Determination method and system and electronic equipment

Abstract

The invention discloses a determination method and system and electronic equipment, and the method comprises the steps: determining a first signal parameter of each cell based on the frequency point information of a cell corresponding to a current connection network and at least one adjacent cell, and each cell at least comprises the cell corresponding to the current connection network and the cell corresponding to the at least one adjacent cell; the first signal parameter represents the signal quality of each cell; determining a recommended value of each neighbor cell in the at least one neighbor cell based on the first signal parameter and interference information, wherein the interference information is used for representing an interference condition between the at least one neighbor cell and a cell corresponding to the current connection network; adjusting the first signal parameter of the at least one adjacent cell based on the recommended value, and determining a second signal parameter; and determining a recommended neighbor cell based on the second signal parameter of each neighbor cell in the at least one neighbor cell.

Description

Technical Field

[0001] The present application relates to the field of communications, and in particular to a determination method, system and electronic device. Background Art

[0002] During the communication process, when selecting a cell, the cell is usually measured directly, and whether to select the cell is determined based on the measurement results. For example, the selected cell is determined by the average value of the signal power of the cell received by the terminal. However, this method does not take into account the overall network environment in which the terminal is located, which may result in the inability to select the cell that best suits the current network environment. Summary of the invention

[0003] In view of this, the present application provides a determination method, system and electronic device, and the specific scheme thereof is as follows:

[0004] A determination method comprising:

[0005] Determine a first signal parameter of each cell based on frequency information of a cell corresponding to the currently connected network and at least one neighboring cell, wherein each cell at least includes a cell corresponding to the currently connected network and a cell corresponding to the at least one neighboring cell, and the first signal parameter represents a signal quality of each cell;

[0006] Determine a recommended value for each of the at least one neighboring cell based on the first signal parameter and interference information, wherein the interference information is used to characterize interference between the at least one neighboring cell and a cell corresponding to the currently connected network;

[0007] Adjusting the first signal parameter of the at least one neighboring cell based on the recommended value to determine the second signal parameter of each neighboring cell;

[0008] Based on the second signal parameter, a recommended neighboring cell is determined.

[0009] Further, the adjusting the first signal parameter of the at least one neighboring cell based on the recommended value to determine the second signal parameter of each neighboring cell includes:

[0010] Sorting the at least one neighboring area according to the size of the recommendation value to obtain a sorting result;

[0011] Determine a suppression factor for each neighboring area based on the ranking result;

[0012] Based on the suppression factor of each neighboring cell, adjusting the first signal parameter of the neighboring cell corresponding to the suppression factor to obtain the second signal parameter of each neighboring cell;

[0013] The adjusting the first signal parameter of the neighboring area corresponding to the suppression factor based on the suppression factor of each neighboring area to obtain the second signal parameter of each neighboring area includes:

[0014] For each neighboring cell, the first signal parameter of the neighboring cell is adjusted based on the suppression factor of the neighboring cell to obtain the second signal parameter of the neighboring cell.

[0015] Furthermore, determining the suppression factor of each neighboring area based on the sorting result includes:

[0016] If in the sorting result, the recommendation value of the first neighboring area is ranked higher than the recommendation value of the second neighboring area, it is determined that the suppression factor corresponding to the first neighboring area is greater than the suppression factor corresponding to the second neighboring area.

[0017] Further, the determining a recommended neighboring area based on the second signal parameter includes:

[0018] Sorting the at least one neighboring area according to the magnitude of the second signal parameter of each neighboring area in the at least one neighboring area to obtain a signal parameter sorting result;

[0019] A specific number of neighboring areas whose second signal parameters meet a condition is determined based on the signal parameter sorting result, and the specific number of neighboring areas are determined as recommended neighboring areas.

[0020] Further, the determining the recommended value of each neighboring cell in the at least one neighboring cell based on the first signal parameter and the interference information includes:

[0021] Inputting a first signal parameter of a cell in the currently connected network and a first signal parameter of a third neighboring area among the at least one neighboring area into a pre-trained deep learning model, and obtaining a recommended value of the third neighboring area output by the deep learning model;

[0022] Among them, the deep learning model is obtained by model training using training data and a simulation environment, and the training data includes at least: the first signal training parameters of the cell corresponding to the connected network and the first signal training parameters of at least one neighboring cell, and the at least one neighboring cell is a cell corresponding to at least one network to be connected.

[0023] Furthermore, the deep learning model is obtained by training the model using training data and a simulation environment, including:

[0024] Based on the simulation environment, in a first state, inputting the training data into a neural network model to obtain a probability of the neural network model outputting a transition from the first state to a second state;

[0025] determining a reward value based on the probability of transitioning from the first state to the second state;

[0026] The deep learning model is trained based on the first state, the second state and the reward value to obtain a trained deep learning model, and the recommendation value of the deep learning model is used to represent the accumulated reward value.

[0027] Further, the determining of the reward value based on the probability of transitioning from the first state to the second state includes:

[0028] If the first state represents a configuration state for adding a target neighboring area, the second state represents a failure in adding the target neighboring area, and a probability of the first state transitioning to the second state is greater than a first target value, determining the reward value to be the first reward value;

[0029] If the first state represents a configuration state of adding a target neighboring area, the second state represents that the target neighboring area is successfully added, and the probability of the first state transitioning to the second state is greater than a second target value, the reward value is determined to be a second reward value, and the second reward value is greater than the first reward value;

[0030] If the first state represents a successful addition of a target neighboring cell, the second state represents a range of a bit error rate when transmitting data services through the cell corresponding to the connected network and the target neighboring cell, and the reward value is determined to be a third reward value based on the range of the bit error rate in the second state, and the third reward value is not less than the first reward value.

[0031] Furthermore, the step of sorting the at least one neighboring area according to the size of the recommended value to obtain a sorting result includes:

[0032] Comparing the recommended value of performing an add operation on each neighboring area in the at least one neighboring area with the recommended value of performing no add operation, wherein the add operation is used to add the neighboring area to the currently connected network;

[0033] Select the neighboring area whose recommendation value for performing the adding operation is greater than the recommendation value for performing the not-adding operation, sort the recommendation values ​​for performing the adding operation on the selected neighboring areas, and obtain a sorting result.

[0034] A determination system, comprising:

[0035] A first determining unit, configured to determine a first signal parameter of each cell based on frequency information of a cell corresponding to a currently connected network and at least one neighboring cell, wherein the cells at least include a cell corresponding to the currently connected network and a cell corresponding to the at least one neighboring cell, and the first signal parameter represents a signal quality of each cell;

[0036] A second determining unit, configured to determine a recommended value for each of the at least one neighboring cell based on the first signal parameter and interference information, wherein the interference information is used to characterize interference between the at least one neighboring cell and a cell corresponding to the currently connected network;

[0037] an adjusting unit, configured to adjust the first signal parameter of the at least one neighboring cell based on the recommended value, and determine a second signal parameter of each neighboring cell;

[0038] The third determining unit is used to determine a recommended neighboring area based on the second signal parameter.

[0039] An electronic device, comprising:

[0040] A processor, configured to determine a first signal parameter of each cell based on frequency information of a cell corresponding to a currently connected network and at least one neighboring cell, wherein each cell includes at least a cell corresponding to the currently connected network and a cell corresponding to the at least one neighboring cell, and the first signal parameter represents a signal quality of each cell; determine a recommended value for each neighboring cell in the at least one neighboring cell based on the first signal parameter and interference information, wherein the interference information is used to represent the interference between at least one neighboring cell and the cell corresponding to the currently connected network; adjust the first signal parameter of the at least one neighboring cell based on the recommended value to determine a second signal parameter for each neighboring cell; and determine a recommended neighboring cell based on the second signal parameter;

[0041] The memory is used to store the program required by the processor to execute the above processing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 A flowchart of a determination method disclosed in an embodiment of the present application;

[0044] Figure 2 A flowchart of a determination method disclosed in an embodiment of the present application;

[0045] Figure 3 A flowchart of a determination method disclosed in an embodiment of the present application;

[0046] Figure 4 A flowchart of a determination method disclosed in an embodiment of the present application;

[0047] Figure 5 A schematic diagram of a corresponding relationship between a state and a reward value disclosed in an embodiment of the present application;

[0048] Figure 6 A schematic diagram of model training for a deep learning model disclosed in an embodiment of the present application;

[0049] Figure 7 A schematic diagram of the structure of a determination system disclosed in an embodiment of the present application;

[0050] Figure 8 A schematic diagram of the structure of an electronic device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0051] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0052] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0053] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0054] The present application discloses a determination method, and its flow chart is as follows: Figure 1 As shown, including:

[0055] Step S11: determining a first signal parameter of each cell based on frequency information of a cell corresponding to the currently connected network and at least one neighboring cell, wherein each cell at least includes a cell corresponding to the currently connected network and at least one neighboring cell, and the first signal parameter represents a signal quality of each cell;

[0056] Step S12: determining a recommended value for each neighboring cell in at least one neighboring cell based on the first signal parameter and interference information, where the interference information is used to characterize interference between the at least one neighboring cell and a cell corresponding to the currently connected network;

[0057] Step S13: adjusting the first signal parameter of at least one neighboring cell based on the recommended value to determine the second signal parameter of each neighboring cell;

[0058] Step S14: Determine a recommended neighboring area based on the second signal parameter.

[0059] During the communication process, when selecting a cell, the signal quality of the cell is usually measured directly, and whether to select the cell is determined based on the measurement results. For example, the selected cell is determined by the average value of the signal power of the cell received by the terminal. However, this method does not take into account the overall network environment in which the terminal is located, which may result in the inability to select the cell that best suits the current network environment.

[0060] Based on this, in this solution, when making neighboring cell recommendations, it is necessary not only to determine the recommended value of the cell based on the first signal parameter used to characterize the signal quality of the cell, but also to determine the recommended value of the cell based on the interference value between the neighboring cell and the cell corresponding to the currently connected network, and the first signal parameter, and further determine the recommended neighboring cell based on the recommended value, so as to ensure that the finally determined recommended neighboring cell is not only determined based on the signal quality of the neighboring cell, but also based on the interference situation between the neighboring cell and the cell corresponding to the currently connected network, thereby ensuring that the recommended cell can meet the current network environment and avoiding the problem of reduced communication quality due to strong interference between the recommended neighboring cell and the cell corresponding to the currently connected network when the recommended neighboring cell is selected for connection.

[0061] When a user's handheld electronic device is at the edge of a cell or in an area with weak signals, the signal of the primary cell PCell cannot meet the communication needs. At this time, a secondary cell SCell needs to be added to enhance signal coverage to ensure that the electronic device can communicate stably. Alternatively, for some services that require high communication continuity, such as voice calls and video calls, secondary cells can be added to ensure communication continuity.

[0062] In wireless communication, the role of the secondary cell SCell is mainly to provide additional wireless resources for the primary cell PCell to improve network capacity and data transmission efficiency. Therefore, when it is necessary to add a secondary cell SCell, the determination method disclosed in this embodiment can be used to recommend the secondary cell.

[0063] When it is necessary to recommend a secondary cell SCell, the cell corresponding to the currently connected network can be determined first, that is, the cells corresponding to the networks to which the electronic device is currently connected, wherein the cells corresponding to the networks to which the electronic device is currently connected can include not only: the cell range covered by the 4g base station to which the electronic device is currently connected, and the cell range covered by the 5g base station to which the electronic device is currently connected, but also include: the WIFI coverage range to which the electronic device is currently connected. Therefore, it is necessary to determine that the cell corresponding to the network to which the electronic device is currently connected includes: the cell covered by the 4g base station, the cell covered by the 5g base station, and the WIFI coverage range (of course, if the electronic device is not currently connected to WIFI, then the cell corresponding to the network to which the electronic device is currently connected does not include the WIFI coverage range).

[0064] After determining the cell corresponding to the currently connected network, it is necessary to further determine the first signal parameters of the cell corresponding to the currently connected network. The measurement can be specifically based on the frequency information of the cell corresponding to the currently connected network to obtain the first signal parameters of each cell. The first signal parameters represent the signal quality of the corresponding cell, that is, the signal quality of each cell corresponding to the currently connected network is obtained. For example, the network currently connected to the electronic device may include a first cell and WIFI, then the first signal parameters of the first cell and the first signal parameters of WIFI need to be obtained.

[0065] In addition, it is also necessary to determine the neighboring cells of the cell corresponding to the network to which the electronic device is currently connected, so as to recommend at least some cells from the neighboring cells as secondary cells. The neighboring cells may be determined by: the network side provides the electronic device with information about the neighboring cells through OTA (Over The Air) signaling, so that the electronic device can determine the neighboring cells. Specifically, the frequency information of the neighboring cells may be transmitted to the electronic device through OTA signaling by the network side, so as to measure the signal quality of the neighboring cells based on the frequency information of the neighboring cells, so as to obtain the first signal parameters of the neighboring cells, so that the electronic device can directly know the first signal parameters of the neighboring cells.

[0066] Among them, the neighboring areas may include: the cell range covered by the 4g base station, the cell range covered by the 5g base station, and may also include: the WIFI coverage range, all of which are networks to be connected.

[0067] After obtaining the first signal parameter of the cell corresponding to the currently connected network and the first signal parameter of at least one neighboring cell, the recommended value of each neighboring cell is determined using the first signal parameter and the interference information.

[0068] The interference information is used to characterize the interference situation between at least one neighboring cell and the cell corresponding to the currently connected network. If the frequency bands between different cells are close, there will be a problem of co-channel interference. Therefore, in this embodiment, when determining the recommended neighboring cell, not only the signal quality of the neighboring cell is considered, but also whether there is co-channel interference between the neighboring cell and the cell corresponding to the currently connected network. For example, the cell corresponding to the currently connected network includes a first cell, wherein the frequency band of the first neighboring cell is close to the frequency band of the first cell, and there is co-channel interference. In this case, when recommending neighboring cells, the first neighboring cell is not recommended, but a neighboring cell with no interference with the cell corresponding to the currently connected network is recommended.

[0069] If there is interference between the cell corresponding to the currently connected network and the neighboring cell, an interference value needs to be determined and used as interference information of the neighboring cell, so that the recommended value of the neighboring cell can be determined by using the interference information and the first signal parameter.

[0070] Among them, the first signal parameter used to determine the recommended value of the neighboring cell includes not only the first signal parameter of the neighboring cell, but also the first signal parameter of the cell corresponding to the currently connected network. The recommended value of the neighboring cell is determined jointly by utilizing the first parameter information of the cell corresponding to the currently connected network, the first parameter information of the neighboring cell, and the interference information between the neighboring cell and the cell corresponding to the currently connected network, so that the recommended value of the neighboring cell can meet both the signal quality of the neighboring cell and the interference information between it and the cell corresponding to the currently connected network.

[0071] The recommended value of the neighboring cell is used as a parameter to adjust the first signal parameter of the neighboring cell, so that the second signal parameter obtained after the adjustment can characterize both the signal quality of the neighboring cell and the interference between the neighboring cell and the cell corresponding to the currently connected network, so that when the second signal parameter is used to determine the recommended neighboring cell, it can be more accurate and more in line with the current network environment of the electronic device, avoiding the problem of poor signal quality in the recommended neighboring cell or interference with other currently connected networks.

[0072] Different neighboring cells may correspond to different recommended values. When adjusting the first signal parameters, the corresponding recommended values ​​need to be selected. That is, when adjusting the first signal parameters of a certain neighboring cell, the recommended values ​​corresponding to the neighboring cell need to be selected, and the recommended values ​​corresponding to other neighboring cells cannot be selected.

[0073] The determination method disclosed in this embodiment determines the first signal parameter of each cell based on the frequency information of the cell corresponding to the currently connected network and at least one neighboring cell, each cell at least includes the cell corresponding to the currently connected network and the cell corresponding to at least one neighboring cell, and the first signal parameter characterizes the signal quality of each cell; determines the recommended value of each neighboring cell in at least one neighboring cell based on the first signal parameter and interference information, and the interference information is used to characterize the interference situation between at least one neighboring cell and the cell corresponding to the currently connected network; adjusts the first signal parameter of at least one neighboring cell based on the recommended value, and determines the second signal parameter of each neighboring cell; determines the recommended neighboring cell based on the second signal parameter. When determining the recommended cell, this scheme not only needs to determine the first signal parameters of each neighboring cell, but also needs to determine the first signal parameters of the cell corresponding to the currently connected network, and determine the recommended value of each neighboring cell based on the first signal parameters and the interference information between the neighboring cell and the cell corresponding to the currently connected network, and determine the recommended neighboring cell based on this, ensuring that when recommending the neighboring cell, it can be determined not only based on the signal quality of each neighboring cell, but also based on the interference information between each neighboring cell and the cell corresponding to the currently connected network, so as to ensure that the finally recommended neighboring cell is a cell that can meet the current connection network, and avoid the situation where the communication quality deteriorates after adding neighboring cells due to interference between different cells.

[0074] This embodiment discloses a determination method, and its flow chart is as follows: Figure 2 As shown, including:

[0075] Step S21: determining a first signal parameter of each cell based on frequency information of a cell corresponding to the currently connected network and at least one neighboring cell, wherein each cell at least includes a cell corresponding to the currently connected network and at least one neighboring cell, and the first signal parameter represents a signal quality of the corresponding cell;

[0076] Step S22: determining a recommended value for each neighboring cell in at least one neighboring cell based on the first signal parameter and interference information, where the interference information is used to characterize interference between the at least one neighboring cell and a cell corresponding to the currently connected network;

[0077] Step S23: sort at least one neighboring area according to the size of the recommended value to obtain a sorting result;

[0078] Step S24: determining the suppression factor of each neighboring area based on the sorting result;

[0079] Step S25: adjusting the first signal parameter of the neighboring area corresponding to the suppression factor based on the suppression factor of each neighboring area to obtain the second signal parameter of each neighboring area;

[0080] Step S26: Determine a recommended neighboring area based on the second signal parameter.

[0081] When determining the recommended neighboring cell, it is necessary to first determine the frequency information of the cell corresponding to the currently connected network, and determine the first signal parameter of the corresponding cell based on the frequency information, as well as determine the frequency information of at least one neighboring cell, and determine the first signal parameter of the corresponding neighboring cell based on the frequency information of the neighboring cell. Afterwards, based on the first signal parameter of the cell corresponding to the currently connected network and the first signal parameter of at least one neighboring cell, and the interference information, determine the recommended value of each neighboring cell, so as to adjust the first signal parameter of the corresponding neighboring cell based on the recommended value of the neighboring cell to obtain the second signal parameter of each neighboring cell, thereby determining the recommended neighboring cell based on the second signal parameter of each neighboring cell.

[0082] Specifically, when the first signal parameter of the neighboring cell is adjusted based on the recommended value of each neighboring cell to obtain the second signal parameter of the corresponding neighboring cell, it can be:

[0083] All neighboring cells determined by the electronic device are sorted according to the recommended value of each neighboring cell to obtain a sorting result, such as: determining that there are 3 neighboring cells that can currently serve as the secondary cell of the electronic device, including: neighboring cell 1, neighboring cell 2, and neighboring cell 3, as shown in Table 1:

[0084] Table 1

[0085] Neighborhood 1 Neighborhood 2 Neighborhood 3 Recommended value 5 9 7 Recommended value sorting 3 1 2

[0086] Among them, the recommended value of neighbor area 1 is 5, the recommended value of neighbor area 2 is 9, and the recommended value of neighbor area 3 is 7. The three neighbor areas are sorted according to the recommended values: neighbor area 2, neighbor area 3, and neighbor area 1.

[0087] After determining the ranking results of the recommended values ​​of each neighboring area, the suppression factor of each neighboring area is determined according to the ranking results. The suppression factor is pre-set, and the numerical value of the suppression factor has nothing to do with the cell, but is only related to the ranking result, that is, the suppression factor is pre-set, and the pre-set suppression factors are sorted. When determining the suppression factor of each neighboring area, the ranking result of each neighboring area is matched with the ranking of the pre-set suppression factors, so as to determine the suppression factor of each neighboring area.

[0088] For example, the preset suppression factors are arranged in order: b1, b2, b3, and the recommended values ​​of each neighboring area are determined to be arranged in order: neighboring area 2, neighboring area 3, neighboring area 1, as shown in Table 2:

[0089] Table 2

[0090] Recommended value sorting results Neighborhood 2 Neighborhood 3 Neighborhood 1 Suppressor ranking b1 b2 b3

[0091] By matching the ranking results of each neighboring area with the ranking of the preset suppression factors, it can be determined that the suppression factor corresponding to neighboring area 2 is b1, the suppression factor corresponding to neighboring area 3 is b2, and the suppression factor corresponding to neighboring area 1 is b3.

[0092] After determining the suppression factor corresponding to each neighboring cell, the first signal parameter of the neighboring cell is adjusted by the suppression factor. Specifically, for each neighboring cell, the first signal parameter of the neighboring cell is adjusted based on the suppression factor of the neighboring cell to obtain the second signal parameter of the neighboring cell.

[0093] For example: the first signal parameter of neighboring cell 2 is adjusted by suppression factor b1 to obtain the second signal parameter of neighboring cell 2; the first signal parameter of neighboring cell 3 is adjusted by suppression factor b2 to obtain the second signal parameter of neighboring cell 3; the first signal parameter of neighboring cell 1 is adjusted by b3 to obtain the second signal parameter of neighboring cell 1.

[0094] The first signal parameter of the neighboring cell is adjusted by the suppression factor to obtain the second signal parameter of the neighboring cell. Since the suppression factor is determined based on the recommended value of the neighboring cell, and the recommended value is determined based on the first signal parameter of the neighboring cell and the interference information, the second signal parameter of the neighboring cell finally obtained is not only related to the signal quality of the cell, but also related to the interference information between the cell and each cell in the currently connected network. This ensures that when the recommended neighboring cell is determined based on the second signal parameter of each neighboring cell and the recommended neighboring cell is sent to the network side, not only the signal quality of the recommended neighboring cell received by the network side can be guaranteed, but also the interference problem between the recommended neighboring cell and each cell in the currently connected network can be avoided, thereby ensuring the communication quality.

[0095] The determination method disclosed in this embodiment determines the first signal parameter of each cell based on the frequency information of the cell corresponding to the current connection network and at least one neighboring cell, each cell at least includes the cell corresponding to the current connection network and the cell corresponding to at least one neighboring cell, and the first signal parameter represents the signal quality of each cell; determines the recommended value of each neighboring cell in at least one neighboring cell based on the first signal parameter and the interference information, and the interference information is used to represent the interference between at least one neighboring cell and the cell corresponding to the current connection network; sorts at least one neighboring cell according to the size of the recommended value to obtain a sorting result; determines the suppression factor of each neighboring cell based on the sorting result; adjusts the first signal parameter of the neighboring cell corresponding to the suppression factor based on the suppression factor of each neighboring cell to obtain the second signal parameter of each neighboring cell; determines the recommended neighboring cell based on the second signal parameter. After determining the recommended value of each neighboring cell, this solution sorts them according to the size of the recommended value, and determines the suppression factor corresponding to each neighboring cell based on the sorting result, so as to adjust the first signal parameter of the neighboring cell based on the suppression factor of each neighboring cell to obtain the second signal parameter of each neighboring cell, so that when the first signal parameter of each neighboring cell is adjusted, it is based on the sorting of its corresponding recommended value among all the recommended values ​​of the neighboring cells, thereby ensuring the accuracy of the adjustment of the first signal parameter.

[0096] Furthermore, in the determination method disclosed in this embodiment, the suppression factor of each neighboring area is determined based on the sorting result, which can be specifically:

[0097] If, in the sorting result, the recommendation value of the first neighboring area is ranked higher than the recommendation value of the second neighboring area, it is determined that the suppression factor corresponding to the first neighboring area is greater than the suppression factor corresponding to the second neighboring area.

[0098] That is, the preset inhibition factors also have corresponding rankings, and the values ​​of inhibition factors at different positions in the rankings are different. For example, the ranking of inhibition factors shown in Table 2 is: b1, b2, b3, wherein b1, b2 and b3 are different from each other, and the values ​​of b1, b2 and b3 have a certain regularity.

[0099] In this embodiment, the values ​​of the suppression factors b1, b2 and b3 decrease in sequence, that is, b1>b2>b3, such as: b1=1, b2=0.95, b3=0.9.

[0100] Since the sorting result of at least one neighboring area according to the size of the recommended value of each neighboring area corresponds to the sorting of the inhibition factor, the inhibition factor corresponding to the neighboring area ranked higher in the sorting result is greater than the inhibition factor corresponding to the neighboring area ranked lower in the sorting result.

[0101] As shown in Table 2, the recommended values ​​of each neighboring area are ranked as follows: neighboring area 2, neighboring area 3, and neighboring area 1, and the suppression factors are ranked as b1, b2, and b3, and b1=1, b2=0.95, and b3=0.9. Then, the suppression factor corresponding to neighboring area 2 is b1=1, the suppression factor corresponding to neighboring area 3 is b2=0.95, and the suppression factor corresponding to neighboring area 1 is b3=0.9.

[0102] After obtaining the suppression factor corresponding to each neighboring area, the first signal parameter of each neighboring area is adjusted to obtain the second signal parameter of each neighboring area. Specifically, the first signal parameter of each neighboring area is divided by the suppression factor corresponding to each neighboring area, and the quotient obtained can be determined as the second signal parameter of the neighboring area.

[0103] For example, the first signal parameter used to characterize the signal quality of neighboring cell 1 is -105db, the first signal parameter of neighboring cell 2 is -106db, and the first signal parameter of neighboring cell 3 is -102db, as shown in Table 3:

[0104] Table 3

[0105] Neighborhood 1 Neighborhood 2 Neighborhood 3 First signal parameter -105db -106db -102db

[0106] After determining the first signal parameter of each neighboring cell, it is necessary to determine the recommended value of each neighboring cell, and sort the recommended values ​​to obtain a sorting result, as shown in Table 1, where the recommended value of neighboring cell 1 is 5, the recommended value of neighboring cell 2 is 9, and the recommended value of neighboring cell 3 is 7. The sorting result of the recommended values ​​is: neighboring cell 2, neighboring cell 3, and neighboring cell 1. The suppression factor of each neighboring cell is determined based on the sorting result of the recommended values.

[0107] The correspondence between the ranking results of the recommended values ​​of each neighboring area and the ranking of the suppression factor can be shown in Table 2. The suppression factor corresponding to neighboring area 1 is b3=0.9, the suppression factor corresponding to neighboring area 2 is b1=1, and the suppression factor corresponding to neighboring area 3 is b2=0.95. The second signal parameters of each neighboring area are calculated as follows: for neighboring area 1, the first signal parameter is divided by b3, which is: -105÷0.9=-116db, so the second signal parameter of neighboring area 1 is -116db; for neighboring area 2, the first signal parameter is divided by b1, which is: -106÷1=-106db, so the second signal parameter of neighboring area 2 is -106db; for neighboring area 3, the first signal parameter is divided by b2, which is -102÷0.95=-107db.

[0108] As shown in Table 4, the relevant information of each neighboring area is as follows:

[0109] Table 4

[0110]

[0111] After the second signal parameters of each neighboring cell are obtained, a recommended neighboring cell is determined based on the second signal parameters of each neighboring cell.

[0112] Based on the determination method disclosed in this embodiment, the higher the ranking of the recommended value of the neighboring area, the larger the corresponding suppression factor, and the lower the ranking of the recommended value of the neighboring area, the smaller the corresponding suppression factor. Since the first signal parameter of each neighboring area is a negative value, the suppression of the first signal parameter is smaller for the neighboring area with a higher ranking of the recommended value, and the suppression of the first signal parameter is larger for the neighboring area with a lower ranking of the recommended value, so as to ensure the accuracy of the obtained second signal parameter, so that the recommended neighboring area can be more in line with the current connection network.

[0113] This embodiment discloses a determination method, and its flow chart is as follows: Figure 3 As shown, including:

[0114] Step S31: determining a first signal parameter of each cell based on frequency information of a cell corresponding to the currently connected network and at least one neighboring cell, wherein each cell at least includes a cell corresponding to the currently connected network and at least one neighboring cell, and the first signal parameter represents a signal quality of the corresponding cell;

[0115] Step S32: determining a recommended value for each neighboring cell in at least one neighboring cell based on the first signal parameter and interference information, where the interference information is used to characterize interference between the at least one neighboring cell and a cell corresponding to the currently connected network;

[0116] Step S33: comparing the recommended value of performing an add operation on each neighboring area in at least one neighboring area with the recommended value of performing no add operation, wherein the add operation is used to add the neighboring area to the currently connected network;

[0117] Step S34, selecting the neighboring area corresponding to the recommended value for performing the adding operation is greater than the recommended value for performing the not adding operation, and sorting the recommended values ​​for performing the adding operation on the selected neighboring areas to obtain a sorting result;

[0118] Step S35: determining the suppression factor of each neighboring area based on the sorting result;

[0119] Step S36: adjusting the first signal parameter of the neighboring area corresponding to the suppression factor based on the suppression factor of each neighboring area to obtain the second signal parameter of each neighboring area;

[0120] Step S37: Determine a recommended neighboring area based on the second signal parameter.

[0121] When adjusting the first signal parameter of a neighboring area based on the recommended value of each neighboring area to obtain the second signal parameter of each neighboring area, it is necessary to sort at least one neighboring area according to the size of the recommended value of each neighboring area to obtain a sorting result, and determine a suppression factor for each neighboring area based on the sorting result, and adjust the first signal parameter of the corresponding neighboring area based on the suppression factor to obtain the second signal parameter of the corresponding neighboring area.

[0122] Among them, when sorting at least one neighboring area according to the size of the recommended value of each neighboring area, the size of the recommended value for performing an adding operation on each neighboring area and the recommended value for performing no adding operation can be determined, and the neighboring areas whose recommended value for performing the adding operation is greater than the recommended value for performing no adding operation are selected, and the recommended values ​​for performing the adding operation on these selected neighboring areas are sorted.

[0123] When determining the recommended value for each neighboring area in at least one neighboring area based on the first signal parameter and the interference information, the recommended value for performing an add operation on each neighboring area and the recommended value for performing a non-add operation on each neighboring area can be obtained. Among them, performing an add operation on a neighboring area means adding the neighboring area to the currently connected network so that the neighboring area is part of the currently connected network; not performing an add operation on a neighboring area means not adding the neighboring area to the currently connected network. The first signal parameter and the interference information can be used to determine the recommended value for each neighboring area when adding it to the currently connected network, that is, the value for recommending adding the neighboring area to the currently connected network and the value for recommending not adding the neighboring area to the currently connected network.

[0124] For a certain neighboring area, if the recommended value for performing an add operation is greater than the recommended value for performing a non-add operation, it can be determined that the neighboring area can be added to the current connection network; if the recommended value for performing an add operation is less than the recommended value for performing a non-add operation, it can be determined that the neighboring area should not be added to the current connection network. Based on this, when at least one neighboring area is sorted according to the size of the recommended value of each neighboring area, only the neighboring areas corresponding to the recommended values ​​for performing an add operation greater than the recommended values ​​for performing a non-add operation are selected, and the recommended values ​​for performing an add operation corresponding to these neighboring areas are sorted, while the neighboring areas corresponding to the recommended values ​​for performing an add operation less than the recommended values ​​for performing a non-add operation can be directly ignored.

[0125] Furthermore, after selecting the neighboring area corresponding to the recommendation value for performing the adding operation which is greater than the recommendation value for performing the not adding operation, the recommendation values ​​for performing the adding operation on the selected neighboring areas are sorted to ensure the accuracy of the sorting.

[0126] For example, there are four neighboring cells, namely, neighboring cell 1, neighboring cell 2, neighboring cell 3 and neighboring cell 4. As shown in Table 5, it is a recommended value table of recommended values ​​for performing the adding operation and recommended values ​​for performing the non-adding operation on the four neighboring cells.

[0127] Table 5

[0128] Recommended values ​​for performing add operations Recommended value for performing no-add operations Neighborhood 1 3 10 Neighborhood 2 10 5 Neighborhood 3 9 4 Neighborhood 4 3 1

[0129] Among them, for neighboring area 1, its recommended value 3 for performing the adding operation is less than the recommended value 10 for performing the not adding operation, therefore, the neighboring area 1 is directly ignored; for neighboring area 2, its recommended value 10 for performing the adding operation is greater than the recommended value 5 for performing the not adding operation, therefore, the recommended value 10 for neighboring area 2 to perform the adding operation is used for sorting; for neighboring area 3, its recommended value 9 for performing the adding operation is greater than the recommended value 4 for performing the not adding operation, therefore, the recommended value 9 for neighboring area 3 to perform the adding operation is used for sorting; for neighboring area 4, its recommended value 3 for performing the adding operation is greater than the recommended value 1 for performing the not adding operation, therefore, the recommended value 3 for neighboring area 4 to perform the adding operation is used for sorting.

[0130] Therefore, for the four neighboring areas in Table 5, neighboring area 2, neighboring area 3, and neighboring area 4 can be used for sorting, that is, the selected neighboring areas. The recommended values ​​for performing the adding operation corresponding to these selected neighboring areas are sorted, and the sorting results can be shown in Table 6:

[0131] Table 6

[0132] Sorting number Neighborhood Recommended values ​​for performing add operations 1 Neighborhood 2 10 2 Neighborhood 3 9 3 Neighborhood 4 3

[0133] The sorting result only includes the neighboring areas whose recommended values ​​for performing the adding operation are greater than the recommended values ​​for performing the not-adding operation, and the sorting result is based on the recommended values ​​for performing the adding operation corresponding to the neighboring areas.

[0134] Furthermore, the method disclosed in this embodiment may also include:

[0135] Determine from the sorting results the neighboring areas whose recommendation values ​​for performing the adding operation are less than a certain value, and delete the neighboring areas from the sorting results, that is, for the neighboring areas whose recommendation values ​​for performing the adding operation are less than a certain value, they are not considered as recommended neighboring areas, and only at least some of the neighboring areas from the neighboring areas whose recommendation values ​​for performing the adding operation are greater than the value are selected as recommended neighboring areas.

[0136] The determination method disclosed in the present embodiment, when determining the second signal parameter of each neighboring cell based on the recommended value of each neighboring cell, only selects the recommended values ​​for performing the adding operation corresponding to the neighboring cells whose recommended values ​​for performing the adding operation are greater than the recommended values ​​for performing the not-adding operation for sorting, so as to ensure that only the recommended values ​​for performing the adding operation are included in the sorting, and only the neighboring cells with larger recommended values ​​for performing the adding operation are sorted, excluding the neighboring cells with larger recommended values ​​for performing the not-adding operation, so as to ensure that the sorted neighboring cells are more recommended for performing the adding operation, thereby ensuring the accuracy of the recommended neighboring cells determined based on this scheme, and avoiding the existence of neighboring cells sorted according to the recommended values ​​for performing the not-adding operation, so that the communication quality is reduced when the neighboring cells are selected as the secondary cell of the electronic device.

[0137] Furthermore, in the determination method disclosed in this embodiment, determining the recommended neighboring area based on the second signal parameter of each neighboring area in at least one neighboring area may be specifically as follows:

[0138] At least one neighboring area is sorted according to the size of the second signal parameter of each neighboring area in at least one neighboring area to obtain a signal parameter sorting result; based on the signal parameter sorting result, a specific number of neighboring areas whose second signal parameters meet the conditions are determined, and the specific number of neighboring areas are determined as recommended neighboring areas.

[0139] After determining the second signal parameter of each neighboring area, the neighboring areas can be sorted according to the second signal parameter to obtain a signal parameter sorting result. The signal parameter sorting result is sorted in descending order, such as sorting according to the second signal parameter in Table 4, that is: -106db, -107db, -116db, and the corresponding neighboring areas are: neighboring area 2, neighboring area 3, and neighboring area 1, respectively.

[0140] If the specific number of neighboring cells for which the second signal parameter satisfies the condition is: 2 neighboring cells, neighboring cells 2 and 3 may be selected as recommended neighboring cells according to the signal parameter sorting result, and sent to the network side, which determines whether to use neighboring cells 2 and 3 as secondary cells of the electronic device;

[0141] If the specific number of neighboring cells whose second signal parameter satisfies the condition is: the neighboring cells whose second signal parameter is greater than the specific value, and the specific number is 3, then according to the signal parameter sorting result, it is determined that neighboring cells 2, 3, and 1 are all the specific number of neighboring cells whose second signal parameter satisfies the condition, and neighboring cells 2, 3, and 1 are selected as recommended neighboring cells, which are sent to the network side, and the network side determines whether to use neighboring cells 2, 3, and 1 as secondary cells of the electronic device;

[0142] Alternatively, when the specific number is 2, two neighboring areas are selected from the three neighboring areas whose second signal parameters are greater than a specific value according to the signal parameter sorting results. The selection criteria may be: according to the signal parameter sorting results, two neighboring areas with large second signal parameters are selected as recommended neighboring areas in descending order, that is, neighboring areas 2 and 3 are determined to be the specific number of neighboring areas whose second signal parameters meet the conditions, and neighboring areas 2 and 3 are selected as recommended neighboring areas, which are sent to the network side, and the network side determines whether to use neighboring areas 2 and 3 as secondary cells of the electronic device.

[0143] After determining the recommended neighboring area, the recommended neighboring area and related information need to be sent to the network side, and the network side determines whether to determine the recommended neighboring area as the secondary cell of the electronic device. Sending the recommended neighboring area and related information to the network side can specifically include: sending the identification information of the recommended neighboring area and the second signal parameters corresponding to the recommended neighboring area to the network side, so that the network side can select the secondary cell of the electronic device from the recommended neighboring area based on the second signal parameters of the recommended neighboring area.

[0144] The determination method disclosed in this embodiment does not directly determine all neighboring areas and the second signal parameters corresponding to each neighboring area as recommended neighboring areas after determining the second signal parameters of each neighboring area, but sorts at least one neighboring area based on the size of the second signal parameter to obtain a signal parameter sorting result, so as to select at least some of the neighboring areas from at least one neighboring area as recommended neighboring areas based on the signal parameter sorting result, thereby reducing the data processing amount on the network side and improving the communication processing efficiency on the network side.

[0145] This embodiment discloses a determination method, and its flow chart is as follows: Figure 4 As shown, including:

[0146] Step S41: determining a first signal parameter of each cell based on frequency information of a cell corresponding to the currently connected network and at least one neighboring cell, wherein each cell at least includes a cell corresponding to the currently connected network and at least one neighboring cell, and the first signal parameter represents a signal quality of each cell;

[0147] Step S42: inputting a first signal parameter of a cell in the currently connected network and a first signal parameter of a third neighboring cell in at least one neighboring cell into a pre-trained deep learning model to obtain a recommended value of the third neighboring cell output by the deep learning model, wherein the deep learning model is obtained by model training using training data and a simulation environment, and the training data at least includes: a first signal training parameter of a cell corresponding to a connected network and a first signal training parameter of at least one neighboring cell, wherein the at least one neighboring cell is a cell corresponding to at least one network to be connected;

[0148] Step S43: adjusting the first signal parameter of at least one neighboring cell based on the recommended value to determine the second signal parameter of each neighboring cell;

[0149] Step S44: Determine a recommended neighboring area based on the second signal parameter.

[0150] When determining the recommended value of each neighboring area in at least one neighboring area based on the first signal parameter and the interference information, a deep learning model can be used for implementation.

[0151] In the determination method disclosed in this embodiment, the deep learning model can be specifically: DQN (Deep Q-Network) model. The DQN model is an algorithm that combines deep learning and reinforcement learning. It uses a deep neural network to approximate the Q-value function, so that it can handle problems in high-dimensional state space.

[0152] In this embodiment, the input of the deep learning model is: the first signal parameter of each cell in the currently connected network and the first signal parameter of the neighboring cell to be connected, and the output of the deep learning model is: the recommended value of the neighboring cell to be connected. The neighboring cell to be connected is one of the at least one neighboring cell, that is, the third neighboring cell of the at least one neighboring cell described in this embodiment.

[0153] It should be noted that it is necessary to use a deep learning model to determine the recommended value for each of at least one neighboring area. Therefore, it is necessary to input the first signal parameters of each of the at least one neighboring area into the deep learning model in sequence. When inputting the first signal parameters of each neighboring area into the deep learning model, it is necessary to simultaneously input the first signal parameters of each cell corresponding to the currently connected network into the deep learning model, so that the recommended value of the neighboring area output by the deep learning model is not only determined based on the signal quality of the neighboring area (ie, the first signal parameter), but also based on whether there is interference between the neighboring area and the cell corresponding to the currently connected network.

[0154] If the signal quality represented by the first signal parameter of a neighboring cell is poor, even if there is no signal interference between the neighboring cell and the cell corresponding to the currently connected network, the recommendation value of the neighboring cell output by the deep learning model will not be large due to its poor signal quality; if a neighboring cell has signal interference with the cell corresponding to the currently connected network, even if the signal quality represented by the corresponding first signal parameter is good, the recommendation value of the neighboring cell output by the deep learning model will not be large.

[0155] Among them, the deep learning model is obtained by model training using training data and a simulation environment. The simulation environment is built based on the application scenario of the deep learning model. The training data includes at least: the first signal training parameters of the cell corresponding to the connected network and the first signal training parameters of at least one neighboring cell. The at least one neighboring cell is a cell corresponding to at least one network to be connected.

[0156] A simulation environment is a virtual environment that simulates the real world or a specific scenario, providing a controllable experimental platform for reinforcement learning. In the simulation environment, the intelligent agent can interact with the simulation environment, perform actions and obtain feedback from the environment.

[0157] Among them, the first signal training parameters of the cell corresponding to the connected network are used to characterize the signal quality of the cell corresponding to the connected network, and the first signal training parameters of at least one neighboring area to be connected are used to characterize the signal quality of at least one neighboring area to be connected. The first signal training parameters of the cell corresponding to the connected network and the first signal training parameters of at least one neighboring area are used as training data, and the model training is performed using the constructed simulation environment to obtain a trained deep learning model, so that the trained deep learning model can determine the recommended value of the neighboring area based on the first signal parameters of the cell corresponding to the currently connected network and the first signal parameters of the neighboring area.

[0158] Specifically, the deep learning model obtained by model training using training data and simulation environment can be:

[0159] Based on the simulation environment, in the first state, the training data is input into the neural network model to obtain the probability of the neural network model outputting the transition from the first state to the second state; the reward value is determined based on the probability of the transition from the first state to the second state; the deep learning model is trained based on the first state, the second state and the reward value to obtain a trained deep learning model, and the recommended value of the deep learning model is used to characterize the cumulative reward value.

[0160] The neural network model may be a DNN (Deep Neural Network) model, which is used for supervised learning and unsupervised learning to handle complex pattern recognition and data fitting problems.

[0161] The simulation environment E includes at least four elements, namely E=<S,A,P,R>, where S represents the state, A represents the action, P represents the probability of state transition, and R represents the reward.

[0162] Among them, S represents the state of the electronic device, and the electronic device can be in different states, such as: configuration state, in the configuration state, it can correspond to: neighboring area adding success state Add success, and neighboring area adding failure state Add Fail, in the neighboring area adding success state, it can correspond to: the state of the bit error rate lower than the first threshold, the state of the bit error rate higher than the first threshold and lower than the second threshold, the state of the bit error rate higher than the second threshold and lower than the third threshold, and the data stagnation state, wherein, when the electronic device is in a state where the bit error rate is lower than the first threshold, the signal quality when the electronic device transmits data through the currently connected network is good; when the electronic device is in a state where the bit error rate is higher than the first threshold and lower than the second threshold, the signal quality when the electronic device transmits data through the currently connected network is poor, when the electronic device is in a state where the bit error rate is higher than the second threshold, the signal quality when the electronic device transmits data through the currently connected network is very poor, and when the electronic device is in a data stagnation state, the electronic device cannot transmit data through the currently connected network.

[0163] A represents the execution of an action of adding or not adding a neighboring cell. The action of adding a neighboring cell is to add the neighboring cell to the currently connected network, and the data of the electronic device is transmitted jointly by the cells already included in the currently connected network before the execution of the adding action and the added neighboring cell; the action of not adding a neighboring cell is to not add the neighboring cell to the currently connected network, and the data of the electronic device is transmitted only through the cells included in the currently connected network.

[0164] P represents the probability that when the electronic device is in state s, action a is executed, and then the electronic device transfers to state s'. This probability can be determined through a neural network model, that is, the neural network model is used to simulate the execution of action a in state s to obtain the probability of state transfer to s'. For example: simulate that the electronic device is currently in a state of successful neighbor addition. In this state, the simulated electronic device performs data services and transmits data through the currently connected network. At this time, the neural network model can output the probability that the electronic device is in a state where the bit error rate is higher than the first threshold and lower than the second threshold.

[0165] R represents the reward value corresponding to the transfer of the electronic device to different states, such as: when the electronic device is in the configuration state, the reward value is 0; when the electronic device is in the neighboring cell adding failure state, the reward value is -10; when the electronic device is in the neighboring cell adding success state, the reward value is 0; when the electronic device is in the state where the bit error rate is lower than the first threshold, the reward value is 10; when the electronic device is in the state where the bit error rate is higher than the first threshold and lower than the second threshold, the reward value is 1; when the electronic device is in the state where the bit error rate is higher than the second threshold and lower than the third threshold, the reward value is -1; when the electronic device is in the data stagnation state, the reward value is -10.

[0166] Among them, the first threshold value can be 10%, the second threshold value can be 20%, and the third threshold value can be 30%. Alternatively, the first threshold value, the second threshold value and the third threshold value can also be other data, which are only exemplified here.

[0167] like Figure 5 As shown, it is a schematic diagram of the correspondence between the state and the reward value. Based on the configuration state, the neighbor cell adding success state and the neighbor cell adding failure state can be determined. In the neighbor cell adding success state, it can correspond to the state where the bit error rate is lower than the first threshold (such as: 10%), the state where the bit error rate is higher than the first threshold and lower than the second threshold (such as: 20%), the state where the bit error rate is higher than the second threshold and lower than the third threshold (such as: 30%), and the data stagnation state.

[0168] Among them, the schematic diagram of model training for the deep learning model can be shown as follows Figure 6 As shown, using the constructed simulation environment, in the current first state S1, action a is executed t , input the training data (i.e., the first signal training parameters of the currently connected network and the first signal training parameters of the neighboring area to be connected) into the neural network model in the simulation environment, and use the output of the neural network model to transfer from the first state to the second state S t+1 The probability P of the transfer to the second state is determined based on the probability P output by the neural network model within the simulation environment. t+1 , and the reward value r obtained by the simulation environment t+1 , the second state S t+1 And the initial first state S1, as a set of data, is used to train the deep learning model. It is necessary to use the simulation environment to obtain multiple sets of data in order to complete the training of the deep learning model.

[0169] Furthermore, if the first state represents the configuration state for adding the target neighboring area, the second state represents the failure to add the target neighboring area, and the probability of the first state transferring to the second state is greater than the first target value, the reward value is determined to be the first reward value; if the first state represents the configuration state for adding the target neighboring area, the second state represents the success of adding the target neighboring area, and the probability of the first state transferring to the second state is greater than the second target value, the reward value is determined to be the second reward value, and the second reward value is greater than the first reward value; if the first state represents the success of adding the target neighboring area, and the second state represents the range of the bit error rate when transmitting data services through the corresponding cell of the connected network and the target neighboring area, the reward value is determined to be a third reward value based on the range of the bit error rate in the second state, and the third reward value is not less than the first reward value.

[0170] Specifically, if the first state is the configuration state, execute action a t , the training data (i.e., the first signal training parameters of the currently connected network and the first signal training parameters of the neighboring area to be connected) are input into the neural network model in the simulation environment, and the neural network model outputs the probability of switching from the configuration state to the neighboring area adding success state, or the probability of switching to the neighboring area adding failure state.

[0171] If the neural network model outputs the probability of a neighboring area adding failure state, and the probability of a neighboring area adding failure state is greater than the first target value, the reward value can be determined to be the first reward value corresponding to the neighboring area adding failure state, such as -10; if the neural network model outputs the probability of a neighboring area adding success state, and the probability of a neighboring area adding success state is greater than the second target value, the reward value can be determined to be the second reward value corresponding to the neighboring area adding success state, such as 0; or, after determining that the neural network model outputs the probability of a neighboring area adding success state, and the probability of a neighboring area adding success state is greater than the second target value, the action of adding the neighboring area is performed, and after determining that the addition is successful, the reward value is determined to be the second reward value corresponding to the neighboring area adding success state;

[0172] Alternatively, it can also be: if it is determined that the probability of the neighboring area adding success state output by the neural network is greater than the probability of the neighboring area adding failure state, the reward value is determined to be the reward value corresponding to the neighboring area adding success state; if it is determined that the probability of the neighboring area adding failure state output by the neural network is greater than the probability of the neighboring area adding success state, the reward value is determined to be the reward value corresponding to the neighboring area adding failure state. For example: the probability of the neighboring area adding success state output by the neural network model is 60%, and the probability of the neighboring area adding failure state is 40%. Since 60%>40%, the reward value is determined to be the reward value corresponding to the neighboring area adding success state.

[0173] In addition, if the first state is the state of successful addition of the neighboring cell, at this time, the current connected network after the successful addition of the neighboring cell is simulated through the simulation environment to execute data services, and the first signal training parameters of each cell corresponding to the current connected network after the successful addition of the neighboring cell are input into the neural network model, and the probability of state transition output by the neural network model can be obtained, that is, the probability of the electronic device switching from the state of successful addition of the neighboring cell to other states. The other states may be: a state in which the bit error rate is lower than the first threshold, a state in which the bit error rate is higher than the first threshold and lower than the second threshold, a state in which the bit error rate is higher than the second threshold and lower than the third threshold, and a data stagnation state.

[0174] If the neural network model outputs the probability of switching from the state of successful neighboring area addition to the state of a bit error rate lower than the first threshold, the reward value can be determined as the reward value corresponding to the state of the bit error rate lower than the first threshold, such as: 10; if the neural network model outputs the probability of switching from the state of successful neighboring area addition to the state of a bit error rate higher than the first threshold and lower than the second threshold, the reward value can be determined as the reward value corresponding to the state of the bit error rate higher than the first threshold and lower than the second threshold, such as: 1; if the neural network model outputs the probability of switching from the state of successful neighboring area addition to the state of a bit error rate higher than the second threshold and lower than the third threshold, the reward value can be determined as the reward value corresponding to the state of the bit error rate higher than the second threshold and lower than the third threshold, such as: -1; if the neural network model outputs the probability of switching from the state of successful neighboring area addition to the state of data stagnation, the reward value can be determined as the reward value corresponding to the data stagnation state, such as: -10.

[0175] By using the training data and the simulation environment, multiple sets of data can be obtained using the above method, and the deep learning model is trained using the obtained multiple sets of data to obtain a trained deep learning model.

[0176] Among them, it should be noted that each set of data obtained using the simulation environment, in addition to including the first signal training parameters of the cells in the connected network and the first signal training parameters of the neighboring cells to be connected, also needs to include a reward value. The deep learning model is obtained by using such multiple sets of data for model training. When the deep learning model is in use, the first signal parameters of each cell in the currently connected network and the first signal parameters of the neighboring cells to be connected are input into the deep learning model, and the deep learning model outputs a recommended value, wherein the recommended value is related to the reward value, and the recommended value represents the cumulative reward value, that is, when an action is performed in one state and the corresponding reward value is obtained when entering the next state, the same action is performed multiple times in the same state, and the multiple reward values ​​obtained when entering the next state are accumulated to obtain the corresponding recommended value, that is, the data required for training the deep learning model includes the reward value, and when the deep learning model is applied, the deep learning model outputs the recommended value.

[0177] The determination method disclosed in this embodiment inputs the first signal parameters of the cell corresponding to the currently connected network and the first signal parameters of the third neighboring area in at least one neighboring area into a pre-trained deep learning model to obtain a recommended value of the third neighboring area output by the deep learning model. The deep learning model is obtained by model training using training data and a simulation environment. The first signal parameters of the cell in the currently connected network and the first signal parameters of the neighboring area are input into the deep learning model to determine the recommended value of the neighboring area, thereby ensuring the accuracy of the recommended value of the determined neighboring area and ensuring that the recommended value of the neighboring area is not only related to the signal quality of the neighboring area, but also related to the interference information between the neighboring area and the cell in the current network, thereby avoiding the problem of interference between the finally recommended neighboring area and the cell in the currently connected network.

[0178] This embodiment discloses a determination system, and its structural diagram is as follows: Figure 7 As shown, including:

[0179] A first determining unit 71 , a second determining unit 72 , an adjusting unit 73 and a third determining unit 74 .

[0180] The first determination unit 71 is used to determine the first signal parameter of each cell based on the frequency information of the cell corresponding to the currently connected network and at least one neighboring cell, each cell at least includes the cell corresponding to the currently connected network and the cell corresponding to at least one neighboring cell, and the first signal parameter represents the signal quality of each cell;

[0181] The second determination unit 72 is used to determine the recommended value of each neighboring cell in at least one neighboring cell based on the first signal parameter and the interference information, where the interference information is used to characterize the interference between the at least one neighboring cell and the cell corresponding to the currently connected network;

[0182] The adjustment unit 73 is used to adjust the first signal parameter of at least one neighboring cell based on the recommended value to determine the second signal parameter of each neighboring cell;

[0183] The third determining unit 74 is configured to determine a recommended neighboring cell based on the second signal parameter.

[0184] Furthermore, the adjustment unit is used to:

[0185] Sorting at least one neighboring area according to the size of the recommended value to obtain a sorting result; determining a suppression factor for each neighboring area based on the sorting result; and adjusting a first signal parameter of the neighboring area corresponding to the suppression factor based on the suppression factor of each neighboring area to obtain a second signal parameter of each neighboring area;

[0186] The adjustment unit can be used in particular for:

[0187] For each neighboring cell, the first signal parameter of the neighboring cell is adjusted based on the suppression factor of the neighboring cell to obtain the second signal parameter of the neighboring cell.

[0188] Furthermore, the adjustment unit is used to:

[0189] If, in the sorting result, the recommendation value of the first neighboring area is ranked higher than the recommendation value of the second neighboring area, it is determined that the suppression factor corresponding to the first neighboring area is greater than the suppression factor corresponding to the second neighboring area.

[0190] Further, the third determining unit is used for:

[0191] At least one neighboring area is sorted according to the size of the second signal parameter of each neighboring area in at least one neighboring area to obtain a signal parameter sorting result; based on the signal parameter sorting result, a specific number of neighboring areas whose second signal parameters meet the conditions are determined, and the specific number of neighboring areas are determined as recommended neighboring areas.

[0192] Further, the second determining unit is used for:

[0193] The first signal parameters of a cell in a currently connected network and the first signal parameters of a third neighboring cell in at least one neighboring cell are input into a pre-trained deep learning model to obtain a recommended value of the third neighboring cell output by the deep learning model; wherein the deep learning model is obtained by model training using training data and a simulation environment, and the training data includes at least: first signal training parameters of a cell corresponding to a connected network and first signal training parameters of at least one neighboring cell, and at least one neighboring cell is a cell corresponding to at least one network to be connected.

[0194] Further, the training of deep learning models includes:

[0195] Based on the simulation environment, in the first state, the training data is input into the neural network model to obtain the probability of the neural network model outputting the transition from the first state to the second state; the reward value is determined based on the probability of the transition from the first state to the second state; the deep learning model is trained based on the first state, the second state and the reward value to obtain a trained deep learning model, and the recommended value of the deep learning model is used to characterize the cumulative reward value.

[0196] Further, the training of deep learning models includes:

[0197] If the first state represents the configuration state of adding the target neighboring area, the second state represents the failure of adding the target neighboring area, the probability of the first state transferring to the second state is greater than the first target value, and the reward value is determined to be the first reward value; if the first state represents the configuration state of adding the target neighboring area, the second state represents the success of adding the target neighboring area, the probability of the first state transferring to the second state is greater than the second target value, and the reward value is determined to be the second reward value, and the second reward value is greater than the first reward value; if the first state represents the success of adding the target neighboring area, the second state represents the range of the bit error rate when transmitting data services through the corresponding cell of the connected network and the target neighboring area, the reward value is determined to be the third reward value based on the range of the bit error rate in the second state, and the third reward value is not less than the first reward value.

[0198] Furthermore, the adjustment unit is used to:

[0199] Compare the recommended value of performing an add operation on each neighboring area in at least one neighboring area with the recommended value of performing no add operation, wherein the add operation is used to add the neighboring area to the currently connected network; select the neighboring area corresponding to the recommended value of performing the add operation that is greater than the recommended value of performing the no add operation, sort the recommended values ​​of performing the add operation on the selected neighboring areas, and obtain the sorting result.

[0200] The determination system disclosed in this embodiment is implemented based on the determination method disclosed in the above embodiment, which will not be described in detail here.

[0201] The determination system disclosed in this embodiment determines the first signal parameter of each cell based on the frequency information of the cell corresponding to the currently connected network and at least one neighboring cell, each cell includes at least the cell corresponding to the currently connected network and the cell corresponding to at least one neighboring cell, and the first signal parameter characterizes the signal quality of each cell; determines the recommended value of each neighboring cell in at least one neighboring cell based on the first signal parameter and interference information, and the interference information is used to characterize the interference situation between at least one neighboring cell and the cell corresponding to the currently connected network; adjusts the first signal parameter of at least one neighboring cell based on the recommended value, and determines the second signal parameter of each neighboring cell; determines the recommended neighboring cell based on the second signal parameter. When determining the recommended cell, this scheme not only needs to determine the first signal parameters of each neighboring cell, but also needs to determine the first signal parameters of the cell corresponding to the currently connected network, and determine the recommended value of each neighboring cell based on the first signal parameters and the interference information between the neighboring cell and the cell corresponding to the currently connected network, and determine the recommended neighboring cell based on this, ensuring that when recommending the neighboring cell, it can be determined not only based on the signal quality of each neighboring cell, but also based on the interference information between each neighboring cell and the cell corresponding to the currently connected network, so as to ensure that the finally recommended neighboring cell is a cell that can meet the current connection network, and avoid the situation where the communication quality deteriorates after adding neighboring cells due to interference between different cells.

[0202] This embodiment discloses an electronic device, and its structural diagram is as follows: Figure 8 As shown, including:

[0203] Processor 81 and memory 82.

[0204] The processor 81 is used to determine the first signal parameter of each cell based on the frequency information of the cell corresponding to the currently connected network and at least one neighboring cell, where each cell includes at least the cell corresponding to the currently connected network and the cell corresponding to at least one neighboring cell, and the first signal parameter characterizes the signal quality of each cell; determine the recommended value of each neighboring cell in at least one neighboring cell based on the first signal parameter and the interference information, and the interference information is used to characterize the interference between at least one neighboring cell and the cell corresponding to the currently connected network; adjust the first signal parameter of at least one neighboring cell based on the recommended value to determine the second signal parameter of each neighboring cell; determine the recommended neighboring cell based on the second signal parameter;

[0205] The memory 82 is used to store the programs required by the processor to execute the above processing procedures.

[0206] The electronic device disclosed in this embodiment is implemented based on the determination method disclosed in the above embodiment, which will not be described in detail here.

[0207] The electronic device disclosed in this embodiment determines the first signal parameter of each cell based on the frequency information of the cell corresponding to the currently connected network and at least one neighboring cell, each cell at least includes the cell corresponding to the currently connected network and the cell corresponding to at least one neighboring cell, and the first signal parameter characterizes the signal quality of each cell; determines the recommended value of each neighboring cell in at least one neighboring cell based on the first signal parameter and interference information, and the interference information is used to characterize the interference situation between at least one neighboring cell and the cell corresponding to the currently connected network; adjusts the first signal parameter of at least one neighboring cell based on the recommended value, and determines the second signal parameter of each neighboring cell; determines the recommended neighboring cell based on the second signal parameter. When determining the recommended cell, this scheme not only needs to determine the first signal parameters of each neighboring cell, but also needs to determine the first signal parameters of the cell corresponding to the currently connected network, and determine the recommended value of each neighboring cell based on the first signal parameters and the interference information between the neighboring cell and the cell corresponding to the currently connected network, and determine the recommended neighboring cell based on this, ensuring that when recommending the neighboring cell, it can be determined not only based on the signal quality of each neighboring cell, but also based on the interference information between each neighboring cell and the cell corresponding to the currently connected network, so as to ensure that the finally recommended neighboring cell is a cell that can meet the current connection network, and avoid the situation where the communication quality deteriorates after adding neighboring cells due to interference between different cells.

[0208] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0209] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0210] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0211] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A determination method, comprising: Determine a first signal parameter of each cell based on frequency information of a cell corresponding to the currently connected network and at least one neighboring cell, wherein each cell at least includes a cell corresponding to the currently connected network and a cell corresponding to the at least one neighboring cell, and the first signal parameter represents a signal quality of each cell; Determine a recommended value for each of the at least one neighboring cell based on the first signal parameter and interference information, wherein the interference information is used to characterize interference between the at least one neighboring cell and a cell corresponding to the currently connected network; Adjusting the first signal parameter of the at least one neighboring cell based on the recommended value to determine the second signal parameter of each neighboring cell; Based on the second signal parameter, a recommended neighboring cell is determined.

2. The method according to claim 1, wherein adjusting the first signal parameter of the at least one neighboring cell based on the recommended value to determine the second signal parameter of each neighboring cell comprises: Sorting the at least one neighboring area according to the size of the recommendation value to obtain a sorting result; Determine a suppression factor for each neighboring area based on the ranking result; Based on the suppression factor of each neighboring cell, adjusting the first signal parameter of the neighboring cell corresponding to the suppression factor to obtain the second signal parameter of each neighboring cell; The adjusting the first signal parameter of the neighboring area corresponding to the suppression factor based on the suppression factor of each neighboring area to obtain the second signal parameter of each neighboring area includes: For each neighboring cell, the first signal parameter of the neighboring cell is adjusted based on the suppression factor of the neighboring cell to obtain the second signal parameter of the neighboring cell.

3. The method according to claim 2, wherein determining the suppression factor of each neighboring area based on the sorting result comprises: If in the sorting result, the recommendation value of the first neighboring area is ranked higher than the recommendation value of the second neighboring area, it is determined that the suppression factor corresponding to the first neighboring area is greater than the suppression factor corresponding to the second neighboring area.

4. The method according to claim 1, wherein determining the recommended neighboring area based on the second signal parameter comprises: Sorting the at least one neighboring area according to the magnitude of the second signal parameter of each neighboring area in the at least one neighboring area to obtain a signal parameter sorting result; A specific number of neighboring areas whose second signal parameters meet a condition is determined based on the signal parameter sorting result, and the specific number of neighboring areas are determined as recommended neighboring areas.

5. The method according to claim 1, wherein determining the recommended value of each neighboring cell in the at least one neighboring cell based on the first signal parameter and the interference information comprises: Inputting a first signal parameter of a cell in the currently connected network and a first signal parameter of a third neighboring area among the at least one neighboring area into a pre-trained deep learning model, and obtaining a recommended value of the third neighboring area output by the deep learning model; Among them, the deep learning model is obtained by model training using training data and a simulation environment, and the training data includes at least: the first signal training parameters of the cell corresponding to the connected network and the first signal training parameters of at least one neighboring cell, and the at least one neighboring cell is a cell corresponding to at least one network to be connected.

6. According to the method of claim 5, the deep learning model is obtained by model training using training data and a simulation environment, comprising: Based on the simulation environment, in a first state, inputting the training data into a neural network model to obtain a probability of the neural network model outputting a transition from the first state to a second state; determining a reward value based on the probability of transitioning from the first state to the second state; The deep learning model is trained based on the first state, the second state and the reward value to obtain a trained deep learning model, and the recommendation value of the deep learning model is used to represent the accumulated reward value.

7. The method according to claim 6, wherein determining the reward value based on the probability of transitioning from the first state to the second state comprises: If the first state represents a configuration state for adding a target neighboring area, the second state represents a failure in adding the target neighboring area, and a probability of the first state transitioning to the second state is greater than a first target value, determining the reward value to be the first reward value; If the first state represents a configuration state of adding a target neighboring area, the second state represents that the target neighboring area is successfully added, and the probability of the first state transitioning to the second state is greater than a second target value, the reward value is determined to be a second reward value, and the second reward value is greater than the first reward value; If the first state represents a successful addition of a target neighboring cell, the second state represents a range of a bit error rate when transmitting data services through the cell corresponding to the connected network and the target neighboring cell, and the reward value is determined to be a third reward value based on the range of the bit error rate in the second state, and the third reward value is not less than the first reward value.

8. According to the method of claim 2, the step of sorting the at least one neighboring area according to the size of the recommendation value to obtain a sorting result comprises: Comparing the recommended value of performing an add operation on each neighboring area in the at least one neighboring area with the recommended value of performing no add operation, wherein the add operation is used to add the neighboring area to the currently connected network; Select the neighboring area whose recommendation value for performing the adding operation is greater than the recommendation value for performing the not-adding operation, sort the recommendation values ​​for performing the adding operation on the selected neighboring areas, and obtain a sorting result.

9. A determination system comprising: A first determining unit, configured to determine a first signal parameter of each cell based on frequency information of a cell corresponding to a currently connected network and at least one neighboring cell, wherein the cells at least include a cell corresponding to the currently connected network and a cell corresponding to the at least one neighboring cell, and the first signal parameter represents a signal quality of each cell; A second determining unit, configured to determine a recommended value for each of the at least one neighboring cell based on the first signal parameter and interference information, wherein the interference information is used to characterize interference between the at least one neighboring cell and a cell corresponding to the currently connected network; an adjusting unit, configured to adjust the first signal parameter of the at least one neighboring cell based on the recommended value, and determine a second signal parameter of each neighboring cell; The third determining unit is used to determine a recommended neighboring area based on the second signal parameter.

10. An electronic device, comprising: A processor, configured to determine a first signal parameter of each cell based on frequency information of a cell corresponding to a currently connected network and at least one neighboring cell, wherein each cell includes at least a cell corresponding to the currently connected network and a cell corresponding to the at least one neighboring cell, and the first signal parameter represents a signal quality of each cell; determine a recommended value for each neighboring cell in the at least one neighboring cell based on the first signal parameter and interference information, wherein the interference information is used to represent the interference between at least one neighboring cell and the cell corresponding to the currently connected network; adjust the first signal parameter of the at least one neighboring cell based on the recommended value to determine a second signal parameter for each neighboring cell; and determine a recommended neighboring cell based on the second signal parameter; The memory is used to store the program required by the processor to execute the above processing process.

Images