Cell transmitting power control method and device, electronic equipment and storage medium

Abstract

The invention provides a cell transmitting power control method and device, electronic equipment and a storage medium, and relates to the technical field of communication. The method comprises the following steps: acquiring a cell network state data set of a first time period; wherein the cell network state data set comprises adjacent cell relation information between cells, first network state information of each cell at each moment in a first time period, connection relation information between each cell and a user terminal, and second network state information of each connected user terminal at each moment; and determining a transmitting power prediction result of each cell according to the cell network state data set so as to control the transmitting power of each cell. According to the method, the transmitting power of each cell is predicted by fully utilizing the adjacent cell relationship among the cells, the connection relationship between each cell and the user terminal and the network state information of each cell and each accessed user terminal.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a method, device, electronic equipment and storage medium for controlling cell transmission power. Background Art

[0002] Downlink power control in 5G systems is a key and fundamental issue. On the one hand, if the base station transmit power is low, path loss and shadow fading cannot be adequately compensated, resulting in weak coverage or even blind spots. On the other hand, if the base station signal transmit power is high, it will cause cross-area coverage and consume more energy, resulting in strong interference and high energy consumption. In the current network, the status of base station cells and users will change over time, so it is necessary to dynamically control the downlink power and adjust the base station transmit power to an appropriate value according to network changes.

[0003] In the related technology, the downlink power is mainly dynamically controlled according to the parameter setting on the base station cell side and the uplink feedback of the user. For example, the transmit power of the edge user-specific PDCCH is adaptively adjusted based on the PDCCH (Physical Downlink Control Channel) CCE (Control Channel Element) aggregation level and PUCCH (Physical Uplink Control Channel) feedback, and the PDSCH power spectrum density is adaptively adjusted based on the scheduling MCS (Modulation and Coding Scheme) of PDSCH (Physical Downlink Shared Channel), bandwidth, and estimated user received signal strength. However, the power control effect is poor, and there is still a problem that the downlink power of the base station does not match the actual demand. Summary of the invention

[0004] The embodiments of the present invention provide a method, device, electronic device and storage medium for controlling cell transmission power, so as to solve the problem that the power control strategy in the prior art does not match the downlink power of the base station with the actual demand.

[0005] In a first aspect, an embodiment of the present invention provides a method for controlling cell transmission power, comprising: obtaining a cell network status data set for a first time period; wherein the cell network status data set includes neighbor relationship information between each cell, first network status information of each cell at each moment in the first time period and connection relationship information with a user terminal, and second network status information of each connected user terminal at each moment; based on the cell network status data set, determining a transmission power prediction result for each of the cells for controlling the transmission power of each of the cells.

[0006] In some embodiments, determining the transmission power prediction result of each of the cells based on the cell network status data set includes: constructing a first cell user map based on the cell network status data set; the first cell user map is an information expression of the network status relationship between the cells and between the cells and the connected user terminals at each moment; inputting the first cell user map into a first power prediction model to obtain the transmission power prediction result of each of the cells; wherein the first power prediction model is trained based on the first cell user map sample and the cell transmission power label corresponding to the first cell user map sample.

[0007] As a possible implementation method, constructing a first cell user graph according to the cell network status data set includes: determining each of the cells as a cell node, and determining the first edge feature information between the cell nodes based on the neighbor relationship information between the cells; for the first moment in the first time period, based on the connection relationship at the first moment, determining each connected user terminal at the first moment as a user node, and determining the second edge feature information between each user node and the cell node; determining the initial feature information of each cell node based on the first network status information of each cell at the first moment, and determining the initial feature information of each user node based on the second network status information of each connected user terminal at the first moment; determining a first cell user graph slice at the first moment based on the cell nodes, the user nodes, the first edge feature information, the second edge feature information, the initial feature information of each cell node and the initial feature information of each user node; determining the cell user graph based on the first cell user graph slices at each moment in the first time period.

[0008] In some embodiments, the first power prediction model includes a spatial fusion layer, a temporal fusion layer and a linear mapping layer; the inputting the first cell user map into the first power prediction model to obtain the transmit power prediction result of each of the cells includes: sending the first cell user map to the spatial fusion layer to obtain a spatial fusion feature set of each of the cell nodes output by the spatial fusion layer; wherein each of the spatial fusion feature sets includes the spatial fusion feature information of the corresponding cell node at each moment; the spatial fusion layer is used to fuse the initial feature information of each of the cell nodes with the initial feature information of its neighboring nodes based on the first cell user map; the spatial fusion feature set of each of the cell nodes is input into the temporal fusion layer to obtain the spatiotemporal fusion feature information of each of the cell nodes output by the temporal fusion layer; the temporal fusion layer is used to fuse the spatial fusion feature information of each of the cell nodes at each moment in the first time period based on the spatial fusion feature set of each of the cell nodes; the spatiotemporal fusion feature information of each of the cell nodes is input into the linear mapping layer to obtain the transmit power prediction result of each of the cells output by the linear mapping layer.

[0009] In some embodiments, determining the transmit power prediction result of each of the cells according to the cell network status data set includes: constructing a plurality of second cell user graphs according to the divided plurality of cell sub-graphs, the first network status information, the connection relationship information, and the second network status information; wherein the plurality of cell sub-graphs are divided based on the neighboring area relationship information of the respective cells, and the plurality of second cell user graphs are connected to the plurality of cell sub-graphs. Figure 1 One corresponds to another; each of the second cell user graphs is an information expression of the network status relationship at each moment between the cells in the corresponding cell subgraph, and between the cells in the corresponding cell subgraph and the connected user terminals; the multiple second cell user graphs are input into the second power prediction model to obtain the transmission power prediction result of each cell; wherein the second power prediction model is trained based on the second cell user graph samples and the cell transmission power labels corresponding to the second cell user graph samples.

[0010] As an example, the multiple cell sub-graphs are divided in the following manner: constructing a cell topology graph based on the neighboring cell relationship of each cell; and segmenting the cell topology graph based on a spectral clustering algorithm to obtain each cell sub-graph.

[0011] In a second aspect, the present invention provides a control device for cell transmission power, comprising: an acquisition module, used to acquire a cell network status data set of a first time period; wherein the cell network status data set includes neighbor relationship information between each cell, first network status information of each cell at each moment in the first time period and connection relationship information with a user terminal, and second network status information of each connected user terminal at each moment; a prediction module, used to determine a transmission power prediction result of each of the cells based on the cell network status data set, so as to control the transmission power of each of the cells.

[0012] In a third aspect, an embodiment of the present invention provides an electronic device, including a processor and a memory storing a computer program, wherein when the processor executes the program, the method for controlling the cell transmission power described in the first aspect is implemented.

[0013] In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method for controlling cell transmission power as described in the first aspect.

[0014] In a fifth aspect, an embodiment of the present invention provides a computer program product, including a computer program, and when the computer program is executed by a processor, the method for controlling the cell transmission power described in the first aspect is implemented.

[0015] The cell transmission power control method, device, electronic device and storage medium provided in the embodiment of the present invention obtain a cell network status data set of the first time period; wherein the cell network status data set includes the neighboring relationship information between each cell, the first network status information of each cell at each moment in the first time period and the connection relationship information between the cell and the user terminal, and the second network status information of each connected user terminal at each moment; according to the cell network status data set, the transmission power prediction result of each cell is determined to control the transmission power of each cell. The present invention makes full use of the neighboring relationship between each cell, the connection relationship between each cell and the user terminal, and the network status information of each cell and each connected user terminal to predict the transmission power of each cell, and uses the predicted transmission power of each cell as the decision basis for controlling the cell transmission power, thereby improving the accuracy of power control, and further improving the signal coverage range and strength of the cell covered by the base station to meet the requirements, so as to improve the network coverage performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 A flowchart of a method for controlling cell transmit power provided in an embodiment of the present invention; Figure 2 A second flow chart of a method for controlling cell transmit power provided in an embodiment of the present invention; Figure 3 A third flow chart of a method for controlling cell transmit power provided in an embodiment of the present invention; Figure 4 is a schematic diagram of a user map of a first cell in an embodiment of the present invention; Figure 5 A fourth flowchart of a method for controlling cell transmit power provided in an embodiment of the present invention; Figure 6 A fifth flow chart of a method for controlling cell transmit power provided in an embodiment of the present invention; Figure 7 A schematic diagram of cell subgraph division provided in an embodiment of the present invention; Figure 8 A schematic diagram of the structure of a device for controlling cell transmission power provided by an embodiment of the present invention; Fig. 9 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Downlink power control in 5G systems is a key and fundamental issue. On the one hand, if the base station transmit power is low, path loss and shadow fading cannot be adequately compensated, resulting in weak coverage or even blind spots. On the other hand, if the base station signal transmit power is high, it will cause cross-area coverage and consume more energy, resulting in strong interference and high energy consumption. In the current network, the status of base station cells and users will change over time, so it is necessary to dynamically control the downlink power and adjust the base station transmit power to an appropriate value according to network changes.

[0020] In the related technology, the downlink power is mainly dynamically controlled according to the parameter setting on the base station cell side and the uplink feedback of the user. For example, the transmit power of the edge user-specific PDCCH is adaptively adjusted based on the PDCCH CCE aggregation level and the PUCCH feedback situation, and the PDSCH power spectrum density is adaptively adjusted based on the PDSCH scheduling MCS, bandwidth, and estimated user received signal strength. However, the power control effect is poor, and there is still a problem that the base station downlink power does not match the actual demand.

[0021] In order to solve the above problems, the present invention provides a method, device, electronic device and storage medium for controlling cell transmission power.

[0022] Figure 1 This is a flow chart of a method for controlling cell transmission power provided by an embodiment of the present invention. Figure 1 As shown, the method may include the following steps.

[0023] Step 101, obtaining a cell network status data set for a first time period; wherein the cell network status data set includes neighboring relationship information between each cell, first network status information of each cell at each moment in the first time period and connection relationship information between the cell and the user terminal, and second network status information of each connected user terminal at each moment.

[0024] In some embodiments, for the scenario of predicting the transmit power of each cell at the current moment, the first time period may be a continuous time period including the current moment and a period of time before, or a continuous time period in the past when the network state is close to the current one. For the scenario of predicting the transmit power of the cell in the future time period, the first time period may be a past time period close to the network state in the future time period.

[0025] In some embodiments of the present invention, the transmit power of all cells in the entire network may be predicted, where each cell is all cells in the entire network, or the transmit power of cells in a target area may be predicted, where each cell is all cells in the target area.

[0026] In some embodiments, the neighboring cell relationship information of each cell can be obtained from the network management data platform. The first network status information of each cell at each moment in the first time period may include the maximum number of users, downlink average PRB utilization, downlink data throughput, interference intensity, etc., which can also be obtained from the network management data platform. The connection relationship information between each cell and the user terminal at each moment in the first time period refers to the information of the user terminal connected to each cell in each time period. For example, if a cell is in an RRC (Radio Resource Control) connection state with a user terminal, it is confirmed that there is a connection relationship between the cell and the user terminal. Each connected user terminal at each moment refers to a user terminal that has accessed the service cell at each moment. The second network state information may include measurement information of each connected user terminal on the service cell, such as RSRP (Reference Signal Receiving Power), RSRQ (Reference Signal Receiving Quality), SINR (Signal to Interference plus Noise Ratio), etc., and may also include measurement information of each connected user terminal on a neighboring cell, such as RSRP, RSRQ, SINR, etc. The above data can be obtained from the network management data platform.

[0027] Step 102: Determine the transmit power prediction result of each cell according to the cell network status data set, so as to control the transmit power of each cell.

[0028] It can be understood that the cell network status data set includes the neighboring cell relationships between each cell in the first time period, the network status information of each cell and each connected user terminal at each moment in the first time period, and also includes data such as the mutual connection between each cell and each user terminal. These data all affect the cell transmission power, so the mutual correlation between the above data can be fully considered to predict the transmission power of each cell.

[0029] In some embodiments, the transmit power prediction result of each cell may include the cell transmit power prediction result of each cell at each moment in a second time period. The second time period may be a future time period in which the network status is close to that of the first time period, and the length of the second time period may be the same as that of the first time period, or may be less than the length of the first time period.

[0030] In other embodiments, if the application scenario is to predict the transmit power of the current cell, and the first time period is a continuous time period including the current moment and a previous period of time, the transmit power prediction result of each cell may include the transmit power prediction value of the corresponding cell at the current moment.

[0031] In some embodiments, the implementation process of determining the transmit power prediction result of each cell according to the cell network status data set may include: inputting the cell network status data set into a trained neural network model to obtain the transmit power prediction result of each cell. The neural network model may be trained based on a large number of cell network status data samples in the past period and their corresponding cell transmit power labels.

[0032] In other embodiments, based on the neighboring cell relationship of each cell, for each cell, the first network status information of the cell and the first network status information of its neighboring cells can be fused to obtain the first fused status information of the cell; the second network status information of the connected user terminal, the second network status information of all connected user terminals connected to the cell is fused to obtain the second fused status information of the cell; the first fused status information and the second fused status information of the cell are fused to obtain the third fused status information of the cell; the information at each moment in the third fused status information is fused to obtain the fourth fused status information; based on the mapping relationship between the preset fourth fused status information and the cell transmission power, the transmission power prediction result of the cell is determined. Among them, the fusion weight value in each step of the fusion process can be obtained based on the respective network status information, or each step of the fusion process can be realized by constructing a multi-layer attention network.

[0033] In some embodiments, the transmission power prediction result of each cell may be directly used for controlling the transmission power of the cell, or may be corrected and the transmission power of each cell may be controlled based on the corrected result.

[0034] As an implementation method, the cell transmission power control method of the present invention can be applied to the network management platform side, executed by the network management platform side, and sends a cell transmission power control instruction to the base station side based on the transmission power prediction results of each cell.

[0035] As another implementation method, the executor of the cell transmission power control method of the present invention can be any electronic device, which establishes a communication connection with the network management platform side, and sends the obtained transmission power prediction results of each cell to the network management platform side, so that the network management platform side can issue a cell transmission power control instruction to the base station side based on the cell transmission power prediction results.

[0036] As another implementation method, the executor of the cell transmission power control method of the present invention can be any electronic device, which establishes a communication connection with the base station side, and sends the obtained transmission power prediction results of each cell to the corresponding base station, which is then sent to the network management platform side, so that the base station side controls the transmission power of the cell based on the cell transmission power prediction results.

[0037] The cell transmission power control method provided by the embodiment of the present invention obtains a cell network status data set of the first time period; wherein the cell network status data set includes the neighboring relationship information between each cell, the first network status information of each cell at each moment in the first time period and the connection relationship information between the cell and the user terminal, and the second network status information of each connected user terminal at each moment; according to the cell network status data set, the transmission power prediction result of each cell is determined to control the transmission power of each cell. The present invention makes full use of the neighboring relationship between each cell, the connection relationship between each cell and the user terminal, and the network status information of each cell and each connected user terminal to predict the transmission power of each cell, and uses the predicted transmission power of each cell as the decision basis for controlling the cell transmission power, thereby improving the accuracy of power control, and further improving the signal coverage range and strength of the cell covered by the base station to meet the requirements, so as to improve the network coverage performance.

[0038] Figure 2 The second flow chart of a method for controlling cell transmission power provided by an embodiment of the present invention. Figure 2 As shown, based on the above embodiments, Figure 1 The implementation process of step 102 in may include the following steps.

[0039] Step 201, construct a first cell user graph according to a cell network status data set; the first cell user graph is an information expression of the network status relationship between various cells and between various cells and various connected user terminals at each moment.

[0040] In some embodiments, the first cell user graph is graph data constructed based on the cell network status data set. The first cell user graph is a dynamic heterogeneous graph, including slices of the first cell user graph at various times in the first time period. Since the cell network status data set contains a large amount of relationship data, expressing information through the cell user graph can more intuitively express the relationship between nodes.

[0041] In some embodiments, for each moment, each cell can be regarded as a cell node, and each connected user terminal can be regarded as a user node. The neighboring cell relationship is expressed by the edges connecting the cell nodes, and the connection relationship between the cell nodes and the user nodes is expressed by the edges connecting the cell nodes and the user terminals. The first network status information of each cell is used as the first network status information of the corresponding cell node, and the second status information of each connected user terminal is used as the second network status information of the corresponding user node.

[0042] As an example, the first cell user graph G={G t} t=1，T1 , where G t G represents the first cell user graph slice at the tth time in T1, where T1 is the first time period. t The network state information includes multiple cell nodes, multiple user nodes, first network state information of each cell node, second network state information of each user node, edges between cell nodes, and edges between cell nodes and user nodes.

[0043] Step 202: input the first cell user map into the first power prediction model to obtain a transmit power prediction result for each cell; wherein the first power prediction model is trained based on the first cell user map sample and the cell transmit power label corresponding to the first cell user map sample.

[0044] In some embodiments, the first power prediction model may be a graph neural network model that has learned the ability to predict the transmit power of a cell based on a user graph of the first cell.

[0045] In some embodiments, the first cell user map sample can be constructed based on a large number of cell network status data sets in historical time periods. For example, the cell network status data sets of multiple time periods in the past and the transmission power labels of each cell corresponding to each time period are obtained, and the duration of each time period is consistent with the duration of the first time period. The first cell user map is constructed based on the cell network status data set of each time period to obtain the first cell user map sample; the first cell user map sample and its corresponding cell transmission power label value are input into the initial first power prediction model for training to obtain the trained first power prediction model.

[0046] It should be noted that the cell transmission power label corresponding to the first cell user map sample can be determined based on actual needs. For example, if it is necessary to predict the transmission power of each cell at the current moment based on the cell network status data set of the continuous time period consisting of the current moment and the previous period, the transmission power setting value at the last moment in each historical period can be used as the corresponding cell transmission power label value. For another example, if the transmission power of each cell in the future time period is predicted based on the cell network status data set of the past time period, the cell transmission power setting value at each moment in each historical period can be used as the corresponding cell transmission power label value during the training process.

[0047] According to the method for controlling cell transmission power according to an embodiment of the present invention, a first cell user graph is constructed based on a cell network status data set; the first cell user graph is an information expression of the network status relationship between each cell and between each cell and each connected user terminal at each moment; the first cell user graph is input into the first power prediction model to obtain the transmission power prediction result of each cell; wherein the first power prediction model is trained based on the first cell user graph sample and the cell transmission power label corresponding to the first cell user graph sample. By constructing the first cell user graph, the network status relationship between cells and between cells and users is fully utilized, and the prediction of cell transmission power is realized through the first power prediction model, which is used as a decision basis for controlling cell transmission power, thereby improving the accuracy of power control.

[0048] Figure 3 The third flow chart of the method for controlling cell transmission power provided by the embodiment of the present invention. Figure 3 As shown, according to the cell network status data set, the implementation process of constructing the first cell user graph may include the following steps.

[0049] Step 301: determine each cell as a cell node, and determine first edge feature information between the cell nodes based on the neighboring cell relationship information between the cells.

[0050] That is to say, each cell is regarded as a node in the first cell user graph, and the neighboring cell relationship between cells is regarded as an edge. If two cells are neighboring cells, the edge weight between the two cell nodes is 1. If there is no neighboring cell relationship between the two cells, the edge weight between the two cell nodes is 0. The first edge feature information between each cell node is the edge weight.

[0051] Step 302: for the first moment in the first time period, based on the connection relationship information at the first moment, each connected user terminal at the first moment is determined as a user node, and second edge feature information between each user node and each cell node is determined.

[0052] In some embodiments, the connected user terminal at the first moment can be determined based on the connection relationship information between each cell and the user terminal at the first moment, and each connected user terminal at the first moment is also used as a node in the first cell user graph, that is, a user node. The connection relationship between each cell and the user terminal is used as the edge between the cell node and the user node. If the connected user terminal A and cell B are in a connected state at the first moment, the edge weight between the user node corresponding to the connected user terminal A at the first moment and the cell node corresponding to cell B is set to 1, otherwise it is set to 0. Among them, the second edge feature information between each user node and each cell node is the edge weight between each user node and each cell node.

[0053] Step 303, based on the first network state information of each cell at the first moment, determine the initial characteristic information of each cell node, and based on the second signal state information of each connected user terminal at the first moment, determine the initial characteristic information of each user node.

[0054] In some embodiments, for each cell, initial characteristic information of a cell node corresponding to the cell may be determined based on first network state information of the cell at a first moment.

[0055] As an example, the initial characteristic information of a cell node can be defined as dimensional column vector, for each cell, the first network state information of the cell at the first moment is encoded and converted to obtain the initial characteristic information of the cell node corresponding to the cell. For example, if the first network state information includes the maximum number of users, the average downlink PRB (Physical Resource Block) utilization, the downlink data throughput, etc., the above information can be spliced ​​and the spliced ​​data can be encoded and converted to obtain the initial characteristic information of the corresponding cell node.

[0056] Step 304: determine a first cell user graph slice at a first moment based on each cell node, each user node, the first edge feature information, the second edge feature information, the initial feature information of each cell node, and the initial feature information of each user node.

[0057] In some embodiments, each cell node, first edge feature information, multiple user nodes determined at the first moment, second edge feature information, initial feature information of each cell node and initial feature information of each user node are combined to form a first cell user graph slice at the first moment.

[0058] Step 305: Determine a first cell user graph based on the first cell user graph slices at each time point in the first time period.

[0059] In some embodiments, the first cell user map includes first cell user map slices at each moment in the first time period, so the first cell user map slices at each moment in the first time period are constructed in sequence according to steps 302 to 304 to finally obtain the first cell user map for the first time period.

[0060] like Figure 4 As shown, the first cell user graph can be a dynamic heterogeneous graph, which includes slices of the first cell user graph at each time in the first time period, and each first cell user graph slice includes cell nodes, user nodes, initial feature information of cell nodes, initial feature information of user nodes, edges between cell nodes, and edges between cell nodes and user nodes.

[0061] According to the cell transmission power control method of an embodiment of the present invention, by sequentially constructing first cell user map slices for each time period within the first time period, a first cell user map for the first time period is obtained, so as to intuitively express the network status relationship between cells and between cells and user terminals, which helps to make full use of this information when predicting cell transmission power, so as to improve the accuracy of cell transmission power prediction, thereby further improving the effect of cell power control.

[0062] Figure 5 FIG4 is a flow chart of a method for controlling cell transmission power provided in an embodiment of the present invention. Figure 5 As shown, based on the above embodiment, the implementation process of inputting the first cell user map into the first power prediction model and obtaining the transmit power prediction result of each cell may include the following steps.

[0063] Step 501, sending the first cell user graph to the spatial fusion layer, obtaining the spatial fusion feature set of each cell node output by the spatial fusion layer; wherein each spatial fusion feature set includes the spatial fusion feature information of the corresponding cell node at each moment; the spatial fusion layer is used to fuse the initial feature information of each cell node with the initial feature information of its neighboring nodes based on the first cell user graph.

[0064] In some embodiments, the first power model may be a hierarchical attention network model, including a spatial fusion layer, a temporal fusion layer, and a linear mapping layer. The spatial fusion layer is used to determine the neighbor nodes of each cell node at each moment based on the first cell user graph, wherein the neighbor nodes may include neighboring cell nodes of the corresponding cell node, and may also include user nodes connected to the corresponding cell node. The spatial fusion layer may also fuse the initial feature information of each cell node, the initial feature information of the adjacent cells, and the initial feature information of the connected user nodes for each cell node, and obtain the spatial fusion feature combination of the cell.

[0065] In some embodiments, the spatial fusion layer may include: for the tth time, for the i-th cell node Ai, according to the first cell user graph, determine the neighbor cell node set and the neighbor user node combination of Ai; based on the initial feature information of Ai and the initial feature information of each neighbor cell node, determine the first fusion weight coefficient of each neighbor cell node to Ai; based on the initial feature information of Ai and the initial feature information of each neighbor user node, determine the second fusion weight coefficient of each neighbor user node to Ai; based on the first fusion weight coefficient and the initial feature information of each neighbor cell, determine the fusion feature of Ai's neighbor cell node; based on the second fusion weight coefficient and the initial feature information of each neighbor user, determine the fusion feature of Ai's neighbor user node; the initial feature information of Ai, the fusion feature of Ai's neighbor cell node and the fusion feature of Ai's neighbor user node are fused again to obtain the spatial fusion feature information of Ai at the tth time. Wherein, the first fusion weight coefficient and the second fusion weight coefficient both represent the contribution degree of the neighbor node to the feature of Ai node.

[0066] As an example, based on the initial feature information of Ai and the initial feature information of each neighboring cell node, the process of determining the first fusion weight coefficient of each neighboring cell node to Ai can be implemented as follows: formula (1).

[0067] (1); in, is the first fusion weight coefficient of the j-th neighboring cell node Aj to the i-th cell node Ai at the t-th moment; represents the set of neighboring cell nodes of cell node Ai at time t; is the initial characteristic information of the cell node Ai at the tth moment; is the activation function; is the attention factor, which can be a trainable column vector with a dimension of 2 ; is a trainable weight matrix with dimension , is the dimension of the initial feature information of the Ai node; Indicates that column vectors are concatenated row by row.

[0068] Based on the first fusion weight coefficient and the initial feature information of each neighbor cell, the implementation method of determining the fusion feature of the Ai neighbor cell node can be as shown in formula (2).

[0069] (2); in, is the fusion feature of Ai’s neighboring cell nodes.

[0070] Based on the initial feature information of Ai and the initial feature information of each neighboring user node, the implementation process of determining the second fusion weight coefficient of each neighboring user node to Ai is shown in the following formula (3).

[0071] (3); in, is the second fusion weight coefficient of the j-th user cell node Bj to the i-th cell node Ai at the t-th moment; represents the set of neighboring user nodes of cell node Ai at time t; is the initial characteristic information of the cell node Ai at the tth moment; is the activation function; is the attention factor, which can be a trainable column vector with a dimension of 2 ; and is the trainable weight matrix, The dimension is , The dimension is , is the dimension of the initial feature information of the user node.

[0072] Based on the second fusion weight coefficient and the initial feature information of each neighbor user, the implementation process of determining the fusion feature of the Ai neighbor user node can be shown in the following formula (4).

[0073] (4); in, Fusion characteristics of Ai neighbor cell users.

[0074] Finally, the initial feature information of Ai, the fused features of Ai's neighboring cell nodes, and the fused features of Ai's neighboring user nodes are fused again to obtain the spatial fused feature information of Ai at the tth moment, as shown in the following formula (5).

[0075] (5); Among them, , and They represent the corresponding weight coefficient matrices, which can be obtained through training; The dimension is , The dimension is , The dimension is , For one Column vector of dimension.

[0076] Step 502, input the spatial domain fusion feature set of each cell node into the time domain fusion layer to obtain the spatiotemporal fusion feature information of each cell node output by the time domain fusion layer; the time domain fusion layer is used to fuse the spatial domain fusion feature information of each cell node at each time in the first time period based on the spatial domain fusion feature set of each cell node.

[0077] In some embodiments, the time domain fusion layer can fuse the spatial domain fusion feature information of each cell node at each time in the first time period based on the scaled dot product attention mechanism.

[0078] In some embodiments, for The spatial fusion feature set of the node is { }, the set can be combined into a matrix , The dimension is .calculate , , ,in , and The dimensions of , is a dimension of The purpose of the mask matrix is ​​to make the cell nodes at time t aggregate only the spatial aggregation features from time t and before when aggregating in the time domain. Therefore Elements in It is defined as the following formula (6).

[0079] (6); Based on the following formula (7), the final spatiotemporal fusion feature information of node Ai can be obtained. It should be noted that the final spatiotemporal fusion feature information of Ai can be a set of spatiotemporal fusion feature information at each moment, or it can be the spatiotemporal fusion feature information at the last moment in the first period, which can be set based on actual needs.

[0080] (7); in, is the spatiotemporal fusion feature information set of node Ai; is a dimension of The matrix of No. Row vector That is The spatiotemporal fusion feature information at the tth moment.

[0081] Step 503: input the spatiotemporal fusion feature information of each cell node into the linear mapping layer to obtain the transmit power prediction result of each cell output by the linear mapping layer.

[0082] According to the method for controlling cell transmission power according to an embodiment of the present invention, the spatial domain information and the time domain information contained in the first cell user map are deeply fused through a first power prediction model including a spatial domain fusion layer, a time domain fusion layer and a linear mapping layer, so as to perform deep feature extraction on the first cell user map, thereby further improving the accuracy of the transmission power prediction of each cell, and further improving the accuracy and effect of the cell power control.

[0083] In order to further improve the computing efficiency, the present invention also provides another embodiment.

[0084] Figure 6 FIG5 is a flow chart of a method for controlling cell transmission power provided in an embodiment of the present invention. Figure 6 As shown, based on the above embodiment, the implementation process of determining the transmit power prediction result of each cell according to the cell network status data set may include the following steps.

[0085] Step 601, constructing a plurality of second cell user graphs according to the divided plurality of cell sub-graphs, the first network state information, the connection relationship information, and the second network state information; wherein the plurality of cell sub-graphs are divided based on the neighboring area relationship information of each cell, and the plurality of second cell user graphs are connected to the plurality of cell sub-graphs. Figure 1 One to one correspondence; each second cell user map is an information expression of the network status relationship at each moment between the cells in the corresponding cell submap, and between the cells in the corresponding cell submap and the connected user terminals.

[0086] It can be understood that the number of cells and user terminals in the entire network is huge, and the amount of data involved in the first cell user graph constructed based on the cell network status data set of the first period is very large, and the computing resources required for model prediction and training are large and time-consuming. In order to improve computing efficiency, the embodiment of the present invention constructs a second cell user graph based on multiple divided cell subgraphs, so that each second cell user graph obtained only involves the information of the cell nodes in the corresponding cell subgraph, and the information of the user nodes connected to the cell nodes. When constructing the second cell user graph and predicting the cell transmission power, distributed parallel computing can be used, thereby greatly improving computing efficiency.

[0087] In some embodiments, the cell subgraph is divided based on the neighboring relationship information between each cell node. Each cell subgraph may include multiple closely related cells, while the relationship between cells in different subgraphs is relatively sparse, so that cells in the same cell subgraph can efficiently exchange information and improve computing efficiency.

[0088] In some embodiments, each cell subgraph may include multiple cell nodes and edges between cell nodes, and the neighboring relationship between cells is used as an edge. If two cells have a neighboring relationship, the edge weight between the corresponding cell nodes is set to 1, otherwise it is set to 0. In other words, the cell subgraph may include first edge feature information between cell nodes.

[0089] In some embodiments, the process of constructing multiple second cell user graphs is consistent with the process of constructing the first cell user subgraph based on the divided multiple cell subgraphs, the first network status information, the connection relationship information, and the second network status information, except that the second cell user graph is constructed based on the cell nodes contained in each cell subgraph, not based on all cell nodes. As an example, when constructing the second cell user graph corresponding to cell subgraph A, the cell nodes in the second cell user graph and the first edge feature information between the cell nodes are consistent with those in cell subgraph A. For each moment, based on the cell nodes in cell subgraph A, the connected user terminals at that moment are obtained and determined as user nodes, and the second edge feature information between user nodes is determined; based on the first network status information at that moment, the initial feature information of the cell nodes in cell subgraph A is determined, based on the second network status information at that moment, the initial feature information of the user nodes is obtained, and based on the above information, the second cell user graph slice in the second cell user graph corresponding to cell subgraph A is determined.

[0090] In some embodiments, the cell sub-map may be divided in the following manner: constructing a cell topology map based on the neighboring cell relationship of each cell; and performing segmentation processing on the cell topology map based on a spectral clustering algorithm to obtain each cell sub-map.

[0091] Specifically, when constructing a cell topology graph, the cells are taken as nodes and the neighboring relationships between cells are taken as edges. That is, if there is a neighboring relationship between two cells, the corresponding edge weight is set to 1, otherwise it is set to 0, and an undirected weighted graph is obtained. Assume that there are nodes, the corresponding adjacency matrix is ​​set to , then the degree matrix can be calculated based on the adjacency matrix , is a diagonal matrix with the elements on the diagonal , so the corresponding Laplacian matrix .

[0092] According to the spectral clustering algorithm, neighboring cells that are closely connected, may overlap and influence each other are divided into the same subgraph. First, a representation matrix is ​​calculated , and then calculate of The eigenvalues ​​and corresponding eigenvectors will be The matrix composed of eigenvectors is normalized by row to obtain the characteristic matrix ,Will Each row of is taken as a sample, a total of Then use k-means and other clustering algorithms to cluster these samples. The samples are clustered and the clustering dimension is , and finally divide the cell topology into A sub-plot of a cell. Figure 7 A schematic diagram of dividing a cell topology map into multiple cell sub-maps.

[0093] Therefore, by segmenting the cell topology map in this step, the cell topology map composed of cells in the entire network is divided into different cell sub-maps, and subsequent processing is performed on each cell sub-map, thereby reducing information interaction costs and improving data processing efficiency.

[0094] Step 602: Input multiple second cell user graphs into a second power prediction model to obtain a transmit power prediction result for each cell; wherein the second power prediction model is trained based on second cell user graph samples and cell transmit power labels corresponding to the second cell user graph samples.

[0095] It should be noted that the difference between the second power prediction model and the first power prediction model is that the second power prediction model is trained based on the second cell user graph sample and its corresponding cell transmission power label, while the first power prediction model is trained based on the first cell user graph sample. Among them, the second cell user graph sample is constructed based on multiple divided cell sub-graphs and cell network status data sets of multiple historical time periods.

[0096] In some embodiments, the second power prediction model may also include a spatial fusion layer, a temporal fusion layer, and a linear mapping layer, and the implementation process and principle are consistent with the first power prediction model, except that the second cell user graph contains less data, and the second power prediction model needs to make predictions based on the data in the second cell user graph. The prediction of cell transmit power based on different second cell user graphs can be completed through distributed parallel computing.

[0097] According to the method for controlling cell transmission power of an embodiment of the present invention, multiple second cell user graphs are constructed based on multiple divided cell sub-graphs, first network status information, connection relationship information, and second network status information, and the multiple second cell user graphs are input into the second power prediction model to obtain the transmission power prediction result of each of the cells. By dividing the cells of the entire network into multiple cell sub-graphs, the present invention can use distributed parallel computing when constructing the cell user graph and predicting the cell transmission power, thereby greatly improving the computing efficiency.

[0098] The following describes a device for controlling cell transmit power provided by an embodiment of the present invention. The device for controlling cell transmit power described below and the method for controlling cell transmit power described above may refer to each other.

[0099] Figure 8 A schematic diagram of the structure of a cell transmission power control device provided in an embodiment of the present invention. Figure 8 As shown, the apparatus may include an acquisition module 810 and a prediction module 820 .

[0100] The acquisition module 810 is used to acquire a cell network status data set for a first period of time; wherein the cell network status data set includes neighboring cell relationship information between each cell, first network status information of each cell at each moment in the first period of time and connection relationship information between each cell and a user terminal, and second network status information of each connected user terminal at each moment; The prediction module 820 is used to determine the transmission power prediction result of each cell according to the cell network status data set, so as to control the transmission power of each cell.

[0101] In some embodiments, the prediction module 820 includes a construction unit 821 and a prediction unit 822; wherein the construction unit 821 is used to construct a first cell user map based on the cell network status data set; the first cell user map is an information expression of the network status relationship between each cell and between each cell and each connected user terminal at each moment; the prediction unit 822 is used to input the first cell user map into a first power prediction model to obtain a transmission power prediction result for each cell; wherein the first power prediction model is trained based on the first cell user map sample and the cell transmission power label corresponding to the first cell user map sample.

[0102] In some embodiments, the construction unit 821 is specifically used to: determine each cell as a cell node, and determine the first edge feature information between each cell node based on the neighbor relationship information between each cell; for the first moment in the first time period, based on the connection relationship information at the first moment, determine each connected user terminal at the first moment as a user node, and determine the second edge feature information between each user node and each cell node; determine the initial feature information of each cell node based on the first network status information of each cell at the first moment, and determine the initial feature information of each user node based on the second signal status information of each connected user terminal at the first moment; determine the first cell user graph slice at the first moment based on each cell node, each user node, the first edge feature information, the second edge feature information, the initial feature information of each cell node and the initial feature information of each user node; determine the first cell user graph based on the first cell user graph slice at each moment in the first time period.

[0103] In some embodiments, the first power prediction model includes a spatial fusion layer, a time domain fusion layer and a linear mapping layer; the prediction unit 822 is also used to: send the first cell user map to the spatial fusion layer, and obtain the spatial fusion feature set of each cell node output by the spatial fusion layer; wherein each spatial fusion feature set includes the spatial fusion feature information of the corresponding cell node at each time; the spatial fusion layer is used to fuse the initial feature information of each cell node with the initial feature information of its neighboring nodes based on the first cell user map; input the spatial fusion feature set of each cell node into the time domain fusion layer, and obtain the spatiotemporal fusion feature information of each cell node output by the time domain fusion layer; the time domain fusion layer is used to fuse the spatial fusion feature information of each cell node at each time in the first time period based on the spatial fusion feature set of each cell node; input the spatiotemporal fusion feature information of each cell node into the linear mapping layer, and obtain the transmission power prediction result of each cell output by the linear mapping layer.

[0104] In some embodiments, the construction unit 821 is further used to: construct multiple second cell user graphs according to the divided multiple cell sub-graphs, the first network state information, the connection relationship information, and the second network state information; wherein the multiple cell sub-graphs are divided based on the neighboring area relationship information of each cell, and the multiple second cell user graphs are divided based on the neighboring area relationship information of each cell. Figure 1One to one correspondence; each second cell user graph is an information expression of the network status relationship at each moment between each cell in the corresponding cell subgraph, and between each cell in the corresponding cell subgraph and each connected user terminal; the prediction unit 822 is also used to: input multiple second cell user graphs into the second power prediction model to obtain the transmission power prediction result of each cell; wherein the second power prediction model is trained based on the second cell user graph sample and the cell transmission power label corresponding to the second cell user graph sample.

[0105] In some embodiments, the apparatus further includes a partitioning module 830 for constructing a cell topology map based on the neighboring cell relationships of each cell; and performing segmentation processing on the cell topology map based on a spectral clustering algorithm to obtain each cell sub-map.

[0106] It should be noted that the explanations in the above embodiment of the method for controlling cell transmit power are also applicable to the device for controlling cell transmit power in the embodiment of the present invention, and will not be described in detail here.

[0107] Fig. 9 An example of a physical structure diagram of an electronic device is shown in FIG. Fig. 9 As shown, the electronic device may include: a processor 910, a communication interface 920, a memory 930 and a communication bus 940, wherein the processor 910, the communication interface 920 and the memory 930 communicate with each other through the communication bus 940. The processor 910 may call a computer program in the memory 930 to execute the steps of the method for controlling the cell transmission power.

[0108] For example, the method includes: obtaining a cell network status data set for a first time period; wherein the cell network status data set includes neighbor relationship information between each cell, first network status information of each cell at each moment in the first time period and connection relationship information between the cell and the user terminal, and second network status information of each connected user terminal at each moment; based on the cell network status data set, determining a transmission power prediction result for each cell for controlling the transmission power of each cell.

[0109] In addition, the logic instructions in the above-mentioned memory 930 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0110] On the other hand, an embodiment of the present invention further provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the steps of the cell transmission power control method provided in the above embodiments.

[0111] The method comprises: obtaining a cell network status data set for a first time period; wherein the cell network status data set comprises neighboring cell relationship information between each cell, first network status information of each cell at each moment in the first time period and connection relationship information between each cell and a user terminal, and second network status information of each connected user terminal at each moment; and determining a transmission power prediction result of each cell according to the cell network status data set, so as to control the transmission power of each cell.

[0112] On the other hand, an embodiment of the present invention further provides a processor-readable storage medium, wherein the processor-readable storage medium stores a computer program, and the computer program is used to enable the processor to execute the steps of the cell transmission power control method provided in the above embodiments.

[0113] For example, the method includes: obtaining a cell network status data set for a first time period; wherein the cell network status data set includes neighbor relationship information between each cell, first network status information of each cell at each moment in the first time period and connection relationship information between the cell and the user terminal, and second network status information of each connected user terminal at each moment; based on the cell network status data set, determining a transmission power prediction result for each cell for controlling the transmission power of each cell.

[0114] The processor-readable storage medium can be any available medium or data storage device that can be accessed by the processor, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc.), optical storage (such as CD, DVD, BD, HVD, etc.), and semiconductor storage (such as ROM, EPROM, EEPROM, non-volatile memory (NANDFLASH), solid-state drive (SSD)), etc.

[0115] The device embodiments described above are merely illustrative, 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 on 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. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0116] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for controlling cell transmission power, characterized in that: include: Acquire a cell network status data set for a first time period; wherein the cell network status data set includes neighboring relationship information between each cell, first network status information of each cell at each moment in the first time period and connection relationship information between the cell and the user terminal, and second network status information of each connected user terminal at each moment; According to the cell network status data set, a transmit power prediction result of each of the cells is determined to control the transmit power of each of the cells.

2. The method for controlling cell transmission power according to claim 1, characterized in that: The step of determining the transmit power prediction result of each cell according to the cell network status data set includes: Constructing a first cell user graph according to the cell network status data set; the first cell user graph is an information expression of the network status relationship between the cells and between the cells and the connected user terminals at each moment; The first cell user map is input into a first power prediction model to obtain a transmit power prediction result for each of the cells; wherein the first power prediction model is trained based on the first cell user map sample and the cell transmit power label corresponding to the first cell user map sample.

3. The method for controlling cell transmission power according to claim 2, characterized in that: The step of constructing a first cell user graph according to the cell network status data set includes: Determine each of the cells as a cell node, and determine first edge feature information between the cell nodes based on the neighboring cell relationship information between the cells; For a first moment in the first time period, based on the connection relationship information at the first moment, each connected user terminal at the first moment is determined as a user node, and second edge feature information between each user node and each cell node is determined; Determine the initial characteristic information of each of the cell nodes based on the first network state information of each cell at the first moment, and determine the initial characteristic information of each of the user nodes based on the second signal state information of each connected user terminal at the first moment; Determine a first cell user graph slice at the first moment based on the respective cell nodes, the respective user nodes, the first edge feature information, the second edge feature information, the initial feature information of each of the cell nodes, and the initial feature information of each of the user nodes; The first cell user graph is determined based on the first cell user graph slices at each time point in the first time period.

4. The method for controlling cell transmission power according to claim 3, characterized in that: The first power prediction model includes a spatial fusion layer, a time domain fusion layer, and a linear mapping layer; the inputting the first cell user graph into the first power prediction model to obtain a transmit power prediction result of each cell includes: The first cell user graph is sent to the spatial fusion layer to obtain a spatial fusion feature set of each cell node output by the spatial fusion layer; wherein each of the spatial fusion feature sets includes spatial fusion feature information of the corresponding cell node at each time; the spatial fusion layer is used to fuse the initial feature information of each cell node with the initial feature information of its neighboring nodes based on the first cell user graph; Inputting the spatial domain fusion feature set of each cell node into the time domain fusion layer, obtaining the spatiotemporal fusion feature information of each cell node output by the time domain fusion layer; the time domain fusion layer is used to fuse the spatial domain fusion feature information of each cell node at each time in the first time period based on the spatial domain fusion feature set of each cell node; The spatiotemporal fusion feature information of each of the cell nodes is input into the linear mapping layer to obtain the transmission power prediction result of each of the cells output by the linear mapping layer.

5. The method for controlling cell transmission power according to claim 1, characterized in that: The step of determining the transmit power prediction result of each cell according to the cell network status data set includes: Constructing a plurality of second cell user graphs according to the divided plurality of cell subgraphs, the first network state information, the connection relationship information, and the second network state information; wherein the plurality of cell subgraphs are divided based on the neighboring cell relationship information of the respective cells, and the plurality of second cell user graphs correspond one-to-one to the plurality of cell subgraphs; each of the second cell user graphs is an information expression of the network state relationship at each moment between the respective cells in the corresponding cell subgraph, and between the respective cells in the corresponding cell subgraph and the respective connected user terminals; The multiple second cell user graphs are input into a second power prediction model to obtain a transmission power prediction result for each of the cells; wherein the second power prediction model is trained based on the second cell user graph samples and the cell transmission power labels corresponding to the second cell user graph samples.

6. The method for controlling cell transmission power according to claim 5, characterized in that: The multiple cell subgraphs are divided in the following manner: Based on the neighboring cell relationships of the cells, construct a cell topology map; Based on the spectral clustering algorithm, the cell topology map is segmented to obtain the various cell sub-maps.

7. A cell transmission power control device, characterized in that: include: An acquisition module, configured to acquire a cell network status data set for a first time period; wherein the cell network status data set includes neighboring cell relationship information between each cell, first network status information of each cell at each moment in the first time period and connection relationship information with a user terminal, and second network status information of each connected user terminal at each moment; The prediction module is used to determine the transmission power prediction result of each cell according to the cell network status data set, so as to control the transmission power of each cell.

8. An electronic device comprising a processor and a memory storing a computer program, characterized in that: When the processor executes the computer program, the method for controlling cell transmission power according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for controlling cell transmission power according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for controlling cell transmission power according to any one of claims 1 to 6 is implemented.

Images