the network selection related information into the initial network selection recommendation model to obtain at least one predicted frequency point output by the network selection recommendation model; input the multiple historical candidate cells and the network selection related information into the network selection recommendation model to obtain the predicted target cell output by the network selection recommendation model; determine the value of the first loss function in combination with at least one of the predicted frequency points, at least one of the historical candidate frequency points and the first loss function; determine the value of the second loss function in combination with the predicted target cell, the historical target cell and the second loss function; and perform parameter adjustment processing on the network selection recommendation model according to the value of the first loss function and the value of the second loss function to achieve training.

[0025] According to the fifth aspect of an embodiment of the present disclosure, an electronic device is also provided, comprising: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to: implement the steps of the network reconstruction method as described above; or, implement the steps of the network reconstruction model training method as described above.

[0026] According to the sixth aspect of an embodiment of the present disclosure, a non-temporary computer-readable storage medium is also provided. When the instructions in the storage medium are executed by a processor, the processor is enabled to execute the reconstruction network selection method as described above; or, execute the training method of the reconstruction network selection model as described above.

[0027] According to the seventh aspect of the embodiments of the present disclosure, a chip is also provided, comprising one or more interface circuits and one or more processors; the interface circuit is used to receive a signal, the signal comprising a computer instruction, and when the processor executes the computer instruction, the chip executes the reconstruction network selection method as described above; or, executes the training method of the reconstruction network selection model as described above.

[0028] The technical solution provided by the embodiments of the present disclosure brings at least the following beneficial effects:

[0029] During the reconstruction network selection process of the terminal device, multiple historical frequency points and network selection related information of the terminal device are obtained; the network selection related information includes the current service cell identifier; the multiple historical frequency points and network selection related information are input into the network selection recommendation model, and at least one candidate frequency point selected by the network selection recommendation model in combination with the network selection related information from the multiple historical frequency points is obtained; cell measurement and cell reselection processing are performed in combination with the at least one candidate frequency point to obtain a target cell that has a neighboring relationship with the current service cell corresponding to the current service cell identifier; reconstruction processing is performed based on the target cell; wherein the selected candidate frequency point is a frequency point that can measure the cell, thereby reducing the number of frequency points that need to be processed, shortening the reconstruction network selection delay, and thus improving the reconstruction network selection efficiency of the terminal device.

[0030] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute improper limitations on the present disclosure.

[0032] Figure 1 A flowchart of a network selection reconstruction method according to an embodiment of the present disclosure;

[0033] Figure 2 A flowchart of a network selection reconstruction method according to another embodiment of the present disclosure;

[0034] Figure 3 A flowchart of a network selection reconstruction method according to another embodiment of the present disclosure;

[0035] Figure 4 Schematic diagram of network selection for reconstruction;

[0036] Figure 5 A flowchart of a training method for reconstructing a network selection model according to an embodiment of the present disclosure;

[0037] Figure 6A schematic diagram of the structure of a network selection reconstruction device according to an embodiment of the present disclosure;

[0038] Figure 7 A schematic diagram of the structure of a training device for reconstructing a network selection model according to an embodiment of the present disclosure;

[0039] Figure 8 is a structural block diagram of an electronic device according to an exemplary embodiment of the present disclosure;

[0040] Fig. 9 The figure is a schematic diagram of the structure of a chip according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0041] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings.

[0042] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0043] The current reconstruction network selection mechanism mainly includes: the terminal device obtains multiple historical frequency points; performs cell measurement processing on each of the multiple historical frequency points; and selects a cell from the multiple cells obtained by the measurement processing for reconstruction processing.

[0044] In the above solution, it is necessary to perform cell measurement processing on each of the multiple historical frequency points, which results in a long network selection reconstruction delay, thereby reducing the network selection reconstruction efficiency of the terminal device.

[0045] Figure 1 The flowchart of the network selection reconstruction method of one embodiment of the present disclosure is as follows. It should be noted that the network selection reconstruction method of the present embodiment can be applied to a network selection reconstruction device, which can be configured in an electronic device or chip so that the electronic device or chip can perform the network selection reconstruction function.

[0046] Among them, the electronic device can be any device with computing capabilities, such as a personal computer (PC), a mobile terminal, a terminal device, a server, etc. The mobile terminal can be, for example, a vehicle-mounted device, a mobile phone, a tablet computer, a personal digital assistant, a wearable device, and other hardware devices with various operating systems, touch screens and / or display screens.

[0047] In addition, the network selection reconstruction device may also be software in an electronic device, etc. The software may be, for example, network selection reconstruction software, etc. In the following embodiments, the execution subject is a terminal device as an example for description.

[0048] like Figure 1 As shown, the method comprises the following steps:

[0049] Step 101, during the process of re-establishing network selection of the terminal device, a plurality of historical frequency points and network selection related information of the terminal device are obtained; the network selection related information includes a current serving cell identifier.

[0050] In the embodiment of the present disclosure, the historical frequency point of the terminal device may be a frequency point used in the cell measurement process, and / or may be a frequency point preset by the operator to which the number in the terminal device belongs.

[0051] In the disclosed embodiment, the network selection related information of the terminal device may include, in addition to the current serving cell identifier, at least one of the following parameters: the public land mobile network (PLMN) identifier of the current serving cell, the frequency of the current serving cell, cell resource information, the capabilities of the terminal device, the location of the terminal device, and the moving speed of the terminal device.

[0052] The cell resource information refers to various resource conditions related to communication services in the cell, such as spectrum resources, channel resources, network equipment resources, user capacity, and service quality parameters.

[0053] The capabilities of a terminal device refer to the various functions and performance parameters that the terminal device possesses during the communication process, such as communication capability and processing capability.

[0054] Step 102: input a plurality of historical frequency points and network selection related information into a network selection recommendation model, and obtain at least one candidate frequency point selected by the network selection recommendation model in combination with the network selection related information from the plurality of historical frequency points.

[0055] In the disclosed embodiment, before step 102, the terminal device may also perform the following process: obtaining the historical reconstruction network selection data of the terminal device; sending the historical reconstruction network selection data to the cloud, and receiving the network selection recommendation model parameters returned by the cloud; the network selection recommendation model parameters are the parameters of the network selection recommendation model obtained after the cloud combines the historical reconstruction network selection data of multiple terminal devices to train the initial network selection recommendation model; according to the network selection recommendation model parameters, the network selection recommendation model of the terminal device is updated. Among them, the initial network selection recommendation model can be, for example, an artificial intelligence model such as a neural network model, a reinforcement learning model, a random forest model, a model based on a machine learning algorithm, etc. The above-mentioned artificial intelligence model can perform feature extraction and decision processing on the input data.

[0056] Among them, after the network selection recommendation model is updated with parameters, the cloud can use at least one of the following two examples to determine whether the network selection recommendation model has been trained; if the network selection recommendation model has been trained, the network selection recommendation model can be used to perform network selection recommendation processing.

[0057] In one example, the cloud can obtain test reconstruction network selection data, which can be selected from historical reconstruction network selection data or preset. The network selection recommendation model is tested using the test reconstruction network selection data to obtain the test accuracy; when the test accuracy is greater than or equal to the specified accuracy threshold, it is determined that the network selection recommendation model training is completed.

[0058] In another example, the cloud can count the training rounds of the network selection recommendation model; when the training rounds are greater than or equal to the specified training rounds, it is determined that the training of the network selection recommendation model is completed.

[0059] The historical reconstruction network selection data of the terminal device may be data of the terminal device in the previous reconstruction network selection process. The historical reconstruction network selection data may include at least one of the following parameters: multiple historical frequency points, network selection related information, at least one historical candidate frequency point selected in combination with the historical frequency point, multiple historical candidate cells obtained by performing cell measurement in combination with at least one historical candidate frequency point, and historical target cells selected in combination with multiple historical candidate cells.

[0060] It should be noted that the historical reconstruction network selection data may be the reconstruction network selection data determined in combination with the network selection recommendation model, and / or may be the reconstruction network selection data determined in combination with the existing reconstruction network selection strategy, which is not specifically limited here.

[0061] Specifically, the cloud can combine multiple historical frequency points in the historical reconstructed network selection data, network selection related information, and at least one historical candidate frequency point obtained by historical frequency point selection to train the network selection recommendation model to obtain a trained network selection recommendation model, so that the trained network selection recommendation model has a frequency point recommendation function.

[0062] Among them, the cloud side combines the historical reconstruction network selection data of multiple terminal devices to train the network selection recommendation model, so that the network selection recommendation model can learn how to select at least one frequency point from multiple historical frequency points, and the at least one frequency point can measure the cell that has a neighboring relationship with the current service cell; then, among the multiple cells measured by the at least one frequency point, the proportion of cells that have a neighboring relationship with the current service cell can be increased, and then when selecting the target cell from the multiple cells, the probability that the selected target cell has a neighboring relationship with the current service cell can be increased, thereby reducing the probability of reconstruction failure and further improving the reconstruction network selection efficiency of the terminal device.

[0063] Step 103: Perform cell measurement and cell reselection processing in combination with at least one candidate frequency point to obtain a target cell that has a neighboring cell relationship with the current serving cell corresponding to the current serving cell identifier.

[0064] In an embodiment of the present disclosure, in one example, the process of the terminal device executing step 103 may be, for example, performing cell measurement processing in combination with at least one candidate frequency point to obtain multiple candidate cells; and arbitrarily selecting a cell from the multiple candidate cells as a target cell.

[0065] In another example, the process of the terminal device executing step 103 may be, for example, performing cell measurement processing in combination with at least one candidate frequency point to obtain multiple candidate cells; sorting the multiple candidate cells according to cell-related parameters to obtain a sorting result; and selecting the cell ranked first from the sorting result as the target cell.

[0066] Among them, cell-related parameters such as cell resources, number of cell users, cell signal quality, etc. can be set according to actual needs.

[0067] Step 104: Perform reconstruction processing based on the target cell.

[0068] In the embodiment of the present disclosure, the process of the terminal device executing step 104 may, for example, be to obtain the terminal context of the terminal device in the network device corresponding to the target cell; the terminal context is obtained by the interaction between the network device corresponding to the target cell and the network device corresponding to the current service cell; and reconstruction processing is performed based on the target cell and the terminal context.

[0069] The terminal context is the data needed by the terminal device during the reconstruction process; if the terminal device does not obtain the terminal context, the reconstruction process may fail. The network device corresponding to the target cell that has a neighboring cell relationship with the current serving cell can interact with the network device corresponding to the current serving cell to obtain the terminal context stored in the network device corresponding to the current serving cell, and provide it to the terminal device for reconstruction processing.

[0070] In the network selection reconstruction method of the disclosed embodiment, during the network selection reconstruction process of the terminal device, multiple historical frequency points and network selection related information are obtained; the network selection related information includes the current service cell identifier; the multiple historical frequency points and network selection related information are input into a network selection recommendation model to obtain at least one candidate frequency point selected from the multiple historical frequency points by the network selection recommendation model in combination with the network selection related information; cell measurement and cell reselection processing are performed in combination with the at least one candidate frequency point to obtain a target cell that has a neighboring relationship with the current service cell corresponding to the current service cell identifier; reconstruction processing is performed based on the target cell; wherein the selected candidate frequency point is a frequency point that can be measured to obtain the cell, thereby reducing the number of frequency points that need to be processed, shortening the network selection reconstruction delay, and thus improving the network selection reconstruction efficiency of the terminal device.

[0071] Figure 2 The flowchart of the network selection reconstruction method of another embodiment of the present disclosure is as follows. It should be noted that the network selection reconstruction method of the present embodiment can be applied to a network selection reconstruction device, which can be configured in an electronic device or chip so that the electronic device or chip can perform the network selection reconstruction function.

[0072] Among them, the electronic device can be any device with computing capabilities, such as a personal computer (PC), a mobile terminal, a terminal device, a server, a controller in a vehicle, etc. The mobile terminal can be, for example, a vehicle-mounted device, a mobile phone, a tablet computer, a personal digital assistant, a wearable device, and other hardware devices with various operating systems, touch screens and / or display screens.

[0073] In addition, the network selection reconstruction device may also be software in an electronic device, etc. The software may be, for example, network selection reconstruction software, etc. In the following embodiments, the execution subject is a terminal device as an example for description.

[0074] like Figure 2 As shown, the method comprises the following steps:

[0075] Step 201, during the process of re-establishing network selection of the terminal device, a plurality of historical frequency points and network selection related information of the terminal device are obtained; the network selection related information includes a current serving cell identifier.

[0076] Step 202: input a plurality of historical frequency points and network selection related information into a network selection recommendation model, and obtain at least one candidate frequency point selected by the network selection recommendation model in combination with the network selection related information from the plurality of historical frequency points.

[0077] Step 203: Perform cell measurement processing in combination with at least one candidate frequency point to obtain multiple candidate cells.

[0078] In an embodiment of the present disclosure, the terminal device may perform cell measurement processing on each candidate frequency point of at least one candidate frequency point in turn to obtain at least one candidate cell under the candidate frequency point.

[0079] Step 204, multiple candidate cells and network selection related information are input into a network selection recommendation model, and a target cell is selected from the multiple candidate cells by the network selection recommendation model in combination with the network selection related information; there is a neighboring cell relationship between the target cell and the current serving cell corresponding to the current serving cell identifier.

[0080] In the disclosed embodiment, the network selection recommendation model can be obtained by training the cloud in combination with the historical reconstruction network selection data of multiple terminal devices. Specifically, the cloud can train the network selection recommendation model in combination with multiple historical candidate cells, network selection related information, and historical target cells in the historical reconstruction network selection data, so that the trained network selection recommendation model has a cell recommendation function.

[0081] The network selection recommendation model can also be obtained by training the terminal device in combination with the historical reconstruction network selection data of multiple terminal devices. The terminal device can obtain the historical reconstruction network selection data of multiple terminal devices through interaction with other terminal devices; and / or the terminal device can obtain the historical reconstruction network selection data of multiple terminal devices communicating with the network device through interaction with the network device. The multiple terminal devices mentioned here may include the terminal device itself that executes the reconstruction network selection method, or may not include the terminal device itself that executes the reconstruction network selection method, and can be set according to actual needs.

[0082] Among them, it should be noted that the training process of the network selection recommendation model may include the following steps: inputting multiple historical frequency points and network selection related information into the initial network selection recommendation model to obtain at least one predicted frequency point output by the network selection recommendation model; inputting multiple historical candidate cells and network selection related information into the network selection recommendation model to obtain the predicted target cell output by the network selection recommendation model; combining at least one predicted frequency point, at least one historical candidate frequency point and a first loss function to determine the value of the first loss function; combining the predicted target cell, the historical target cell and the second loss function to determine the value of the second loss function; according to the value of the first loss function and the value of the second loss function, adjusting the parameters of the network selection recommendation model to achieve training.

[0083] Among them, the network selection recommendation model trained through the above training process can learn how to select at least one frequency point from multiple historical frequency points, and can also learn how to select a target cell that has a neighboring relationship with the current service cell from multiple candidate cells, thereby increasing the probability of selecting a target cell that has a neighboring relationship with the current service cell, avoiding reconstruction failure, and further improving the reconstruction network selection efficiency of the terminal equipment.

[0084] In the embodiment of the present disclosure, the target cell and the current serving cell have a neighboring cell relationship, which means that the target cell and the current serving cell are adjacent and have a handover relationship.

[0085] In the embodiment of the present disclosure, as an alternative to steps 202 to 204, the terminal device may perform the following steps: perform cell measurement processing in combination with multiple historical frequency points to obtain multiple candidate cells; input the multiple candidate cells and network selection related information into a network selection recommendation model, and obtain a target cell selected from the multiple candidate cells by the network selection recommendation model in combination with the network selection related information.

[0086] Among them, this replacement scheme is relative to Figure 2 In the illustrated embodiment, the network selection reconstruction delay is relatively high, and the network selection reconstruction efficiency is consistent.

[0087] Step 205: Perform reconstruction processing based on the target cell.

[0088] It should be noted that the detailed description of steps 201 to 202 and step 205 can be found in Figure 1 Steps 101 to 102 and step 104 in the illustrated embodiment will not be described in detail herein.

[0089] In the network selection reconstruction method of the disclosed embodiment, during the network selection reconstruction process of the terminal device, multiple historical frequency points and network selection related information are obtained; the network selection related information includes the current service cell identifier; the multiple historical frequency points and network selection related information are input into a network selection recommendation model, and at least one candidate frequency point selected by the network selection recommendation model in combination with the network selection related information from the multiple historical frequency points is obtained; cell measurement processing is performed in combination with at least one candidate frequency point to obtain multiple candidate cells; the multiple candidate cells and network selection related information are input into the network selection recommendation model, and a target cell selected by the network selection recommendation model in combination with the network selection related information from the multiple candidate cells is obtained; there is a neighboring cell relationship between the target cell and the current service cell corresponding to the current service cell identifier; reconstruction processing is performed based on the target cell; wherein the selected candidate frequency point is a frequency point that can be measured to obtain the cell, reducing the number of frequency points that need to be processed and shortening the network selection reconstruction delay; the selected target cell is a cell that has a neighboring cell relationship with the current service cell, thereby further improving the network selection reconstruction efficiency of the terminal device.

[0090] Figure 3 The flowchart of the network selection reconstruction method of another embodiment of the present disclosure is as follows. It should be noted that the network selection reconstruction method of the present embodiment can be applied to a network selection reconstruction device, which can be configured in an electronic device or chip so that the electronic device or chip can perform the network selection reconstruction function.

[0091] Among them, the electronic device can be any device with computing capabilities, such as a personal computer (PC), a mobile terminal, a terminal device, a server, a controller in a vehicle, etc. The mobile terminal can be, for example, a vehicle-mounted device, a mobile phone, a tablet computer, a personal digital assistant, a wearable device, and other hardware devices with various operating systems, touch screens and / or display screens.

[0092] In addition, the network selection reconstruction device may also be software in an electronic device, etc. The software may be, for example, network selection reconstruction software, etc. In the following embodiments, the execution subject is a terminal device as an example for description.

[0093] like Figure 3 As shown, the method comprises the following steps:

[0094] Step 301, during the process of re-establishing network selection of the terminal device, a plurality of historical frequency points and network selection related information of the terminal device are obtained; the network selection related information includes a current serving cell identifier.

[0095] Step 302: determine whether to trigger a reconstruction network selection mechanism based on a network selection recommendation model, and determine whether the network selection recommendation model has a network selection recommendation capability.

[0096] In the embodiment of the present disclosure, whether to trigger the reconstruction network selection mechanism based on the network selection recommendation model can be configured in at least one of the following ways: by a network device; by a terminal device; or by a user using the terminal device.

[0097] In which, whether to trigger the reconstruction network selection mechanism based on the network selection recommendation model is configured by the user using the terminal device. The terminal device can display a configuration interface according to the user's operation; a configuration switch can be displayed in the configuration interface; the user can configure whether to trigger the reconstruction network selection mechanism based on the network selection recommendation model by operating the configuration switch.

[0098] In an embodiment of the present disclosure, the process of a terminal device determining whether a network selection recommendation model has the network selection recommendation capability may, for example, be to determine whether the network selection recommendation model has been trained, and to determine whether the accuracy of the network selection recommendation model meets a preset accuracy condition; if the network selection recommendation model has been trained, and the accuracy of the network selection recommendation model meets the preset accuracy condition, determining that the network selection recommendation model has the network selection recommendation capability; if the network selection recommendation model has not been trained, or the accuracy of the network selection recommendation model does not meet the preset accuracy condition, determining that the network selection recommendation model does not have the network selection recommendation capability.

[0099] Among them, when the network selection recommendation model has not been trained, or the accuracy of the network selection recommendation model does not meet the preset accuracy conditions, the accuracy of the network selection recommendation model in selecting frequencies and selecting target cells is difficult to guarantee. Therefore, cell measurement and cell reselection can be performed in combination with multiple historical frequencies, and reconstruction processing can be performed based on the cells obtained through the selection processing.

[0100] Step 303, when the reconstruction network selection mechanism based on the network selection recommendation model is not triggered, or when the network selection recommendation model does not have the network selection recommendation capability, cell measurement and cell reselection processing are performed in combination with multiple historical frequency points, and reconstruction processing is performed based on the cell obtained by the selection processing.

[0101] Wherein, when the reconstruction network selection mechanism based on the network selection recommendation model is not triggered, or when the network selection recommendation model does not have the network selection recommendation capability, in one example, the terminal device can sequentially perform cell measurement processing on each of the multiple frequency points in combination with the frequency point to obtain a measurement result. When the measurement result is a measured cell, reconstruction processing is performed based on the cell, and it is determined whether the reconstruction is successful; if the reconstruction is successful, the process ends. If the reconstruction fails, reconstruction processing is performed on another measured cell; if all cells at the frequency point fail during reconstruction processing, cell measurement processing and reconstruction processing are performed in combination with the next frequency point until the reconstruction is successful.

[0102] Wherein, when the reconstruction network selection mechanism based on the network selection recommendation model is not triggered, or when the network selection recommendation model does not have the network selection recommendation capability, in another example, the terminal device can sequentially perform cell measurement processing on each of the multiple frequency points in combination with the frequency point to obtain a measurement result. Based on the measurement results of each frequency point, multiple candidate cells are obtained. For each of the multiple candidate cells in turn, reconstruction processing is performed based on the candidate cell, and it is determined whether the reconstruction is successful; if the reconstruction is successful, the process ends; if the reconstruction fails, reconstruction processing is performed based on the next candidate cell until the reconstruction is successful.

[0103] Step 304, when the reconstruction network selection mechanism based on the network selection recommendation model is triggered and the network selection recommendation model has the network selection recommendation capability, multiple historical frequency points and network selection related information are input into the network selection recommendation model, and at least one candidate frequency point selected by the network selection recommendation model in combination with the network selection related information from the multiple historical frequency points is obtained.

[0104] Step 305: Perform cell measurement and cell reselection processing in combination with at least one candidate frequency point to obtain a target cell that has a neighboring cell relationship with the current serving cell corresponding to the current serving cell identifier.

[0105] Step 306: Perform reconstruction processing based on the target cell.

[0106] It should be noted that the detailed description of step 301, step 304 to step 306 can be found in Figure 1 Steps 101 to 104 in the illustrated embodiment will not be described in detail herein.

[0107] In the network selection reconstruction method of the disclosed embodiment, during the network selection reconstruction process of the terminal device, multiple historical frequency points and network selection related information of the terminal device are obtained; the network selection related information includes the current service cell identifier; it is determined whether to trigger the network selection reconstruction mechanism based on the network selection recommendation model, and whether the network selection recommendation model has the network selection recommendation capability; when the network selection reconstruction mechanism based on the network selection recommendation model is not triggered, or when the network selection recommendation model does not have the network selection recommendation capability, cell measurement and cell reselection processing are performed in combination with multiple historical frequency points, and reconstruction processing is performed based on the cell obtained by the selection processing; when the network selection reconstruction mechanism based on the network selection recommendation model is triggered, and the network selection recommendation model has the network selection recommendation capability, multiple historical frequency points are used to measure the cell and reselect the cell; Historical frequency points and network selection related information are input into a network selection recommendation model, and at least one candidate frequency point selected by the network selection recommendation model in combination with the network selection related information from multiple historical frequency points is obtained; cell measurement and cell reselection processing are performed in combination with the at least one candidate frequency point to obtain a target cell that has a neighboring relationship with the current serving cell corresponding to the current serving cell identifier; reconstruction processing is performed based on the target cell; wherein, based on whether a reconstruction network selection mechanism based on the network selection recommendation model is triggered and whether the network selection recommendation model has the network selection recommendation capability, it is determined whether to perform reconstruction network selection processing in combination with the network selection recommendation model, which can further increase the probability that the selected target cell has a neighboring relationship with the current serving cell, and further improve the reconstruction network selection efficiency of the terminal device.

[0108] The following examples are used to illustrate this. Figure 4 As shown in Figure 1, it is a schematic diagram of network reconstruction. Figure 4The following steps may be included. Step 401, start. Step 402, the terminal triggers reconstruction. Step 403, determine whether to turn on the AI ​​big model (network selection recommendation model) to reconstruct the network selection optimization; if so, execute step 404; if not, execute step 411. Step 404, the AI ​​big model determines whether it has the ability to optimize the scenario (determine whether the network selection recommendation model has the network selection recommendation capability); if so, execute step 405; if not, execute step 411. Step 405, apply for the AI ​​big model to recommend neighboring frequency points (select at least one candidate frequency point from multiple historical frequency points). Step 406, the AI ​​big model recommends neighboring frequency points (at least one candidate frequency point) based on calculation. Step 407, the terminal recommends a frequency measurement cell (combining at least one candidate frequency point to perform cell measurement to obtain multiple candidate cells). Step 408, the AI ​​big model selects the most suitable cell from the measured cells based on calculation (selects the target cell from multiple candidate cells). Step 409, initiate reconstruction to the network (reconstruction processing based on the target cell). Step 410: Feedback the result (reconstruction network selection data) to the AI ​​big model for data training. Step 411: Use the terminal's own reconstruction network selection strategy (combine multiple historical frequency points to perform cell measurement and cell reselection, and perform reconstruction based on the cell obtained by the selection process); then execute step 409. Step 412: End.

[0109] Figure 5 The flowchart of the network selection reconstruction method of another embodiment of the present disclosure is as follows. It should be noted that the network selection reconstruction method of the present embodiment can be applied to a network selection reconstruction device, which can be configured in an electronic device or chip so that the electronic device or chip can perform the network selection reconstruction function.

[0110] Among them, the electronic device can be any device with computing capabilities, such as a personal computer (PC), a mobile terminal, a terminal device, a server, a controller in a vehicle, etc. The mobile terminal can be, for example, a vehicle-mounted device, a mobile phone, a tablet computer, a personal digital assistant, a wearable device, and other hardware devices with various operating systems, touch screens and / or display screens.

[0111] In addition, the network selection reconstruction device may also be software in an electronic device, etc. The software may be, for example, network selection reconstruction software, etc. In the following embodiments, the execution subject is a cloud as an example for explanation.

[0112] like Figure 5 As shown, the method comprises the following steps:

[0113] Step 501: Acquire historical reconstruction network selection data of multiple terminal devices.

[0114] In an embodiment of the present disclosure, in one example, the cloud can be connected to a network device to obtain historical reconstruction network selection data of multiple terminal devices from at least one network device. In another example, the cloud can be connected to a terminal device to obtain historical reconstruction network selection data from multiple terminal devices.

[0115] In the disclosed embodiment, the historical reconstruction network selection data includes at least one of the following parameters in the historical reconstruction network selection process: multiple historical frequency points, network selection related information, at least one historical candidate frequency point selected in combination with the historical frequency points, multiple historical candidate cells obtained by performing cell measurements in combination with at least one historical candidate frequency point, and historical target cells obtained by selecting from multiple historical candidate cells; the reconstruction processing result of the historical reconstruction network selection process is successful reconstruction of the network selection.

[0116] Wherein, when the reconstruction network selection fails, at least one parameter in the historical reconstruction network selection process can be determined as historical reconstruction network selection data for training the network selection recommendation model. Wherein, the historical reconstruction network selection data when the reconstruction network selection fails can be used as a negative sample for training the network selection recommendation model.

[0117] Step 502: Obtain an initial network selection recommendation model.

[0118] The initial network selection recommendation model may be, for example, an artificial intelligence model such as a neural network model, a reinforcement learning model, a random forest model, a model based on a machine learning algorithm, etc. The above artificial intelligence model can perform feature extraction and decision processing on the input data.

[0119] Step 503: training the initial network selection recommendation model in combination with the historical reconstructed network selection data to obtain a trained network selection recommendation model.

[0120] In the disclosed embodiment, the training process of the network selection recommendation model in the cloud in combination with the historical reconstructed network selection data when the network selection is successfully reconstructed may include the following steps: inputting a plurality of historical frequency points and network selection related information into the initial network selection recommendation model to obtain at least one predicted frequency point output by the network selection recommendation model; inputting a plurality of historical candidate cells and network selection related information into the network selection recommendation model to obtain a predicted target cell output by the network selection recommendation model; determining the value of the first loss function in combination with at least one predicted frequency point, at least one historical candidate frequency point and a first loss function; determining the value of the second loss function in combination with the predicted target cell, the historical target cell and the second loss function; and adjusting the parameters of the network selection recommendation model according to the value of the first loss function and the value of the second loss function to achieve training.

[0121] In the training method for the reconstruction network selection model of the embodiment of the present disclosure, historical reconstruction network selection data of multiple terminal devices are obtained; an initial network selection recommendation model is obtained; the initial network selection recommendation model is trained and processed in combination with the historical reconstruction network selection data to obtain a trained network selection recommendation model; wherein the trained network selection recommendation model can learn how to select at least one candidate frequency point from multiple historical frequency points, and can learn how to select a target cell that has a neighboring relationship with the current service cell from multiple candidate cells, thereby improving the recommendation accuracy of the network selection recommendation model and further improving the probability of successful reconstruction.

[0122] Figure 6 The present invention is a schematic diagram of the structure of a network selection reconstruction device according to an embodiment of the present invention.

[0123] like Figure 6 As shown, the reconstruction network selection device may include: a first acquisition module 601, a second acquisition module 602, a measurement selection module 603 and a reconstruction processing module 604.

[0124] Among them, the first acquisition module 601 is used to obtain multiple historical frequency points and network selection related information of the terminal device during the reconstruction network selection process of the terminal device; the network selection related information includes the current service cell identifier; the second acquisition module 602 is used to input the multiple historical frequency points and the network selection related information into the network selection recommendation model, and obtain at least one candidate frequency point selected by the network selection recommendation model in combination with the network selection related information from the multiple historical frequency points; the measurement selection module 603 is used to perform cell measurement and cell reselection processing in combination with at least one of the candidate frequencies, and obtain a target cell that has a neighboring relationship with the current service cell corresponding to the current service cell identifier; the reconstruction processing module 604 is used to perform reconstruction processing based on the target cell.

[0125] In one embodiment of the present disclosure, the measurement selection module 603 is specifically used to perform cell measurement processing in combination with at least one of the candidate frequency points to obtain multiple candidate cells; input the multiple candidate cells and the network selection related information into the network selection recommendation model, and obtain the target cell selected by the network selection recommendation model in combination with the network selection related information from the multiple candidate cells.

[0126] In one embodiment of the present disclosure, the device also includes: a determination module, used to determine whether to trigger a reconstruction network selection mechanism based on a network selection recommendation model, and determine whether the network selection recommendation model has a network selection recommendation capability; the measurement selection module 603 is also used to perform cell measurement and cell reselection processing in combination with the multiple historical frequency points when the reconstruction network selection mechanism based on the network selection recommendation model is not triggered, or when the network selection recommendation model does not have a network selection recommendation capability, and perform reconstruction processing based on the cell obtained by the selection processing.

[0127] In one embodiment of the present disclosure, the determination module is specifically used to determine whether the network selection recommendation model has been trained and whether the accuracy of the network selection recommendation model meets the preset accuracy condition; when the network selection recommendation model has been trained and the accuracy of the network selection recommendation model meets the preset accuracy condition, determine that the network selection recommendation model has the network selection recommendation capability; when the network selection recommendation model has not been trained or the accuracy of the network selection recommendation model does not meet the preset accuracy condition, determine that the network selection recommendation model does not have the network selection recommendation capability.

[0128] In one embodiment of the present disclosure, the device also includes: a third acquisition module, a sending and receiving module and a parameter updating module; the third acquisition module is used to obtain the historical reconstructed network selection data of the terminal device; the sending and receiving module is used to send the historical reconstructed network selection data to the cloud, and receive the network selection recommendation model parameters returned by the cloud; the network selection recommendation model parameters are the parameters of the network selection recommendation model obtained after the cloud combines the historical reconstructed network selection data of multiple terminal devices to train the initial network selection recommendation model; the parameter updating module is used to perform parameter update processing on the network selection recommendation model of the terminal device according to the network selection recommendation model parameters.

[0129] In one embodiment of the present disclosure, the reconstruction processing module 604 is specifically used to obtain the terminal context of the terminal device in the network device corresponding to the target cell; the terminal context is obtained by the interaction between the network device corresponding to the target cell and the network device corresponding to the current service cell; and reconstruction processing is performed based on the target cell and the terminal context.

[0130] In one embodiment of the present disclosure, the network selection related information also includes at least one of the following: the public land mobile network PLMN identifier of the current service cell, the frequency of the current service cell, cell resource information, the capabilities of the terminal device, the location of the terminal device, and the moving speed of the terminal device.

[0131] In the network reconstruction device of the disclosed embodiment, during the network reconstruction process of the terminal device, multiple historical frequency points and network selection related information are obtained; the network selection related information includes the current service cell identifier; the multiple historical frequency points and network selection related information are input into the network selection recommendation model, and at least one candidate frequency point selected from the multiple historical frequency points by the network selection recommendation model in combination with the network selection related information is obtained; cell measurement and cell reselection processing are performed in combination with the at least one candidate frequency point to obtain a target cell that has a neighboring relationship with the current service cell corresponding to the current service cell identifier; reconstruction processing is performed based on the target cell; wherein the selected candidate frequency point is a frequency point that can be measured to obtain the cell, thereby reducing the number of frequency points that need to be processed, shortening the reconstruction network selection delay, and thus improving the reconstruction network selection efficiency of the terminal device.

[0132] Figure 7 A schematic diagram of the structure of a training device for reconstructing a network selection model according to an embodiment of the present disclosure.

[0133] like Figure 7 As shown, the training device for reconstructing the network selection model may include: a first acquisition module 701, a second acquisition module 702 and a training processing module 703.

[0134] Among them, the first acquisition module 701 is used to obtain the historical reconstructed network selection data of multiple terminal devices; the second acquisition module 702 is used to obtain the initial network selection recommendation model; the training processing module 703 is used to train the initial network selection recommendation model in combination with the historical reconstructed network selection data to obtain a trained network selection recommendation model.

[0135] In one embodiment of the present disclosure, the historical reconstruction network selection data includes at least one of the following parameters in the historical reconstruction network selection process: multiple historical frequency points, network selection related information, at least one historical candidate frequency point selected in combination with the historical frequency points, multiple historical candidate cells obtained by performing cell measurements in combination with at least one of the historical candidate frequencies, and historical target cells selected in combination with the multiple historical candidate cells; the reconstruction processing result of the historical reconstruction network selection process is successful reconstruction of the network selection.

[0136] In one embodiment of the present disclosure, the training processing module is specifically used to input the multiple historical frequency points and the network selection related information into the initial network selection recommendation model to obtain at least one predicted frequency point output by the network selection recommendation model; input the multiple historical candidate cells and the network selection related information into the network selection recommendation model to obtain the predicted target cell output by the network selection recommendation model; determine the value of the first loss function in combination with at least one of the predicted frequency points, at least one of the historical candidate frequency points and the first loss function; determine the value of the second loss function in combination with the predicted target cell, the historical target cell and the second loss function; and perform parameter adjustment processing on the network selection recommendation model according to the value of the first loss function and the value of the second loss function to achieve training.

[0137] In the training device for reconstructing the network selection model of the embodiment of the present disclosure, historical reconstruction network selection data of multiple terminal devices are obtained; an initial network selection recommendation model is obtained; the initial network selection recommendation model is trained and processed in combination with the historical reconstruction network selection data to obtain a trained network selection recommendation model; wherein the trained network selection recommendation model can learn how to select at least one candidate frequency point from multiple historical frequency points, and can learn how to select a target cell that has a neighboring relationship with the current service cell from multiple candidate cells, thereby improving the recommendation accuracy of the network selection recommendation model and further improving the probability of successful reconstruction.

[0138] According to the fifth aspect of an embodiment of the present disclosure, an electronic device is also provided, comprising: a processor; a memory for storing processor executable instructions, wherein the processor is configured to: implement the steps of the network reconstruction method as described above; or, implement the steps of the network reconstruction model training method as described above.

[0139] In order to implement the above embodiments, the present disclosure also proposes a non-transitory computer-readable storage medium.

[0140] When the instructions in the storage medium are executed by the processor, the processor is enabled to execute the network reconstruction and selection method as described above; or, execute the network reconstruction and selection model training method as described above.

[0141] In order to implement the above embodiments, the present disclosure also provides a computer program product.

[0142] When the computer program product is executed by a processor of an electronic device, the electronic device is enabled to execute the above method.

[0143] Figure 8 The figure is a structural block diagram of an electronic device according to an exemplary embodiment. Figure 8The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0144] like Figure 8 As shown, the electronic device 1000 includes a processor 111, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 112 or a program loaded from a memory 116 to a random access memory (RAM) 113. In the RAM 113, various programs and data required for the operation of the electronic device 1000 are also stored. The processor 111, the ROM 112, and the RAM 113 are connected to each other via a bus 114. An input / output (I / O) interface 115 is also connected to the bus 114.

[0145] The following components are connected to the I / O interface 115: a memory 116 including a hard disk, etc.; and a communication part 117 including a network interface card such as a local area network (LAN) card, a modem, etc., and the communication part 117 performs communication processing via a network such as the Internet; a drive 118 is also connected to the I / O interface 115 as needed.

[0146] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 117. When the computer program is executed by the processor 111, the above-mentioned functions defined in the method of the present disclosure are executed.

[0147] In an exemplary embodiment, a storage medium including instructions is also provided, such as a memory including instructions, and the instructions can be executed by the processor 111 of the electronic device 1000 to complete the above method. Optionally, the storage medium can be a non-transitory computer-readable storage medium, for example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0148] In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0149] Fig. 9 FIG. 1 is a schematic diagram of the structure of a chip according to an embodiment of the present disclosure. Fig. 9 As shown, the chip includes a processor 901 and an interface circuit 902. The number of the processor 901 can be one or more, and the number of the interface circuit 902 can be one or more.

[0150] Optionally, the interface circuit 902 is used to receive a signal, and the signal includes a computer instruction. When the processor 901 executes the computer instruction, the chip executes the reconstruction network selection method or the training method of the reconstruction network selection model described in the above embodiments of the present disclosure.

[0151] The collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in this application are in compliance with relevant laws and regulations and do not violate public order and good morals.

[0152] In addition, the word "exemplary" is used herein to indicate serving as an example, instance, or diagram. Any aspect or design described as "exemplary" in this article is not necessarily understood to be advantageous compared to other aspects or designs. On the contrary, the use of the word exemplary is intended to present concepts in a specific way. As used herein, the term "or" is intended to represent an inclusive "or" rather than an exclusive "or". That is, unless otherwise specified or clear from the context, "X applies A or B" is intended to represent any one of the natural inclusive arrangements. That is, if X applies A; X applies B; or X applies both A and B, "X applies A or B" is satisfied under any of the aforementioned examples. In addition, unless otherwise specified or clearly pointed to a singular form from the context, the articles "one" and "an" as used in this application and the appended claims are generally understood to mean "one or more".

[0153] Likewise, although the present disclosure has been shown and described with respect to one or more implementations, equivalent variations and modifications will occur to those skilled in the art after reading and understanding the specification and drawings. The present disclosure includes all such modifications and variations and is limited only by the scope of the claims. In particular, with respect to the various functions performed by the components (e.g., elements, resources, etc.) described above, unless otherwise noted, the terms used to describe such components are intended to correspond to any component (functionally equivalent) that performs the specific functions of the described components, even if the structure is not equivalent to the disclosed structure. In addition, although the specific features of the present disclosure may have been disclosed with respect to only one of several implementations, such features may be combined with one or more other features of other implementations as may be desired and conducive to any given or specific application. In addition, with respect to "including", "having", "having", "having", or variations thereof used in the specific embodiments or claims, such terms are intended to be inclusive in a manner similar to the term "comprising".

[0154] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0155] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A network selection reconstruction method, characterized in that: The method comprises: In the process of re-establishing network selection of the terminal device, a plurality of historical frequency points and network selection related information of the terminal device are obtained; the network selection related information includes a current serving cell identifier; Inputting the multiple historical frequency points and the network selection related information into a network selection recommendation model, and obtaining at least one candidate frequency point selected by the network selection recommendation model in combination with the network selection related information from the multiple historical frequency points; Performing cell measurement and cell reselection processing in combination with at least one of the candidate frequency points to obtain a target cell that has a neighboring cell relationship with the current serving cell corresponding to the current serving cell identifier; Reconstruction processing is performed based on the target cell.

2. The method according to claim 1, characterized in that The performing cell measurement and cell reselection processing in combination with at least one of the candidate frequency points to obtain a target cell having a neighboring cell relationship with the current serving cell corresponding to the current serving cell identifier includes: Performing cell measurement processing in combination with at least one of the candidate frequency points to obtain multiple candidate cells; The multiple candidate cells and the network selection related information are input into the network selection recommendation model, and the target cell selected from the multiple candidate cells by the network selection recommendation model in combination with the network selection related information is obtained.

3. The method according to claim 1 or 2, characterized in that: Before inputting the multiple historical frequency points and the network selection related information into a network selection recommendation model and obtaining at least one candidate frequency point selected by the network selection recommendation model in combination with the network selection related information from the multiple historical frequency points, the method further includes: Determining whether to trigger a reconstruction network selection mechanism based on a network selection recommendation model, and determining whether the network selection recommendation model has a network selection recommendation capability; When the reconstruction network selection mechanism based on the network selection recommendation model is not triggered, or when the network selection recommendation model does not have the network selection recommendation capability, cell measurement and cell reselection processing are performed in combination with the multiple historical frequency points, and reconstruction processing is performed based on the cell obtained by the selection processing.

4. The method according to claim 3, characterized in that The determining whether the network selection recommendation model has a network selection recommendation capability includes: Determining whether the network selection recommendation model has been trained, and determining whether the accuracy of the network selection recommendation model meets a preset accuracy condition; When the network selection recommendation model is trained and the accuracy of the network selection recommendation model meets the preset accuracy condition, determining that the network selection recommendation model has the network selection recommendation capability; When the network selection recommendation model has not been trained, or the accuracy of the network selection recommendation model does not meet the preset accuracy condition, it is determined that the network selection recommendation model does not have the network selection recommendation capability.

5. The method according to claim 1 or 2, characterized in that: The method further comprises: Acquire historical reconstruction network selection data of the terminal device; Sending the historical reconstructed network selection data to the cloud, and receiving the network selection recommendation model parameters returned by the cloud; the network selection recommendation model parameters are parameters of the network selection recommendation model obtained by training the initial network selection recommendation model by the cloud in combination with the historical reconstructed network selection data of multiple terminal devices; According to the network selection recommendation model parameters, parameter updating processing is performed on the network selection recommendation model of the terminal device.

6. The method according to claim 1, characterized in that The performing reconstruction processing based on the target cell includes: Acquire a terminal context of the terminal device in the network device corresponding to the target cell; the terminal context is obtained by interaction between the network device corresponding to the target cell and the network device corresponding to the current serving cell; Reconstruction processing is performed based on the target cell and the terminal context.

7. The method according to claim 1, characterized in that The network selection related information also includes at least one of the following: the public land mobile network PLMN identifier of the current serving cell, the frequency of the current serving cell, cell resource information, terminal device capabilities, terminal device location, and terminal device movement speed.

8. A method for training a network selection recommendation model, characterized in that: The method comprises: Obtain historical reconstruction network selection data for multiple terminal devices; Obtain an initial network selection recommendation model; The initial network selection recommendation model is trained in combination with the historical reconstructed network selection data to obtain a trained network selection recommendation model.

9. The method according to claim 8, characterized in that The historical reconstruction network selection data includes at least one of the following parameters in the historical reconstruction network selection process: multiple historical frequency points, network selection related information, at least one historical candidate frequency point selected in combination with the historical frequency points, multiple historical candidate cells obtained by cell measurement based on at least one of the historical candidate frequencies, and historical target cells selected in combination with the multiple historical candidate cells; the reconstruction processing result of the historical reconstruction network selection process is successful reconstruction of the network selection.

10. The method according to claim 9, characterized in that The training process of the initial network selection recommendation model in combination with the historical reconstructed network selection data to obtain a trained network selection recommendation model includes: Inputting the multiple historical frequency points and the network selection related information into the initial network selection recommendation model to obtain at least one predicted frequency point output by the network selection recommendation model; Inputting the multiple historical candidate cells and the network selection related information into the network selection recommendation model to obtain the predicted target cell output by the network selection recommendation model; Determine a value of the first loss function by combining at least one of the predicted frequency points, at least one of the historical candidate frequency points, and the first loss function; Determine a value of the second loss function by combining the predicted target cell, the historical target cell and the second loss function; According to the value of the first loss function and the value of the second loss function, parameter adjustment processing is performed on the network selection recommendation model to achieve training.

11. A network selection reconstruction device, characterized in that: The device comprises: A first acquisition module is used to acquire multiple historical frequency points and network selection related information of the terminal device during the network selection reconstruction process of the terminal device; the network selection related information includes a current serving cell identifier; A second acquisition module is used to input the multiple historical frequency points and the network selection related information into a network selection recommendation model, and obtain at least one candidate frequency point selected by the network selection recommendation model in combination with the network selection related information from the multiple historical frequency points; A measurement selection module, configured to perform cell measurement and cell reselection processing in combination with at least one of the candidate frequency points, and obtain a target cell that has a neighboring cell relationship with the current serving cell corresponding to the current serving cell identifier; A reconstruction processing module is used to perform reconstruction processing based on the target cell.

12. A training device for a network selection recommendation model, characterized in that: The device comprises: A first acquisition module is used to acquire historical reconstruction network selection data of multiple terminal devices; The second acquisition module is used to acquire an initial network selection recommendation model; The training processing module is used to train the initial network selection recommendation model in combination with the historical reconstructed network selection data to obtain a trained network selection recommendation model.

13. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; Wherein, the processor is configured to: Implement the steps of the network reconstruction method as described in any one of claims 1 to 7; or implement the steps of the network reconstruction model training method as described in any one of claims 8 to 10.

14. A non-temporary computer-readable storage medium, when the instructions in the storage medium are executed by a processor, enables the processor to execute the reconstruction network selection method as described in any one of claims 1 to 7; or, execute the training method of the reconstruction network selection model as described in any one of claims 8 to 10.

15. A chip, characterized in that: The invention comprises one or more interface circuits and one or more processors; the interface circuit is used to receive a signal, the signal comprises a computer instruction, and when the processor executes the computer instruction, the chip executes the reconstruction network selection method described in any one of claims 1 to 7; or executes the training method of the reconstruction network selection model described in any one of claims 8 to 10.

Images