Method and system for adjusting network resources

Abstract

The embodiment of the invention provides a network resource adjusting method and system, and the method comprises the steps: obtaining network parameters of all cells in a CCFD network based on user equipment, the network parameters comprise cell energy consumption efficiency and at least one of the following channel relationships: a channel relationship between different base stations, and a channel relationship between user equipment on an interference applying side and user equipment on an interfered side; wherein the user equipment comprises the user equipment on the disturbing side and the user equipment on the disturbed side; and determining a base station resource adjustment strategy corresponding to the network parameter based on the network parameter, so that the transmitting power of each cell is adjusted according to the base station resource adjustment strategy. According to the invention, the problem that the adjustment effect of the communication quality of the user equipment is not ideal in the prior art is solved, and the effect of improving the communication quality of the user equipment is further achieved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of communications, and in particular, to a method and system for adjusting network resources. Background Art

[0002] With the rapid development of mobile Internet and artificial intelligence technology, smart wireless devices such as smart phones, augmented reality (AR), virtual reality (VR) and other smart applications such as telemedicine, Industry 4.0, and autonomous driving have entered a stage of explosive growth. The network provided by traditional wireless systems cannot meet the scenarios of different business types. The system network based on duplex technology can effectively allocate resources and manage interference according to business changes, thereby achieving multi-dimensional scalability adaptation such as network coverage, business scheduling, and system anti-interference.

[0003] However, duplex technology involves a variety of interference factors. Moreover, the transmission power of the base station is closely related to the influence of interference factors. Low power leads to low mutual interference, but naturally poor communication quality. High power will aggravate mutual interference, greatly affecting the communication quality of user equipment connected to the base station.

[0004] In the related art, the dimensions of interference factors considered are relatively single, and the effect of adjusting the communication quality of the user equipment is not ideal. Therefore, it is urgent to improve the adjustment method of the communication quality of the user equipment connected to the base station. Summary of the invention

[0005] The embodiments of the present invention provide a method and system for adjusting network resources, so as to at least solve the problem that the adjustment effect on the communication quality of user equipment in the related art is not ideal.

[0006] According to an embodiment of the present invention, a method for adjusting network resources is provided, comprising:

[0007] Acquiring network parameters of each cell in the CCFD network based on the user equipment, the network parameters including cell energy consumption efficiency and at least one of the following channel relationships: channel relationships between different base stations, channel relationships between disturbing-side user equipment and disturbed-side user equipment; wherein the user equipment includes the disturbing-side user equipment and the disturbed-side user equipment;

[0008] A base station resource adjustment strategy corresponding to the network parameter is determined based on the network parameter, so that the transmission power of each cell is adjusted according to the base station resource adjustment strategy.

[0009] According to another embodiment of the present invention, a network resource adjustment system is provided, comprising:

[0010] An environment mapping model is used to obtain network parameters of each cell in the CCFD network based on user equipment, wherein the network parameters include cell energy consumption efficiency and at least one of the following channel relationships: channel relationships between different base stations, and channel relationships between disturbing-side user equipment and disturbed-side user equipment; wherein the user equipment includes the disturbing-side user equipment and the disturbed-side user equipment;

[0011] The DQN network model is used to determine the base station resource adjustment strategy corresponding to the network parameter based on the network parameter, so as to adjust the transmission power of each cell according to the base station resource adjustment strategy.

[0012] According to yet another embodiment of the present invention, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the steps of any one of the above method embodiments when run.

[0013] According to yet another embodiment of the present invention, there is provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0014] Through the present invention, since more dimensions of parameters can be collected according to different network parameters, the transmission power of the base station can be adjusted to adapt to different interference environments and communication requirements. By adjusting the transmission power, the interference level can be reduced, the communication quality can be improved, and the communication experience of the user equipment can be improved. This method realizes the dynamic adjustment of the communication quality of the user equipment, can flexibly allocate resources according to actual conditions, and effectively solves the problem of unsatisfactory adjustment effect on the communication quality of the user equipment in the related technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a schematic diagram of the relationship between the transmission channel and the interference channel between the user equipment and the base station in the CCFD network according to an embodiment of the present invention;

[0016] Figure 2 is a flow chart of a method for adjusting network resources according to an embodiment of the present invention;

[0017] Figure 3 is a schematic diagram of the information transmission relationship between user equipment, cells, and base stations in a CCFD network according to an embodiment of the present invention;

[0018] Figure 4 is a flow chart of a method for supplementing a channel relationship of a disturbed side user equipment by a disturbed side base station according to an embodiment of the present invention;

[0019] Figure 5is a schematic diagram of CLI measurement mapping of multiple user equipment according to an embodiment of the present invention;

[0020] Figure 6 is a flow chart of a method for obtaining preceding environmental variables, optimizing subsequent state environmental variables, incentive variables, and action variables according to an embodiment of the present invention;

[0021] Figure 7 is a flow chart of a method for obtaining a target value function according to an embodiment of the present invention;

[0022] Figure 8 is a flow chart of a method for obtaining an optimized value function according to an embodiment of the present invention;

[0023] Fig. 9 is a flow chart of a method for obtaining weights of an optimized network module according to an embodiment of the present invention;

[0024] Fig.10 Schematic diagram of the structure of an environment mapping model and a DQN network model according to an embodiment of the present invention. DETAILED DESCRIPTION

[0025] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings and in combination with the embodiments.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0027] Figure 1 is a schematic diagram of the relationship between the transmission channel and the interference channel between the user equipment and the base station in the CCFD network according to an embodiment of the present invention, such as Figure 1 As shown,

[0028] In the CCFD multi-cell networking scenario, there are a lot of cross-link interference CLI (Cross-Link Interference) and self-interference SI (Self-Interference) outside the inter-cell transmission channel between user equipment and base stations. CCFD cell stations use different resource allocation methods (time domain: different frame structures; frequency domain: different RB scheduling strategies), which increases the complexity of network channel management and interference management and reduces the system spectrum utilization. Effective resource allocation and interference optimization are the key technologies to improve the overall network anti-interference capability and system energy efficiency. Among them, the RB scheduling strategy, whose full name is Round-Robin scheduling strategy, is a common process scheduling algorithm. It is a simple and average round-robin scheduling strategy, also known as polling scheduling. In the RB scheduling strategy, a process is assigned a time slice, that is, a fixed period of time. After each process is executed in turn for a period of time, the time slice is used up, the process is suspended, and handed over to the next process waiting to be executed. This round-robin execution method gives each process a chance to be executed, thereby achieving fairness. The RB scheduling strategy is suitable for multi-tasking operating systems, in which multiple processes need to run simultaneously. It is characterized by simplicity, fairness, and cyclic execution.

[0029] In particular, interference management is a key technical problem in CCFD networking, where there is mutual interference in the uplink and downlink of the time domain between the aggressor and the victim. In this case, the interference information between base stations and serving users in the system can be fed back through the CLI (Cross-Link Interference) measurement report. Among them, CLI includes channel interference measurements between base stations and user equipment. Through inter-station interaction, it can synchronize the neighboring area time and frequency domain scheduling strategy for interference management, and perform power, time and frequency domain, and beam avoidance adjustments to complete collaborative scheduling. The aggressor and the victim use part of the resources in the downlink scheduling time slot to perform interference detection through characteristic data, and can use measurement methods such as SRS / CSI / DRMS.

[0030] Among them, beam avoidance is a mechanism in communication used to avoid interference between multiple communication devices (such as radar, satellite communication, etc.) on the same spectrum band. In wireless communication, spectrum resources are limited, so multiple devices are often required to share the same frequency band for communication. However, when multiple devices send signals on the same frequency band at the same time, mutual interference problems will arise. Beam avoidance is based on beamforming technology, which performs beamforming at the transmitting and receiving ends by using adjustable antenna arrays. Using beamforming technology, the transmitting end can focus the signal into a narrow beam so that it is transmitted only in a specific direction instead of propagating in all directions. The receiving end uses beamforming technology to enhance the signal received from a specific direction, thereby suppressing interference signals from other directions. By using beam avoidance, the overlapping use of spectrum resources can be reduced and the capacity and reliability of the communication system can be improved.

[0031] Among them, SRS (Signal-to-Noise Ratio) is the abbreviation of signal-to-noise ratio, which is used to represent the ratio of signal to noise. In a communication system, the signal represents the useful information transmitted, while the noise represents interference factors unrelated to the signal. A higher SRS value means a stronger signal relative to the noise, which usually indicates better signal quality.

[0032] CSI (Channel State Information) is the abbreviation of channel state information, which is used to describe the channel status in wireless communication systems. CSI usually includes information such as channel gain and channel phase, which can be used to optimize the performance of communication systems. By obtaining and using CSI, beamforming, interference management, power control and other technologies can be performed to improve the capacity and efficiency of communication systems.

[0033] DRMS (Delay Spread Root Mean Square) is the abbreviation of delay spread root mean square, which is used to describe the delay spread of the path in wireless communication. Delay spread refers to the influence of different paths caused by multipath propagation on the signal during transmission, which may lead to problems such as signal superposition and delay spread. By measuring the DRMS ​​value, the delay compensation mechanism of the system can be evaluated and adjusted to improve the reliability and performance of communication.

[0034] These measurement methods are mainly used to evaluate and optimize the performance of wireless communication systems. SRS is used to measure signal quality, CSI is used to understand channel conditions and optimize communication systems, and DRMS ​​is used to evaluate and adjust delay spread issues.

[0035] In this embodiment, a method for adjusting network resources is provided. Figure 2 is a flow chart of a method for adjusting network resources according to an embodiment of the present invention; Figure 2As shown, the process includes the following steps:

[0036] Step S201, based on the user equipment, network parameters of each cell in the CCFD network are obtained, where the network parameters include cell energy efficiency and at least one of the following channel relationships: channel relationships between different base stations, and channel relationships between disturbing-side user equipment and disturbed-side user equipment; wherein the user equipment includes disturbing-side user equipment and disturbed-side user equipment;

[0037] In an exemplary implementation, for example, user equipment A communicates with cell A of base station A to form transmission channel A, and user equipment B communicates with cell B of base station A to form transmission channel B. Under the influence of geographical factors, transmission channel A and transmission channel B may intersect. Therefore, when user equipment B sends a signal through transmission channel B, it may cause interference to user equipment A. At this time, for user equipment A, user equipment A is a user equipment on the disturbed side, and user equipment B is a user equipment on the disturbing side. Of course, when user equipment A sends a signal through transmission channel A, it may cause interference to user equipment B. At this time, for user equipment B, user equipment B is a user equipment on the disturbed side, and user equipment A is a user equipment on the disturbing side. The determination of the disturbing side user equipment and the disturbed side user equipment is based on the user equipment that proposes the adjustment requirement. The disturbed side user equipment pointed out in the embodiment of the present invention refers to the user equipment that proposes the adjustment requirement.

[0038] Step S202: determining a base station resource adjustment strategy corresponding to the network parameter based on the network parameter, so as to adjust the transmission power of each cell according to the base station resource adjustment strategy.

[0039] Through the above steps, since more dimensions of the collected parameters can be adjusted according to different network parameters, the transmit power of the base station can be adjusted to adapt to different interference environments and communication requirements. By adjusting the transmit power, the interference level can be reduced, the communication quality can be improved, and the communication experience of the user equipment can be improved. This method realizes the dynamic adjustment of the communication quality of the user equipment, can flexibly allocate resources according to actual conditions, and effectively solves the problem of unsatisfactory adjustment effect on the communication quality of the user equipment in the related technology.

[0040] In one implementation, obtaining network parameters of each cell in the CCFD network based on the user equipment includes:

[0041] Through the environmental mapping model, cross-link interference CLI characteristic data is inserted into the downlink scheduling time slot of each cell in the CCFD network, and the cross-link interference CLI characteristic data is transmitted to the user equipment using space division propagation, so that the user equipment can feedback the energy consumption efficiency and channel relationship of each cell in the CCFD network based on the cross-link interference CLI characteristic data.

[0042] In an exemplary embodiment, the cross-link interference CLI characteristic data is CLI RS (Cross-Link Interference Resource Set, cross-link interference reference signal), which includes SRS-RSRP (Reference signal received power, reference signal received power) and CLI-RSSI (Received signal strength indicator, reference signal strength indication). The cell where the interfering side user equipment communicates and the cell where the interfered side user equipment communicates both send CLI RS (cross-link interference reference signal). After space division propagation, the user equipment feeds back the CLI RS measurement report to the corresponding transmission channel cell in the uplink scheduling time slot.

[0043] In one embodiment, after the cross-link interference CLI characteristic data is transmitted to the user equipment by space division propagation, so that the user equipment feeds back the energy consumption efficiency and channel relationship of each cell to each cell in the CCFD network based on the cross-link interference CLI characteristic data, the method further includes:

[0044] The cell energy efficiency and channel relationship of each cell are summarized through the environment mapping model, and the relationship between the channel relationship of the user equipment on the disturbing side and the channel relationship of the user equipment on the disturbed side is determined to supplement the channel relationship of the user equipment on the disturbed side.

[0045] Figure 3 is a schematic diagram of the information transmission relationship between user equipment, cells, and base stations in a CCFD network according to an embodiment of the present invention, Figure 4 is a flow chart of a method for supplementing a channel relationship of a disturbed side user equipment by a disturbed side base station according to an embodiment of the present invention, such as Figure 4 As shown,

[0046] In one implementation, the cell energy efficiency and channel relationship of each cell are summarized through an environment mapping model, and the relationship between the channel relationship of the disturbing side user equipment and the channel relationship of the disturbed side user equipment is determined to supplement the channel relationship of the disturbed side user equipment, including:

[0047] Step S401, a first channel relationship is obtained by aggregating data of multiple cells corresponding to the disturbed side base station through the disturbed side base station;

[0048] Step S402, the interfering side base station aggregates data of multiple cells corresponding to the interfering side base station to obtain a second channel relationship;

[0049] Step S403, the disturbed side base station supplements the channel relationship of the disturbed side user equipment based on the first channel relationship and the second channel relationship; wherein the disturbed side base station is a base station communicating with the disturbed side user equipment, and the disturbing side base station is a base station communicating with the disturbing side user equipment.

[0050] In an exemplary embodiment, Figure 3 and Figure 4 As shown, in Figure 3 In the system architecture, user equipment A communicates with cell A of base station A to form transmission channel A. User equipment B communicates with cell A of base station A to form transmission channel B. User equipment C communicates with cell B of base station A to form transmission channel C. User equipment D communicates with cell C of base station B to form transmission channel D. User equipment E communicates with cell C of base station B to form transmission channel E.

[0051] In the process of user equipment A communicating with cell A, or in the process of user equipment C communicating with cell B, or in the process of user equipment D communicating with cell C, or in the process of user equipment E communicating with cell C, for the above four communication processes, interference may be caused to user equipment B. Then user equipment B is the disturbed side user equipment, user equipment A or user equipment C or user equipment D or user equipment E may be the disturbing side user equipment, base station A is the disturbed side base station, and base station B is the disturbing side base station. Of course, when user equipment A and user equipment C transmit data through media such as Bluetooth, user equipment A and user equipment C also establish transmission channel N at this time. When user equipment B and transmission channel N have an intersection, the transmission signal between user equipment A and user equipment C will also cause interference to user equipment B.

[0052] Base station A acts as a base station on the victim side. Cell A receives data on transmission channel A sent by user equipment A, and receives data on transmission channel B sent by user equipment B. Cell B receives data on transmission channel C sent by user equipment C. Since there is also a communication relationship between user equipment A and user equipment C, and transmission channel N is established, user equipment A will send data on transmission channel N to cell A, or user equipment C will send data on transmission channel N to cell B. Cell A and cell B transmit information through the station, and base station A summarizes the data received by cell A and cell B. Therefore, it can be known that the channel relationship of user equipment B includes: relationship 1, communicating with cell A through transmission channel B, and relationship 2, the information transmitted by transmission channel N will also cause interference to user equipment B. Therefore, by combining the data of cell B, relationship 2 is supplemented, so as to supplement the channel relationship of the user equipment on the victim side.

[0053] Base station B, as an interfering base station, aggregates the data of transmission channel D and transmission channel E received by cell C. When user equipment D communicates with user equipment A, transmission channel M is established. When user equipment B intersects with transmission channel M, the transmission signal between user equipment D and user equipment A will also cause interference to user equipment B.

[0054] Cell C receives data on transmission channel D sent by user equipment D and data on transmission channel E sent by user equipment E. At the same time, user equipment D sends data on transmission channel M to cell C, and base station B summarizes the data on transmission channel D, transmission channel E, and transmission channel M into a second channel relationship. Base station A summarizes the data of cell A and cell B to obtain the first channel relationship. Base station B and base station A send the data on transmission channel D, transmission channel E, and transmission channel M to base station A through inter-station information exchange. Base station A summarizes the first channel relationship and the second channel relationship, and it can be learned that the channel relationship of user equipment B includes: relationship 1, communicating with cell A through transmission channel B, and relationship 2, the information transmitted by transmission channel M will also cause interference to user equipment B. Therefore, by combining the data of cell B, relationship 2 is supplemented, so as to supplement the channel relationship of the user equipment on the disturbed side.

[0055] Of course, in one implementation, base station A and base station B send their respective data to a server, and the server aggregates the data of base station A and the data of base station B.

[0056] In one implementation, the cross-link interference CLI characteristic data includes at least one of the following dimensions of cell energy efficiency and channel relationship: frequency domain, time domain, and spatial distance information.

[0057] In one embodiment, after the cross-link interference CLI characteristic data is transmitted to the user equipment by space division propagation, so that the user equipment feeds back the energy consumption efficiency and channel relationship of each cell to each cell in the CCFD network based on the cross-link interference CLI characteristic data, the method further includes:

[0058] The network parameters received from multiple cells are subjected to dimensionality reduction mapping and normalization processing to generate a time-frequency domain measurement bitmap, so as to organize the discrete user measurement data into a continuous distribution data set in the time domain.

[0059] Figure 5 is a schematic diagram of CLI measurement mapping of multiple user equipment according to an embodiment of the present invention, such as Figure 5 As shown,

[0060] In an exemplary implementation, taking the channel relationship of network parameters as an example, user equipment 1 (ue-1) has 9 transmission channels located at different locations, user equipment 2 (ue-2) has 9 transmission channels located at different locations, and user equipment 3 (ue-3) has 9 transmission channels located at different locations. Since the locations of the transmission channels of user equipment 1, user equipment 2, and user equipment 3 are the same, there are also different situations. The transmission channels of the three user devices are aligned, dimensionally reduced, mapped, and normalized, and projected into the same area of ​​the same plane (the time-frequency domain measurement bitmap of the cell), and the positions commonly occupied by multiple user devices are colored multiple times (weight superposition).

[0061] For system-level interference management, the traditional method is to manually adjust the cell frame structure or scheduling strategy according to the service type and resource utilization to avoid cross interference. This method requires coordinated adjustment of the entire network, but it cannot take into account the real-time nature of network service changes and may sacrifice some user perception. In addition, it may aggravate the problem of co-frequency interference and affect network performance.

[0062] Since CCFD cells support uplink and downlink duplex in the time-frequency domain, the flexible frame structure (FlexibleFrame) is a necessary and sufficient condition for commercial deployment, which means that the constraints on resource adjustment of traditional cells can be broken. Elastic adjustment of cell resources within the system can be achieved within the CLI interference constraint range. This enables dynamic adjustment based on the service needs of the coverage area to ensure optimal adjustment of user perception and system energy efficiency. How to adjust power resource allocation and scheduling within the system based on service characteristics, energy efficiency, and user perception has become a new challenge, and it is necessary to achieve optimal benefits in a multi-dimensional model.

[0063] In one implementation, the network parameters received by multiple cells are subjected to dimensionality reduction mapping and normalization processing to generate a time-frequency domain measurement bitmap, so as to organize discrete user measurement data into a continuous distribution data set in the time domain, and further includes:

[0064] The network parameters between different base stations are shared through the interface, the network parameters received by the cells between multiple base stations are reduced in dimension and normalized, and a time-frequency domain measurement bitmap is generated to organize the discrete user measurement data into a continuous distribution data set in the time domain.

[0065] In one implementation, determining a base station resource adjustment strategy corresponding to the network parameter based on the network parameter so as to adjust the transmit power of each cell according to the base station resource adjustment strategy includes:

[0066] When the network parameters are obtained through the environment mapping model, the base station power adjustment strategy corresponding to the network parameters is obtained through the DQN network model.

[0067] In one implementation, when the network parameters are obtained through the environment mapping model, the base station power adjustment strategy corresponding to the network parameters is obtained through the DQN network model, including:

[0068] The network parameters are obtained through the optimized network of the DQN network model, and the database of the optimized network is traversed based on the network parameters to obtain the base station power adjustment strategy corresponding to the network parameters, wherein the mapping relationship between the network parameters and the base station power adjustment strategy is stored in the database.

[0069] Figure 6 is a flow chart of a method for obtaining preceding environmental variables, optimizing subsequent state environmental variables, incentive variables, and action variables according to an embodiment of the present invention, such as Figure 6 As shown,

[0070] In one implementation, when the network parameters are obtained through the environment mapping model, after the base station power adjustment strategy corresponding to the network parameters is obtained through the DQN network model, the following method is also performed through the environment mapping model:

[0071] Step S601, defining the network parameters before the cell adjustment as the preceding environment variables;

[0072] Step S602, defining the adjusted network parameters of the cell as optimization post-order state environment variables;

[0073] Step S603, defining the cell energy efficiency as an incentive variable;

[0074] Step S604, defining the cell power adjustment strategy as an action variable;

[0075] Step S605 , storing the preceding environmental variables, the optimized subsequent state environmental variables, the incentive variables, and the action variables as historical data in the experience pool module of the DQN network model to train the DQN network model.

[0076] Figure 7 is a flow chart of a method for obtaining a target value function according to an embodiment of the present invention. Figure 7 As shown,

[0077] In one embodiment, after storing the preceding environmental variables, the optimized subsequent state environmental variables, the incentive variables, and the action variables in the experience pool of the DQN network model, the method further includes executing the following method through the target network module of the DQN network model:

[0078] Step S701, obtaining the previous environment variables sent by the experience pool and optimizing the subsequent state environment variables;

[0079] Step S702, obtaining a base station power adjustment strategy sent by the optimization network module;

[0080] Step S703, based on the preceding environmental variables, the optimized subsequent state environmental variables, and the base station power adjustment strategy, obtain the target value function, so that the error correction module of the DQN network model obtains the weight of the optimized network module based on the target value function; wherein the preceding environmental variables, the optimized subsequent state environmental variables and the base station power adjustment strategy are data of different adjustment processes.

[0081] Figure 8 is a flow chart of a method for obtaining an optimized value function according to an embodiment of the present invention. Figure 8 As shown,

[0082] In one embodiment, after storing the preceding environmental variables, the optimized subsequent state environmental variables, the incentive variables, and the action variables in the experience pool module of the DQN network model, the following method is also performed by optimizing the network:

[0083] Step S801, obtaining the preceding environment variables, optimizing the following state environment variables, and the action variables sent by the experience pool;

[0084] Step S802, based on the preceding environmental variables, the optimized subsequent state environmental variables, and the action variables, an optimized value function is obtained, so that the error correction module obtains the weight of the optimized network module based on the optimized value function; wherein the preceding environmental variables, the optimized subsequent state environmental variables, and the action variables are data of the same adjustment process.

[0085] Fig. 9 is a flow chart of a method for obtaining weights of optimized network modules according to an embodiment of the present invention. Fig. 9 As shown,

[0086] In one embodiment, after storing the preceding environmental variables, the optimized subsequent state environmental variables, the incentive variables, and the action variables in the experience pool module of the DQN network model, the following method is also performed by the error correction module:

[0087] Step S901, obtaining the incentive variable sent by the experience pool module;

[0088] Step S902, based on the target value function, the optimization value function, and the incentive variable, the weight of the optimized network module is obtained.

[0089] In one embodiment, after obtaining the target value function, the method further includes:

[0090] When the optimized network traverses the database of the optimized network based on the network parameters and fails to obtain the base station power adjustment strategy corresponding to the network parameters, the average of multiple target value functions is used as the base station power adjustment strategy.

[0091] In an exemplary embodiment, Fig.10 As shown in the figure, after the DQN network model is trained and applied multiple times, the environment mapping model defines the network parameters before the cell adjustment as the preceding environment variables S t , the network parameters after cell adjustment are defined as the optimized post-order state environment variable S t+1 , the cell energy efficiency is defined as the incentive variable r t , define the cell power adjustment strategy as action variable a t . And set the preceding environment variable S t , optimize the post-order state environment variable S t+1 , incentive variable r t , action variable a t Input into the experience pool module to facilitate training of the DQN network model.

[0092] The experience pool module converts the preceding environment variable S t , action variable a t Input into the optimization network module for training, so that the optimization network module can obtain the optimization value function. t+1 Input to the target network module, and after every preset number of iterations of the optimization network module, the previous environment variable S t , action variable a t And the optimized subsequent state action variable a obtained after adjusting the weight t+1 Sent to the target network module so that the target network module can be based on the previous environment variable S t , action variable a t , optimize the post-order state environment variable S t+1 , optimize the subsequent state action variable a t+1 Get the target value function.

[0093] The error correction module receives the optimization value function Q1 sent by the optimization network model, receives the target value function Q2 sent by the target network model, and receives the incentive variable r sent by the experience pool module. t , based on the optimization value function Q1, the target value function Q2, and the incentive variable r t , using the gradient descent method, the weight θ of the optimized network module is obtained by minimizing the expected prediction error.

[0094] The optimization network module receives the weight θ sent by the error correction module and performs model training and updating so as to obtain corresponding adjustment actions according to the network parameters.

[0095] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by adding the necessary general hardware platform by software, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0096] In this embodiment, a network resource adjustment system is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the system described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0097] Fig.10 is a schematic diagram of the structure of the environment mapping model and the DQN network model according to an embodiment of the present invention, Fig.10 As shown, the system includes:

[0098] An environment mapping model is used to obtain network parameters of each cell in the CCFD network, the network parameters including cell energy efficiency and at least one of the following channel relationships: channel relationship between different base stations, channel relationship between different user equipments;

[0099] The DQN network model is used to obtain the base station power adjustment strategy corresponding to the network parameters based on the network parameters;

[0100] The environment mapping model is also used to adjust the transmit power of each cell based on the base station power adjustment strategy.

[0101] Furthermore, the system is also used to: obtain network parameters of each cell in the CCFD network based on the user equipment, including:

[0102] Through the environment mapping model, cross-link interference CLI feature data is inserted into the downlink scheduling time slot of each cell in the CCFD network, and the cross-link interference CLI feature data is transmitted to the user equipment by space division propagation, so that the user equipment can feedback the energy efficiency and channel relationship of each cell to each cell in the CCFD network based on the cross-link interference CLI feature data. Among them, the cross-link interference CLI feature data includes at least one of the following dimensions of cell energy efficiency and channel relationship: frequency domain, time domain, and space distance information.

[0103] Furthermore, the system is also used to: transmit the cross-link interference CLI characteristic data to the user equipment by using the space division propagation method, so that the user equipment feeds back the energy consumption efficiency and channel relationship of each cell to each cell in the CCFD network based on the cross-link interference CLI characteristic data, and further includes:

[0104] The cell energy efficiency and channel relationship of each cell are summarized through the environment mapping model, and the relationship between the channel relationship of the user equipment on the disturbing side and the channel relationship of the user equipment on the disturbed side is determined to supplement the channel relationship of the user equipment on the disturbed side.

[0105] Furthermore, the system is also used to: summarize the cell energy consumption efficiency and channel relationship of each cell through the environment mapping model, determine the relationship between the channel relationship of the user equipment on the disturbing side and the channel relationship of the user equipment on the disturbed side, so as to supplement the channel relationship of the user equipment on the disturbed side, including:

[0106] The disturbed side base station aggregates data of multiple cells corresponding to the disturbed side base station to obtain a first channel relationship;

[0107] The interfering side base station aggregates data of multiple cells corresponding to the interfering side base station to obtain a second channel relationship;

[0108] The disturbed side base station supplements the channel relationship of the disturbed side user equipment based on the first channel relationship and the second channel relationship; wherein the disturbed side base station is a base station communicating with the disturbed side user equipment, and the disturbing side base station is a base station communicating with the disturbing side user equipment.

[0109] Furthermore, the system is also used to: transmit the cross-link interference CLI characteristic data to the user equipment by using the space division propagation method, so that the user equipment feeds back the energy consumption efficiency and channel relationship of each cell to each cell in the CCFD network based on the cross-link interference CLI characteristic data, and further includes:

[0110] The network parameters received from multiple cells are subjected to dimensionality reduction mapping and normalization processing to generate a time-frequency domain measurement bitmap, so as to organize the discrete user measurement data into a continuous distribution data set in the time domain.

[0111] Furthermore, the system is also used to: perform dimension reduction mapping and normalization processing on the network parameters received by multiple cells to generate a time-frequency domain measurement bitmap to organize discrete user measurement data into a continuous distribution data set in the time domain, and also includes:

[0112] The network parameters between different base stations are shared through the interface, the network parameters received by the cells between multiple base stations are reduced in dimension and normalized, and a time-frequency domain measurement bitmap is generated to organize the discrete user measurement data into a continuous distribution data set in the time domain.

[0113] Furthermore, the system is also used to: determine a base station resource adjustment strategy corresponding to the network parameter based on the network parameter, so as to adjust the transmission power of each cell according to the base station resource adjustment strategy, including:

[0114] When the network parameters are obtained through the environment mapping model, the base station power adjustment strategy corresponding to the network parameters is obtained through the DQN network model.

[0115] Furthermore, the system is also used to: when the network parameters are obtained through the environment mapping model, obtain the base station power adjustment strategy corresponding to the network parameters through the DQN network model, including:

[0116] The network parameters are obtained through the optimized network of the DQN network model, and the database of the optimized network is traversed based on the network parameters to obtain the base station power adjustment strategy corresponding to the network parameters, wherein the mapping relationship between the network parameters and the base station power adjustment strategy is stored in the database.

[0117] Furthermore, the system is also used for: in the case of obtaining network parameters through the environment mapping model, after obtaining the base station power adjustment strategy corresponding to the network parameters through the DQN network model, it also includes executing the following method through the environment mapping model:

[0118] The network parameters before cell adjustment are defined as pre-order environment variables;

[0119] The adjusted network parameters of the cell are defined as optimization post-order state environment variables;

[0120] The cell energy efficiency is defined as the incentive variable;

[0121] The cell power adjustment strategy is defined as an action variable;

[0122] The preceding environmental variables, the optimized subsequent state environmental variables, the incentive variables, and the action variables are stored as historical data in the experience pool module of the DQN network model to train the DQN network model.

[0123] Furthermore, the system is also used to: after storing the preceding environmental variables, the optimized subsequent state environmental variables, the incentive variables, and the action variables in the experience pool of the DQN network model, it also includes executing the following method through the target network module of the DQN network model:

[0124] Obtain the preceding environment variables sent by the experience pool and optimize the following state environment variables;

[0125] Obtaining the base station power adjustment strategy sent by the optimization network module;

[0126] Based on the preceding environmental variables, the optimized subsequent state environmental variables, and the base station power adjustment strategy, a target value function is obtained, so that the error correction module of the DQN network model obtains the weight of the optimized network based on the target value function; wherein the preceding environmental variables, the optimized subsequent state environmental variables, and the base station power adjustment strategy are data of different adjustment processes.

[0127] Furthermore, the system is also used to: after storing the preceding environmental variables, the optimized subsequent state environmental variables, the incentive variables, and the action variables in the experience pool module of the DQN network model, it also includes executing the following method by optimizing the network:

[0128] Obtain the preceding environment variables sent by the experience pool, optimize the following state environment variables, and action variables;

[0129] Based on the preceding environmental variables, the optimized subsequent state environmental variables and the action variables, an optimized value function is obtained so that the error correction module obtains the weight of the optimized network module based on the optimized value function; wherein the preceding environmental variables, the optimized subsequent state environmental variables and the action variables are data of the same adjustment process.

[0130] Furthermore, the system is also used to: after storing the preceding environmental variables, the optimized subsequent state environmental variables, the incentive variables, and the action variables in the experience pool module of the DQN network model, it also includes executing the following method through the error correction module:

[0131] Get the incentive variables sent by the experience pool module;

[0132] Based on the target value function, the optimization value function and the incentive variables, the weights of the optimized network modules are obtained.

[0133] Furthermore, the system is also used for: after obtaining the target value function, further comprising:

[0134] When the optimized network traverses the database of the optimized network based on the network parameters and fails to obtain the base station power adjustment strategy corresponding to the network parameters, the average of multiple target value functions is used as the base station power adjustment strategy.

[0135] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0136] The above method and system are explained below with reference to examples:

[0137] For example, the current CCFD system has m∈{1,2,3,…,M+1} cells, where user j∈{1,2,3,…,N}, where M+1<,x jm Indicates the access relationship between user j and cell m. There are several service user equipments in the current network and they are distributed in an incoherent manner.

[0138] Due to the asymmetric uplink and downlink characteristics of the CCFD system, there is cross-link interference between base stations and user equipment. Each service cell inserts CLI characteristic data in the downlink scheduling time slot. After space division propagation within the system, each user equipment can receive the CLI measurement characteristic data sent by each cell. This time, the user cross link is defined as follows:

[0139]

[0140] (Formula 1)

[0141] Among them, x j‘m‘ is the non-access channel connection relationship, P j‘m‘ is the transmission power of other cells or other users in the non-access channel to user j, k j‘m‘ In order to reduce the dimension by measuring the relationship between the bitmap and the channel, the interference channel gain is normalized. The dimension reduction process projects the spatial domain to the frequency domain scheduling through the histogram method.

[0142] The user signal to interference plus noise ratio SINR (Signal to Interference plus Noise Ratio) is defined by formula 2 as:

[0143]

[0144] Among them, γ jm is the signal-to-noise ratio between user j and cell m, P jm is the transmission power of cell m to user j, h jm is the channel gain between user j and cell m, δ 2 is additive white Gaussian noise. jm It is the cross-layer co-frequency cross-link interference of other cells. jm is the internal self-interference of cell m.

[0145] Definition: The actual transmission rate of user j associated with cell m is:

[0146] τ jm =Wlog2(1+γ jm )

[0147] (Formula 3)

[0148] Definition: The resource relationship of the entire network system is:

[0149]

[0150] in, For the entire network resources, is the actual resource of cell m, is the cell resource adjustment factor, Reserve resources for the cell.

[0151] Taking into account the maximum and minimum resource constraints of the CCFD system cell, the user perception rate constraints, the CLI / SI interference constraints, and the scheduling cooperation constraints, the optimization problem with the maximum energy efficiency as the goal is defined, as shown in the following formula:

[0152] P1:

[0153] stC1:x jm ={0,1}

[0154]

[0155]

[0156]

[0157]

[0158] C6:P min ≤P m ≤P max

[0159] (Formula 5)

[0160] Where X = [x jm ],P=[P jm ], C1 and C2 represent user cell association constraints, C3 represents user perceived rate constraints, C4 and C5 represent system CLI / SI interference constraints, and C6 represents cell resource constraints. It can be seen that problem P1 is non-convex and is a nonlinear optimization problem.

[0161] The optimization goal of this problem is to effectively improve the system energy efficiency while ensuring the perception of network users, and to optimize the system network power consumption resource gradient convergence on the basis of not decreasing the user perception gradient.

[0162] Due to the nonlinear characteristics of problem P1 and the randomness of network eigenvalues, the solution process is very complicated. Problem P1 is divided into four sub-problem solution processes: user measurement and normalization processing, energy efficiency experience playback, optimization network and target network model training, and gradient error correction.

[0163] Initialize the optimization network model and the target network model according to the network scale and performance requirements. The initialization parameters include: number of iterations M, number of scheduling time slots per iteration T, learning rate λ, and exploration rate ε;

[0164] Initialize the experience pool according to the network scale and performance requirements, set the number of sample sets N, and initialize the model value function Q.

[0165] According to the maximum and minimum limits of base station power and the current base station service type, the initial power configuration of each cell is performed; and the power adjustment strategy is quantified to divide the cell transmission power into |A| orders of magnitude, as shown in the following formula:

[0166] A∈{0,P min ,…,P max}

[0167] (Formula 6)

[0168] For the entire network system, or a specified cell set, or a user set, periodic CLI measurements are performed and the optimization process of the number of iterations M is started. During the measurement period, the user feeds back the transmission channel and interference channel CLI reports to the access cell, and the corresponding overall measurement results are obtained by aggregating the data. Through the histogram method, the dimension reduction processing is performed, and the spatial domain is projected into the frequency domain to obtain the time-frequency domain interference bitmap, so as to obtain the channel relationship k of each user for the entire network j‘m‘ , as the environment variable(s) input to the DQN network.

[0169] In each iteration cycle, for each scheduling time slot, the DQN network outputs the power adjustment strategy and action variable (a) according to the environment variable (S) combined with the current network weight, and acts on the actual CCFD system. Because DQN can only process discrete action spaces, it is necessary to use the ε-greedy method to perform random sampling in the power adjustment strategy quantization set A, as shown in the formula:

[0170]

[0171] If the corresponding action can be traversed in the power adjustment strategy quantization set, it will act on the real environment; if there is no corresponding action, the action corresponding to the optimal value function is selected and acted on the real environment.

[0172] In the real environment, the transmit power of each cell is adjusted π(a|s). The relationship between the measured bitmap and the channel before adjustment is used as the preceding environmental variable (s t ), the adjusted measurement bitmap and channel relationship is used as the optimization post-order state environment variable (s t+1 ), the energy efficiency of the base station side is used as the incentive variable for optimizing the network model (r t ), the current adjustment action is used as the action variable (a t The above variables are stored in the experience pool as experience input, and this process is repeated until the number of experience items meets the DQN training requirements. When the experience pool is full, the priority experience playback mechanism is used to randomly sample from the experience pool and the reward (r t ) and the next state (s t +1 ) is sent to the optimization network to calculate the optimization target.

[0173] Due to the strong correlation between the previous and next sequences in the reinforcement learning training process, the traditional "actor-critic" algorithm is difficult to converge. Therefore, this invention defines: after the optimization network model DQN is iterated a certain number of times, the model parameters are synchronized to the target network. The synchronized parameters include the weight θ. After synchronization, the weight of the target network is θ μ .

[0174] The optimization network model and the target network model value function (Q function) use the gradient descent method to update the model weights by minimizing the expected prediction error, as shown in the following formula:

[0175]

[0176] in To solve the function gradient, from Formula 7 we can know that the DQN weight Q weight θ is updated as follows:

[0177] θ μ+1 ←λθ μ +(1-λ)θ μ+1

[0178] (Formula 9).

[0179] The present invention decomposes the nonlinear optimization problem of the entire network into multiple processes including user measurement and normalization processing, energy efficiency experience playback, optimization network and target network model training, and gradient error correction, and obtains the resource configuration allocation adjustment strategy for each cell after optimization through iterative calculation.

[0180] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0181] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0182] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0183] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0184] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0185] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0186] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for adjusting network resources, characterized in that: include: Acquiring network parameters of each cell in the CCFD network based on the user equipment, the network parameters including cell energy consumption efficiency and at least one of the following channel relationships: channel relationships between different base stations, channel relationships between disturbing-side user equipment and disturbed-side user equipment; wherein the user equipment includes the disturbing-side user equipment and the disturbed-side user equipment; A base station resource adjustment strategy corresponding to the network parameter is determined based on the network parameter, so that the transmission power of each cell is adjusted according to the base station resource adjustment strategy.

2. The method according to claim 1, characterized in that Obtain network parameters of each cell in the CCFD network based on user equipment, including: The cross-link interference CLI characteristic data is inserted into the downlink scheduling time slot of each cell in the CCFD network through the environmental mapping model, and the cross-link interference CLI characteristic data is transmitted to the user equipment by using the space division propagation method, so that the user equipment can feedback the energy consumption efficiency of each cell and the channel relationship to each cell in the CCFD network based on the cross-link interference CLI characteristic data.

3. The method according to claim 2, characterized in that The method further comprises: transmitting the cross-link interference CLI characteristic data to the user equipment by using a space division propagation method, so that the user equipment feeds back the energy consumption efficiency of each cell and the channel relationship to each cell in the CCFD network based on the cross-link interference CLI characteristic data; The cell energy consumption efficiency and the channel relationship of each cell are summarized through an environment mapping model, and the relationship between the channel relationship of the disturbing side user equipment and the channel relationship of the disturbed side user equipment is determined to supplement the channel relationship of the disturbed side user equipment.

4. The method according to claim 2, characterized in that: Summarizing the cell energy consumption efficiency and the channel relationship of each cell through an environment mapping model, determining the relationship between the channel relationship of the disturbing-side user equipment and the channel relationship of the disturbed-side user equipment to supplement the channel relationship of the disturbed-side user equipment, including: Aggregating data of multiple cells corresponding to the disturbed side base station through the disturbed side base station to obtain a first channel relationship; A second channel relationship is obtained by aggregating data of multiple cells corresponding to the interfering side base station through the interfering side base station; The disturbed side base station supplements the channel relationship of the disturbed side user equipment based on the first channel relationship and the second channel relationship; wherein the disturbed side base station is a base station that communicates with the disturbed side user equipment, and the disturbing side base station is a base station that communicates with the disturbing side user equipment.

5. The method according to claim 2, characterized in that: The cross-link interference CLI characteristic data includes at least factors of the cell energy consumption efficiency and the channel relationship in one of the following dimensions: frequency domain, time domain, and spatial distance information.

6. The method according to claim 2, characterized in that The method further comprises: transmitting the cross-link interference CLI characteristic data to the user equipment by using a space division propagation method, so that the user equipment feeds back the energy consumption efficiency of each cell and the channel relationship to each cell in the CCFD network based on the cross-link interference CLI characteristic data; The network parameters received by the plurality of cells are subjected to dimensionality reduction mapping and normalization processing to generate a time-frequency domain measurement bitmap, so as to organize the discrete user measurement data into a continuous distribution data set in the time domain.

7. The method according to claim 6, characterized in that The network parameters received by the plurality of cells are subjected to dimensionality reduction mapping and normalization processing to generate a time-frequency domain measurement bitmap, so as to organize the discrete user measurement data into a continuous distribution data set in the time domain, and further comprising: The network parameters between different base stations are shared through the interface, the network parameters received by the cells between the multiple base stations are subjected to dimensionality reduction mapping and normalization processing, and a time-frequency domain measurement bitmap is generated to organize the discrete user measurement data into a continuous distribution data set in the time domain.

8. The method according to claim 1, characterized in that Determining a base station resource adjustment strategy corresponding to the network parameter based on the network parameter, so as to adjust the transmit power of each cell according to the base station resource adjustment strategy, includes: In the case where the network parameters are obtained through the environment mapping model, the base station power adjustment strategy corresponding to the network parameters is obtained through the DQN network model.

9. The method according to claim 8, characterized in that In the case where the network parameters are obtained through the environment mapping model, a base station power adjustment strategy corresponding to the network parameters is obtained through the DQN network model, including: The network parameters are obtained through the optimized network of the DQN network model, and the database of the optimized network is traversed based on the network parameters to obtain the base station power adjustment strategy corresponding to the network parameters, wherein the database stores a mapping relationship between the network parameters and the base station power adjustment strategy.

10. The method according to claim 8, characterized in that In the case where the network parameters are obtained through the environment mapping model, after the base station power adjustment strategy corresponding to the network parameters is obtained through the DQN network model, the following method is also performed through the environment mapping model: defining the network parameters before the cell adjustment as pre-order environment variables; defining the adjusted network parameters of the cell as optimization post-order state environment variables; The cell energy efficiency is defined as an incentive variable; defining the cell power adjustment strategy as an action variable; The preceding environmental variables, the optimized subsequent state environmental variables, the incentive variables, and the action variables are stored as historical data in the experience pool module of the DQN network model to train the DQN network model.

11. The method according to claim 10, characterized in that After storing the preceding environmental variables, the optimized subsequent state environmental variables, the incentive variables, and the action variables in the experience pool of the DQN network model, the method further includes executing the following method through the target network module of the DQN network model: Obtaining the preceding environment variables sent by the experience pool and optimizing the following state environment variables; Obtaining the base station power adjustment strategy sent by the optimization network module; Based on the preceding environmental variables, the optimized subsequent state environmental variables, and the base station power adjustment strategy, a target value function is obtained so that the error correction module of the DQN network model obtains the weight of the optimized network module based on the target value function; wherein the preceding environmental variables, the optimized subsequent state environmental variables, and the base station power adjustment strategy are data of different adjustment processes.

12. The method according to claim 11, characterized in that After storing the preceding environmental variables, the optimized subsequent state environmental variables, the incentive variables, and the action variables in the experience pool module of the DQN network model, the method further includes executing the following method through the optimized network: Obtain the preceding environment variables, optimized subsequent state environment variables, and action variables sent by the experience pool; Based on the preceding environmental variables, the optimized subsequent state environmental variables and the action variables, an optimized value function is obtained so that the error correction module obtains the weight of the optimized network module based on the optimized value function; wherein the preceding environmental variables, the optimized subsequent state environmental variables and the action variables are data of the same adjustment process.

13. The method according to claim 12, characterized in that After storing the preceding environmental variables, the optimized subsequent state environmental variables, the incentive variables, and the action variables in the experience pool module of the DQN network model, the method further includes executing the following method through the error correction module: Acquire the incentive variable sent by the experience pool module; Based on the target value function, the optimization value function, and the incentive variable, the weight of the optimization network module is obtained.

14. The method according to claim 11, characterized in that After obtaining the target value function, it also includes: When the optimized network traverses the database of the optimized network based on the network parameters and fails to obtain the base station power adjustment strategy corresponding to the network parameters, the average of the multiple target value functions is used as the base station power adjustment strategy.

15. A network resource adjustment system, characterized in that: include: An environment mapping model is used to obtain network parameters of each cell in the CCFD network based on user equipment, wherein the network parameters include cell energy consumption efficiency and at least one of the following channel relationships: channel relationships between different base stations, and channel relationships between disturbing-side user equipment and disturbed-side user equipment; wherein the user equipment includes the disturbing-side user equipment and the disturbed-side user equipment; The DQN network model is used to determine the base station resource adjustment strategy corresponding to the network parameter based on the network parameter, so as to adjust the transmission power of each cell according to the base station resource adjustment strategy.

16. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method described in any one of claims 1 to 14 when executed by a processor.

17. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method described in any one of claims 1 to 14 are implemented.

Images