Base station resource allocation optimization method and device and readable storage medium

Abstract

The invention provides a base station resource allocation optimization method and apparatus, and a readable storage medium. The method comprises the steps of obtaining performance data of a base station; according to the performance data, determining a resource adaptive allocation strategy through a deep reinforcement learning model; according to the performance data, position information of an interference source is determined through a weighted average algorithm, and an interference isolation strategy is determined; and carrying out allocation optimization processing on the base station resources based on the resource adaptive allocation strategy and the interference isolation strategy. According to the base station resource allocation optimization method provided by the invention, intelligent resource scheduling and interference management of the mobile communication base station are realized through deep reinforcement learning, interference analysis, cooperative load and a self-healing mechanism. According to the scheme, a highly dynamic network environment can be effectively handled, the resource utilization rate of the base station and the user experience are improved through real-time sensing and intelligent adjustment, and the stability and high-efficiency operation of the network are guaranteed.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a base station resource allocation optimization method, device and readable storage medium. Background Art

[0002] With the rapid development of mobile communication technology, mobile networks need to support massive device access and handle growing data traffic. This demand is even more urgent in 5G and the upcoming 6G networks.

[0003] Mobile communication base stations are the core nodes of wireless networks, responsible for signal transmission, coverage and resource allocation. Currently, in terms of resource scheduling and interference management of mobile communication base stations, commonly used wireless network optimization methods include static resource allocation and preset rule scheduling, load balancing based on traffic classification, and interference management based on channel switching.

[0004] However, existing technologies still have shortcomings such as lack of dynamic adaptability, lack of precision in interference management, and limited energy efficiency optimization effects.

[0005] Therefore, how to better implement base station resource scheduling and interference management has become a problem that needs to be solved. Summary of the invention

[0006] The technical problem to be solved by the present application is to provide a base station resource allocation optimization method, device and readable storage medium to solve the problems existing in the prior art in view of the above-mentioned deficiencies in the prior art.

[0007] In a first aspect, the present application provides a method for optimizing base station resource allocation, the method

[0008] The law includes:

[0009] S1. Obtain performance data of the base station;

[0010] S2. Determine a resource adaptive allocation strategy through a deep reinforcement learning model based on the performance data;

[0011] S3. Determine the location information of the interference source through a weighted average algorithm according to the performance data, and determine an interference isolation strategy;

[0012] S4. Performing allocation optimization processing of base station resources based on the resource adaptive allocation strategy and the interference isolation strategy.

[0013] In some embodiments, the performance data of the base station includes at least one of the following:

[0014] Channel quality indication CQI;

[0015] Signal strength RSSI;

[0016] User density UE Density;

[0017] Traffic Demand

[0018] Transmission delay Latency.

[0019] In some embodiments, S2 includes:

[0020] S21, performing feature extraction and dimensionality reduction processing on the performance data to obtain a feature vector;

[0021] S22. Determine a resource adaptive allocation strategy through a deep reinforcement learning model according to the feature vector;

[0022] The resource adaptive allocation strategy includes at least one of bandwidth allocation, power control, and spectrum selection.

[0023] In some embodiments, S3 includes:

[0024] S31, assigning corresponding weights according to the channel quality indicator CQI of each base station;

[0025] S32, determining the location information of the interference source by a weighted average algorithm according to the weight corresponding to each base station;

[0026] S33, determining an interference isolation strategy according to the location information of the interference source;

[0027] The interference isolation strategy includes channel switching or adjusting power output.

[0028] In some embodiments, it also includes:

[0029] Obtain current load data and historical load data of the base station;

[0030] Predicting future load data of the base station using a sliding window prediction model according to the current load data and the historical load data;

[0031] Determining a load adjustment strategy for the base station according to the future load data;

[0032] The load adjustment strategy includes user diversion or output power adjustment.

[0033] In some embodiments, it also includes:

[0034] Obtain real-time fault monitoring data and historical fault data of base stations;

[0035] According to the real-time fault monitoring data and the historical fault data, future fault data of the base station is predicted by a long short-term memory network LSTM model;

[0036] determining a resource adjustment strategy of the base station according to the future fault data;

[0037] The resource adjustment strategy includes at least one of bandwidth reallocation, power adjustment, spectrum adjustment, and load migration.

[0038] In some embodiments, it also includes:

[0039] A fault self-repair strategy is configured for the base station, where the fault self-repair strategy is used to adjust resources when a fault occurs in the base station so that the base station can be restored to a normal working state.

[0040] In a second aspect, the present application provides a base station resource allocation optimization device, the device comprising:

[0041] A data acquisition module, configured to acquire performance data of a base station;

[0042] A resource allocation module, configured to determine a resource adaptive allocation strategy through a deep reinforcement learning model based on the performance data;

[0043] An interference isolation module, configured to determine the location information of the interference source through a weighted average algorithm according to the performance data, and determine an interference isolation strategy;

[0044] The optimization processing module is configured to perform allocation optimization processing of base station resources based on the resource adaptive allocation strategy and the interference isolation strategy.

[0045] In a third aspect, the present application provides a base station resource allocation optimization device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement the base station resource allocation optimization method described in the first aspect above.

[0046] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the base station resource allocation optimization method described in the first aspect above is implemented.

[0047] The present application provides a method, device and readable storage medium for optimizing base station resource allocation, the method comprising: obtaining performance data of the base station; determining a resource adaptive allocation strategy through a deep reinforcement learning model based on the performance data; determining the location information of the interference source through a weighted average algorithm based on the performance data, and determining an interference isolation strategy; performing allocation optimization processing of base station resources based on the resource adaptive allocation strategy and the interference isolation strategy. The present application provides a method for optimizing base station resource allocation, which realizes intelligent resource scheduling and interference management of mobile communication base stations through deep reinforcement learning, interference analysis, collaborative load and self-healing mechanism. The solution can effectively cope with highly dynamic network environments, improve the resource utilization and user experience of base stations through real-time perception and intelligent adjustment, and ensure the stability and high-efficiency operation of the network. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0049] Figure 1 A flowchart of a base station resource allocation optimization method provided in an embodiment of the present application;

[0050] Figure 2 A schematic diagram of the structure of a base station resource allocation optimization device provided in an embodiment of the present application;

[0051] Figure 3 A schematic diagram of the structure of another base station resource allocation optimization device provided in an embodiment of the present application.

[0052] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0053] In order to enable those skilled in the art to better understand the technical solution of the present application, the implementation mode of the present application will be further described in detail below with reference to the accompanying drawings.

[0054] It should be understood that the specific embodiments and drawings described herein are only used to explain the present application, rather than to limit the present application.

[0055] It can be understood that, in the absence of conflict, the various embodiments in the present application and the various features in the embodiments can be combined with each other.

[0056] It can be understood that, for the convenience of description, the drawings of the present application only show the parts related to the present application, while the parts unrelated to the present application are not shown in the drawings.

[0057] It can be understood that each unit and module involved in the embodiments of the present application may correspond to only one physical structure, or may be composed of multiple physical structures, or multiple units and modules may be integrated into one physical structure.

[0058] It can be understood that the terms "first", "second", etc. in the embodiments of the present application are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.

[0059] It is understandable that, in the absence of conflict, the functions and steps marked in the flowcharts and block diagrams of the present application may occur in an order different from that marked in the drawings.

[0060] It is understood that the flowcharts and block diagrams of the present application illustrate the possible architectures, functions, and operations of the systems, devices, equipment, and methods according to the various embodiments of the present application. Among them, each box in the flowchart or block diagram may represent a unit, module, program segment, code, which contains executable instructions for implementing the specified functions. Moreover, each box or combination of boxes in the block diagram and flowchart may be implemented by a hardware-based system that implements the specified functions, or by a combination of hardware and computer instructions.

[0061] It can be understood that the units and modules involved in the embodiments of the present application can be implemented by software or hardware, for example, the units and modules can be located in a processor.

[0062] In the prior art, in terms of resource scheduling and interference management of mobile communication base stations, the solutions closest to the present application include the following commonly used wireless network optimization methods:

[0063] (1) Static resource allocation and preset rule scheduling:

[0064] This type of method relies on pre-set scheduling rules, such as fixed bandwidth allocation, power control, spectrum switching, etc. The base station allocates resources under preset conditions to meet the needs of specific users or traffic. This method is suitable for low-dynamic environments, but it will show a certain lag in scenarios where the number of users and demand changes greatly. Network energy efficiency optimization methods: Some traditional methods optimize energy efficiency to a certain extent through fixed algorithms (such as power control, signal spectrum allocation, etc.). However, these methods are mostly based on preset parameters, and their optimization capabilities are limited when burst traffic increases or under high load conditions, making it difficult to achieve adaptive adjustment.

[0065] (2) Load balancing based on traffic classification:

[0066] By analyzing and classifying the demand types of different user traffic (such as real-time video, high-bandwidth demand traffic, etc.), some solutions have achieved load balancing to alleviate congestion. However, such methods still lack adaptability to network changes and it is difficult to achieve dynamic adjustment of resources.

[0067] (3) Interference management based on channel switching:

[0068] Interference management mainly relies on spectrum switching or fixed interference detection rules to isolate interference sources. The base station regularly detects channel interference and switches spectrum when the interference exceeds the threshold. However, this method cannot accurately locate the interference source, resulting in frequent spectrum switching and waste of resources, and cannot achieve refined interference isolation.

[0069] The shortcomings of the prior art are mainly concentrated in the following aspects:

[0070] (1) Lack of dynamic adaptability:

[0071] The static rules and fixed algorithms in the existing technology cannot adapt to complex network environments, especially in high-load or interference environments, and performance is easily affected. Changes in the network environment (such as an increase in the number of users and traffic bursts) will cause static rules to fail, thereby affecting service quality.

[0072] (2) Lack of precision in interference management:

[0073] Existing interference management methods rely on simple spectrum switching and are unable to accurately locate and effectively isolate interference sources, resulting in waste of spectrum resources and degradation of signal quality, especially in high-interference environments.

[0074] (3) The effect of energy efficiency optimization is limited:

[0075] Energy efficiency optimization in existing technologies is mainly based on fixed parameter control, without combining real-time traffic and load prediction, so it is difficult to maintain efficient energy consumption management when the load surges or traffic is concentrated. Lack of automatic fault repair capabilities: Most base stations require manual intervention when they fail or their performance degrades, which makes it difficult to automatically repair them in real time, affecting network stability and user experience.

[0076] To address the above-mentioned defects, this application is based on machine learning. First, network environment data is collected in real time in the base station, including characteristics such as channel quality, user density, signal strength and transmission delay. On this basis, a deep reinforcement learning model is used to adaptively schedule resources.

[0077] Specifically, the base station generates a dynamic resource allocation strategy, including bandwidth allocation, spectrum selection, and power control, based on real-time network status (such as user demand, interference level, etc.). The deep reinforcement learning model optimizes resource allocation through a reward mechanism to balance service quality, energy consumption, and interference management. The model continuously adjusts resource allocation based on feedback, allowing the base station to adapt to the ever-changing network environment.

[0078] At the same time, the solution uses data fusion and interference analysis technology between adjacent base stations to identify and locate interference sources, and isolate interference sources through channel switching or power adjustment to improve signal quality and connection stability. In terms of load management, it predicts future traffic demand and reallocates resources based on load conditions to ensure that traffic is optimized in high-density areas while avoiding excessive energy consumption.

[0079] In addition, the solution has designed a self-healing mechanism for base stations, which uses a long short-term memory network (LSTM) to learn and predict the historical status of base stations and identify potential failures or performance degradation in advance. When an abnormality is detected, the base station automatically triggers the self-healing mechanism and ensures the normal operation of the network by switching to backup resources or adjusting power and spectrum.

[0080] The purposes of this application include:

[0081] (1) Adaptive resource scheduling: Deep reinforcement learning is used to optimize resource allocation in real time according to network status to meet dynamically changing user needs and traffic loads and avoid the lag of static resource allocation.

[0082] (2) Improve the accuracy of interference management: Through the coordinated spatiotemporal data analysis of multiple base stations, the interference source can be located and intelligently isolated, thereby reducing spectrum switching and signal quality loss.

[0083] (3) Improve energy efficiency: Use edge computing nodes to predict load conditions, intelligently adjust power output and resource allocation, and achieve efficient energy management. Automatic fault repair: The base station has a built-in self-healing mechanism that automatically adjusts resource configuration when performance degrades or the load is too high through fault prediction and automatic resource reconstruction, thereby improving network stability.

[0084] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0085] The present application provides a base station resource allocation optimization method, the working process of the method can be implemented by electronic devices, such as computers, handheld smart terminals, etc. For the convenience of explanation, the method execution subject is described as a computer in each embodiment of the present application.

[0086] Figure 1 A schematic diagram of a base station resource allocation optimization method provided in an embodiment of the present application, such as Figure 1 As shown, the present application provides a base station resource allocation optimization method, the method includes S1-S4, which are as follows:

[0087] S1. Obtain performance data of the base station;

[0088] In some embodiments, the performance data of the base station includes at least one of the following:

[0089] Channel Quality Indicator (CQI): A quantitative indicator of channel quality that reflects the reliability of the current signal.

[0090] Signal Strength (RSSI): Indication of the signal strength received by the base station, representing the stability of the connection.

[0091] User density (UE Density): The number of user devices connected to a base station.

[0092] Traffic Demand: The amount of data transmitted by a base station.

[0093] Transmission delay (Latency): refers to the communication delay from the base station to the user end.

[0094] S2. Determine a resource adaptive allocation strategy through a deep reinforcement learning model based on the performance data;

[0095] In some embodiments, S2 includes:

[0096] S21, performing feature extraction and dimensionality reduction processing on the performance data to obtain a feature vector;

[0097] S22. Determine a resource adaptive allocation strategy through a deep reinforcement learning model according to the feature vector;

[0098] The resource adaptive allocation strategy includes at least one of bandwidth allocation, power control, and spectrum selection.

[0099] Specifically, the collected data is first subjected to feature extraction and dimensionality reduction processing to form a feature vector, as follows:

[0100] X=[CQI,UE Density,Traffic Demand,RSSI,Latency]

[0101] Then, a deep reinforcement learning model is used to adaptively allocate resources based on the current network environment to optimize base station bandwidth, power, and spectrum usage. The deep reinforcement learning model is constructed as follows:

[0102] State (St): The feature vector X extracted in the previous stage constitutes the current network state, including channel quality, user density, etc.

[0103] Action (At): Resource allocation strategy output by the model, such as bandwidth allocation, power control, and spectrum selection.

[0104] Reward (Rt): The reward value is determined based on the network performance after resource allocation, including quality of service (QoS), energy consumption, and interference.

[0105] Reinforcement learning process: Through the implementation of action At (i.e. resource allocation strategy), the performance indicators of the base station (such as QoS, network delay, energy efficiency) are understood and the reward Rt is calculated:

[0106] R t =α·QoS(S t , A t )-β·Energy(S t , A t )-γ·Interference(S t , A t )

[0107] in:

[0108] α, β, and γ are weight coefficients used to balance QoS, energy consumption, and interference.

[0109] QoS(St,At) evaluates the impact of the current resource allocation strategy on service quality.

[0110] Energy(St,At) represents the power consumed by the resource allocation strategy.

[0111] Interference(St,At) indicates the current interference level.

[0112] In addition, the Deep Q Network (DQN) model is used to optimize the Q value through the Experience Replay and Target Network mechanisms. Based on the current state and rewards, reinforcement learning continuously optimizes the model by continuously updating the Q value, so that the strategy selected by the model maximizes the future cumulative rewards, and optimizes the resource allocation of the base station under different network conditions:

[0113]

[0114] Here, η is the learning rate and δ is the discount factor used to balance short-term and long-term rewards.

[0115] In this step, the deep reinforcement learning model generates the optimal resource allocation strategy. Based on the trained optimal strategy, the base station can intelligently schedule bandwidth, power, and spectrum resources to adaptively meet the needs of different users and traffic. For example:

[0116] Bandwidth allocation: prioritize bandwidth allocation to users with high QoS requirements.

[0117] Power control: Increase the power when the signal strength is weak, and reduce the power when the signal strength is too strong.

[0118] Spectrum selection: Select channels based on interference levels and avoid high-interference frequency bands.

[0119] S3. Determine the location information of the interference source through a weighted average algorithm according to the performance data, and determine an interference isolation strategy;

[0120] In some embodiments, S3 includes:

[0121] S31, assigning corresponding weights according to the channel quality indicator CQI of each base station;

[0122] S32, determining the location information of the interference source by a weighted average algorithm according to the weight corresponding to each base station;

[0123] S33, determining an interference isolation strategy according to the location information of the interference source;

[0124] The interference isolation strategy includes channel switching or adjusting power output.

[0125] This step uses data between multiple base stations to identify and locate interference sources and automatically adjust channels to isolate interference.

[0126] Specifically, in order to estimate the location of the interference source, the signal quality indicator of each base station (such as the CQI indicator of the base station) can be used, and through data sharing between adjacent base stations, the base stations provide signal interference information to each other to build a data set.

[0127] Assume that the impact of interference sources on the signal quality of surrounding base stations follows a certain attenuation law, for example, the closer the base station is to the interference source, the more obvious the signal quality degradation, and the farther the base station is from the interference source, the higher the signal quality. The CQI value of a base station is an indicator of the quality of its received signal.

[0128] In this solution, the larger the CQI value, the better the signal quality, indicating that the base station receives a stronger signal and is farther away from the interference source; the lower the CQI value, the worse the signal quality, indicating that the base station receives a weaker signal and is closer to the interference source. Therefore, the inverse of the CQI value can be used to indicate the severity of the interference to the base station, that is, the closer to the interference source, the lower the CQI and the greater the weight.

[0129] Based on the above relationship, the weighted average method can be used to estimate the location of the interference source. The specific ideas are as follows:

[0130] Base stations with poor signal quality (i.e., lower CQI) may be closer to the interference source, so these base stations should have a greater influence, while base stations with higher CQI (good signal quality and farther away from the interference source) have less influence on the estimated interference source location. The inverse of the base station's CQI value is used as the weight, because the lower the CQI, the worse the signal quality and the closer the base station is to the interference source, and a greater weight should be given. Weight w i The calculation formula is:

[0131] w i =1 / CQI i

[0132] Since the coverage range of a base station is limited, usually a few hundred meters, assume that the number of base stations within a certain range around base station i is n-1. Among these n base stations, the coordinates of each base station are marked as (x i ,y i ), in the coverage area formed by these n base stations, determine the location of the specific interference source (x int ,y int ), as follows:

[0133]

[0134] Since the weight is the inverse of the CQI, the base station with a lower CQI (closer to the interference source) contributes more to the interference source location estimation and has a higher weight. The base station with a higher CQI (farther away from the interference source) has less impact on the location estimation and has a lower weight. Finally, the location estimation of the interference source is obtained by combining the location of each base station and its weight through weighted averaging.

[0135] The location of the interference source is determined through the above steps. After locating the interference source, if the interference intensity of the interference source is large, the base station isolates the interference source by switching channels or adjusting power output, and automatically switches the interfered channel to a low-interference frequency band to ensure signal quality.

[0136] S4. Performing allocation optimization processing of base station resources based on the resource adaptive allocation strategy and the interference isolation strategy.

[0137] In some embodiments, the method of the present application further comprises:

[0138] Obtain current load data and historical load data of the base station;

[0139] Predicting future load data of the base station using a sliding window prediction model according to the current load data and the historical load data;

[0140] Determining a load adjustment strategy for the base station according to the future load data;

[0141] The load adjustment strategy includes user diversion or output power adjustment.

[0142] Specifically, this embodiment ensures efficient use of network resources by performing load prediction and energy efficiency optimization on base stations. The specific steps are as follows:

[0143] (1) Using the sliding window prediction model, the future load L is predicted based on the current and historical load conditions. t+1 , as follows:

[0144]

[0145] Where N is the length of the sliding window, which is used to smooth short-term fluctuations.

[0146] (2) When the load is high, such as greater than a certain threshold, the load adjustment strategy requires the base station to divert some users to low-load base stations in advance to avoid overload. For example, non-real-time traffic users such as video data can be preferentially diverted to neighboring base stations to improve overall load balance.

[0147] (3) The base station dynamically adjusts power output according to user density and load conditions to optimize energy efficiency per unit bandwidth and reduce energy consumption.

[0148] In this application, the geographical location and number of users are crucial to the power regulation of the base station. In areas with dense users, the base station needs higher power to cover more users; while in areas with sparse users, the base station can reduce power output to avoid energy waste. By monitoring user distribution and connection status in real time, the base station can dynamically adjust the power of the coverage area.

[0149] Among them, load refers to the amount of traffic that the base station needs to handle at a certain moment. When the load is high, the base station needs to increase power to ensure the stability of data transmission and avoid affecting the user experience due to network congestion or signal attenuation. On the contrary, when the load is low, the base station can optimize energy efficiency and reduce energy consumption by reducing power output. In the case of uneven user distribution or large fluctuations in network load, simply changing power output often leads to low energy efficiency. Therefore, the base station will dynamically adjust the balance between power and bandwidth according to the current bandwidth demand and power consumption to ensure that each unit of bandwidth can provide the best energy efficiency, that is, to provide sufficient network coverage and data transmission capacity with as little power as possible.

[0150] Specifically, assuming that the power output of base station i is P, the bandwidth allocation is B, and the energy efficiency is E, the energy efficiency can be expressed as:

[0151]

[0152] The goal of the base station is to minimize E, that is, to minimize the power P under a given bandwidth B. That is, to make the energy consumption per unit bandwidth as low as possible. In other words, through reasonable power allocation and bandwidth allocation, the value of E is minimized. By optimizing power output and bandwidth allocation, the base station can ensure efficient energy utilization while meeting the user's connection needs and data transmission needs.

[0153] In wireless communication networks, base stations may experience performance degradation due to overload, network failure, or equipment failure. In this case, if not repaired in time, it may lead to poor user experience or network unavailability. Therefore, the goal of base station self-healing mechanism is to automatically identify problems and adjust resources to maintain or restore normal operation of the network.

[0154] The goal of the self-healing mechanism is that the base station needs to be able to detect performance degradation or signs of failure. For example, users cannot connect to the network, signal quality is reduced, resources are overloaded, etc. Once a problem is detected, the base station can take some automated measures to repair the fault, such as adjusting spectrum resources, reallocating bandwidth, adjusting power, or migrating part of the load to other base stations.

[0155] Based on the above principles, in some embodiments, the method of the present application further includes:

[0156] Obtain real-time fault monitoring data and historical fault data of base stations;

[0157] According to the real-time fault monitoring data and the historical fault data, predicting future fault data of the base station through the LSTM model;

[0158] determining a resource adjustment strategy of the base station according to the future fault data;

[0159] The resource adjustment strategy includes at least one of bandwidth reallocation, power adjustment, spectrum adjustment, and load migration.

[0160] Specifically, the base station needs to monitor and evaluate the network status, user connection quality and other performance indicators in real time. For example:

[0161] Real-time traffic monitoring: The base station monitors the current load situation and assesses whether the traffic exceeds expectations or whether there is network congestion.

[0162] Signal quality monitoring: For example, the signal quality can be judged in real time through indicators such as CQI and SINR (Signal to Interference plus Noise Ratio). If the signal quality in a certain base station area is seriously degraded, it may indicate that there is an interference source or a fault has occurred.

[0163] User experience monitoring: If a large number of users are disconnected, connection fails, or data transmission rate is too low, it indicates that the base station may be faulty or overloaded.

[0164] The core of the base station self-healing mechanism is the ability to predict failures in advance and issue warnings based on historical data. The base station can use the LSTM model to learn and predict time series data such as load fluctuations and changes in signal strength. By training on historical data, the LSTM model is able to identify potential failure modes. For example, if certain indicators of the base station (such as signal quality, number of connections, etc.) fluctuate abnormally, the LSTM model can detect these changes in advance and issue warnings. For example, if it is detected that the load in a certain period will exceed expectations, the base station can issue an early warning and start automatically adjusting resources to avoid failures.

[0165] Once a base station detects a potential problem and confirms a failure or performance degradation, resource self-adjustment will be initiated. The specific resource adjustment methods are as follows:

[0166] (1) Bandwidth reallocation: If a surge in traffic from certain user groups causes the base station to be overloaded, the base station can dynamically adjust bandwidth allocation based on the real-time load. For example, the bandwidth for some non-real-time applications can be reduced to allocate more bandwidth to real-time communication traffic (such as voice or video calls).

[0167] (2) Power adjustment: When the base station detects that the signal quality is decreasing or the interference is increasing, it can adjust the transmission power. For example, in some areas with dense users or places with large interference, the base station can increase the power coverage, and reduce the power in areas with low load to save energy and improve energy efficiency.

[0168] (3) Spectrum adjustment: If an interference source affects a certain frequency band, the base station can dynamically switch to a frequency band with less interference, thereby reducing the impact of interference on the network. Through spectrum management, the base station can optimize the transmission quality of the signal and reduce the degradation of network quality caused by spectrum interference.

[0169] (4) Load Migration: When a base station is overloaded or fails, some users can be migrated to other base stations or small base stations in the vicinity. For example, if the traffic of a base station is close to saturation, the system can guide the adjacent base station to absorb some user traffic.

[0170] In some embodiments, the method of the present application further comprises:

[0171] A fault self-repair strategy is configured for the base station, where the fault self-repair strategy is used to adjust resources when a fault occurs in the base station so that the base station can be restored to a normal working state.

[0172] Specifically, a fault self-repair strategy is configured for the base station in advance. Once a fault or performance degradation is detected, the base station will execute the fault self-repair strategy, for example:

[0173] Automatic repair path: Ensure that the base station returns to normal working condition by adjusting resource configuration. If the problem is still not solved after adjustment, the base station may automatically switch to a backup node or notify the network administrator for manual intervention.

[0174] Restoring user connections: By dynamically allocating resources, users whose connections have been interrupted can resume their connections as quickly as possible, thus ensuring service continuity and user experience.

[0175] In addition, once the self-healing mechanism is activated and resource adjustments are made, the adjustment effect needs to be continuously monitored and real-time feedback needs to be provided. If the problem is found to be resolved, the adjustment status will be continuously monitored and maintained; if the problem is not effectively resolved, further adjustments may be made or more advanced recovery strategies may be enabled.

[0176] This application provides a base station resource allocation optimization method, which realizes intelligent resource scheduling and interference management of mobile communication base stations through deep reinforcement learning, interference analysis, collaborative load and self-healing mechanism. This solution can effectively cope with highly dynamic network environments, improve the resource utilization rate and user experience of base stations through real-time perception and intelligent adjustment, and ensure the stability and high-efficiency operation of the network.

[0177] It should be understood that, although the various steps in the flowcharts in the above-described embodiments are sequentially displayed according to the indications of the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps is not strictly limited in order, and they can be executed in other orders. Moreover, at least a portion of the steps in the figure may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily to be carried out sequentially, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0178] Figure 2 A schematic diagram of a base station resource allocation optimization device provided in an embodiment of the present application, such as Figure 2 As shown, the present application provides a base station resource allocation optimization device, the device comprising:

[0179] A data acquisition module 11, which is configured to acquire performance data of a base station;

[0180] A resource allocation module 12, which is configured to determine a resource adaptive allocation strategy through a deep reinforcement learning model based on the performance data;

[0181] An interference isolation module 13, which is configured to determine the location information of the interference source through a weighted average algorithm according to the performance data, and determine an interference isolation strategy;

[0182] The optimization processing module 14 is configured to perform allocation optimization processing of base station resources based on the resource adaptive allocation strategy and the interference isolation strategy.

[0183] Regarding the limitation on the base station resource allocation optimization device, reference may be made to the limitation on the base station resource allocation optimization method in the above-mentioned embodiments of the present application, and this embodiment will not be repeated here.

[0184] Figure 3 Another schematic diagram of the base station resource allocation optimization device provided in an embodiment of the present application is as follows Figure 3 As shown, in some embodiments, the present application provides a base station resource allocation optimization device, including a memory 22 and a processor 21, the memory stores a computer program, and the processor is configured to run the computer program to execute the base station resource allocation optimization method in the above-mentioned embodiments of the present application.

[0185] The memory is connected to the processor, the memory may be a flash memory or a read-only memory or other memory, and the processor may be a central processing unit or a single-chip microcomputer.

[0186] In some embodiments, the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the base station resource allocation optimization method in the above-mentioned embodiments of the present application is implemented.

[0187] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program modules or other data). Computer-readable storage media include, but are not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable read only memory), flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory), digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0188] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present application, but the present application is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and substance of the present application, and these modifications and improvements are also considered to be within the scope of protection of the present application.

Claims

1. A method for optimizing base station resource allocation, characterized in that: The method comprises: S1. Obtain performance data of the base station; S2. Determine a resource adaptive allocation strategy through a deep reinforcement learning model based on the performance data; S3. Determine the location information of the interference source through a weighted average algorithm according to the performance data, and determine an interference isolation strategy; S4. Performing allocation optimization processing of base station resources based on the resource adaptive allocation strategy and the interference isolation strategy.

2. The base station resource allocation optimization method according to claim 1, characterized in that: The performance data of the base station includes at least one of the following: Channel quality indication CQ I; Signal strength RSS I; User density UE Densi ty; Traffic demandTraffi c Demand; Transmission delay Latency.

3. The base station resource allocation optimization method according to claim 2, characterized in that: S2, including: S21, performing feature extraction and dimensionality reduction processing on the performance data to obtain a feature vector; S22. Determine a resource adaptive allocation strategy through a deep reinforcement learning model according to the feature vector; The resource adaptive allocation strategy includes at least one of bandwidth allocation, power control, and spectrum selection.

4. The base station resource allocation optimization method according to claim 2, characterized in that: S3, including: S31, assigning corresponding weights according to the channel quality indicator CQ I of each base station; S32, determining the location information of the interference source by a weighted average algorithm according to the weight corresponding to each base station; S33, determining an interference isolation strategy according to the location information of the interference source; The interference isolation strategy includes channel switching or adjusting power output.

5. The base station resource allocation optimization method according to claim 1, characterized in that: Also includes: Obtain current load data and historical load data of the base station; Predicting future load data of the base station using a sliding window prediction model according to the current load data and the historical load data; Determining a load adjustment strategy for the base station according to the future load data; The load adjustment strategy includes user diversion or output power adjustment.

6. The base station resource allocation optimization method according to claim 1, characterized in that: Also includes: Obtain real-time fault monitoring data and historical fault data of base stations; According to the real-time fault monitoring data and the historical fault data, future fault data of the base station is predicted by a long short-term memory network LSTM model; determining a resource adjustment strategy of the base station according to the future fault data; The resource adjustment strategy includes at least one of bandwidth reallocation, power adjustment, spectrum adjustment, and load migration.

7. The base station resource allocation optimization method according to claim 6, characterized in that: Also includes: A fault self-repair strategy is configured for the base station, where the fault self-repair strategy is used to adjust resources when a fault occurs in the base station so that the base station can be restored to a normal working state.

8. A base station resource allocation optimization device, characterized in that: The device comprises: A data acquisition module, configured to acquire performance data of a base station; A resource allocation module, configured to determine a resource adaptive allocation strategy through a deep reinforcement learning model based on the performance data; An interference isolation module, configured to determine the location information of the interference source through a weighted average algorithm according to the performance data, and determine an interference isolation strategy; The optimization processing module is configured to perform allocation optimization processing of base station resources based on the resource adaptive allocation strategy and the interference isolation strategy.

9. A base station resource allocation optimization device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement the base station resource allocation optimization method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the base station resource allocation optimization method according to any one of claims 1 to 7 is implemented.

Images