Wireless communication network scheduling method and system based on artificial intelligence

By adopting artificial intelligence scheduling methods in wireless communication networks, combining roulette selection and dynamic Bayesian networks, the problem of taking into account load balancing and user needs and prediction of future network performance changes is solved, and the stability and user experience of the network are improved. .

CN118488587BActive Publication Date: 2024-11-12BEIJING CHINACOMM DESIGN & CONSULTING CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410932486.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-12
Publication Date
2024-11-12
Estimated Expiration
2044-07-12

AI Technical Summary

Technical Problem

The existing wireless communication network scheduling methods are difficult to take into account load balancing and user needs, resulting in a lack of user experience, a decline in network service quality and satisfaction, and future network performance changes are difficult to accurately predict, and emergencies cannot be dealt with in a timely manner.

Method used

Using an artificial intelligence-based method, by clarifying that computing resource scheduling is based on user preferences, it is carried out in two stages, and using the roulette selection algorithm to select nodes, combined with the construction of a dynamic Bayesian network for parameter estimation, predict future network performance changes, and enhance network Stability and reliability.

Benefits of technology

It realizes load balancing and satisfying user preferences, improves network service quality and user experience, enhances the ability to respond to emergencies, and improves the stability and reliability of the network.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118488587B_ABST
    Figure CN118488587B_ABST
Patent Text Reader

Abstract

The present invention discloses a wireless communication network scheduling method and system based on artificial intelligence, and the method includes: wireless communication network data collection, network scheduling, network parameter prediction, inter-regional collaborative scheduling and performance monitoring. The present invention belongs to the field of network scheduling technology, and specifically refers to a wireless communication network scheduling method and system based on artificial intelligence. This scheme clarifies that computing resource scheduling is based on user preferences, and different preferences have different scheduling strategies, which are carried out in two stages. The calculation of the comprehensive score is used to reflect the importance of the node in resource allocation scheduling, and the roulette wheel selection algorithm is used to select computing resource nodes to balance the load and meet user preferences; a dynamic Bayesian network is constructed to perform parameter estimation, and the dynamic Bayesian network parameters are estimated using historical data to predict future network performance change trends, enhance foresight, improve the stability and reliability of the network, and enhance the ability to respond to emergencies.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of network scheduling, and in particular to a wireless communication network scheduling method and system based on artificial intelligence. Background Art

[0002] Wireless communication network scheduling is a method that uses advanced algorithms and strategies to maximize the use of network resources and optimize network performance. It aims to reasonably allocate network resources, improve the efficiency, capacity and service quality of wireless networks, and meet the communication needs of different users in different scenarios. However, the existing wireless communication network scheduling method has technical problems such as different user needs, the difficulty of existing technologies in balancing load and user needs, poor user experience, and reduced quality and satisfaction of network services; and the difficulty of accurately predicting future changes in network performance and the inability to schedule in time to deal with potential emergencies. Summary of the invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a wireless communication network scheduling method and system based on artificial intelligence. In view of the technical problems that different user needs exist, the prior art is difficult to balance load balancing and user needs, the user experience is insufficient, and the quality and satisfaction of network services are reduced, a clear computing resource scheduling is adopted based on user preferences, and different preferences have different scheduling strategies, which are carried out in two stages. The comprehensive score is calculated to reflect the importance of the node in resource allocation and scheduling, and a roulette wheel selection algorithm is used to select computing resource nodes to balance the load and meet user preferences; in view of the technical problem that future network performance changes are difficult to accurately predict and scheduling cannot be performed in time to deal with potential emergencies, a dynamic Bayesian network is constructed to estimate parameters, and historical data is used to estimate the dynamic Bayesian network parameters. The future network performance change trend is predicted based on the current network status, foresight is enhanced, the stability and reliability of the network are improved, and the ability to respond to emergencies is enhanced.

[0004] The technical solution adopted by the present invention is as follows: The wireless communication network scheduling method based on artificial intelligence provided by the present invention comprises the following steps:

[0005] Step S1: wireless communication network data collection, specifically collecting network status parameters, network load information, user location information and historical traffic pattern information;

[0006] Step S2: Perform network scheduling, specifically to clarify that computing resource scheduling is based on user preferences and different preferences have different scheduling strategies, which is carried out in two stages. The comprehensive score is calculated to reflect the importance of the node in resource allocation scheduling, and the roulette wheel selection algorithm is used to select computing resource nodes to balance the load and meet user preferences;

[0007] Step S3: network parameter prediction, specifically, constructing a dynamic Bayesian network, performing parameter estimation, estimating dynamic Bayesian network parameters using historical data, predicting future network performance change trends based on the current network status, and improving network stability and reliability;

[0008] Step S4: inter-regional collaborative scheduling, specifically, information interaction and sharing between nodes in different regions, control entities to negotiate, and conduct traffic diversion;

[0009] Step S5: Performance monitoring, specifically, continuously monitoring the effect of wireless communication network scheduling and evaluating performance indicators.

[0010] Furthermore, in step S1, the wireless communication network data collection includes the following steps:

[0011] Step S11: Collect network status parameters, including signal strength, bandwidth and delay information, and perform data cleaning and integration;

[0012] Step S12: Collect network load information, including the amount of data being transmitted in the current network and the number of active users;

[0013] Step S13: Collect user location information, understand the distribution of users, and perform targeted scheduling optimization for different areas;

[0014] Step S14: Collect historical traffic pattern information, and infer future traffic trends by analyzing changes in network traffic over a period of time.

[0015] Further, in step S2, the network scheduling comprises the following steps:

[0016] Step S21: Implement the scheduling model in stages, clarify that the computing resource scheduling in the scheduling model is based on user preferences, different preferences correspond to different scheduling strategies, and the execution is divided into two stages;

[0017] Step S22: The first stage of the implementation model is used to set different priority indicators to make the evaluation more targeted and reasonable. Based on the resource status information, each computing resource node is evaluated using a node scoring algorithm with corresponding preferences, and different priority indicators are set for different preferences.

[0018] Step S23: Implement the second stage of the model to ensure that the selected nodes are more in line with the actual needs of users with different preferences. The target computing resource node is selected according to the scoring results. Users with different network resource request preferences pay attention to different status parameters of the computing resource node.

[0019] Step S24: Calculate the comprehensive score to reflect the importance of the node in resource allocation scheduling. The scheduling model designs a specific node scoring algorithm for various resource request preferences and preference combinations. The input parameters are the corresponding priority indicators. Based on the standardized data matrix, the comprehensive score is calculated. The formula used is as follows:

[0020] ;

[0021] ;

[0022] Where n represents the total number of data points, which are used to determine how to allocate and schedule resources to the corresponding nodes, i represents the index of the data point, represents the normalized data matrix of the jth node, P ij represents the value associated with the i-th data point at the j-th node, which is used to evaluate the relationship between the data point and the node. m represents the total number of nodes. E k represents the normalized data matrix of the kth node, w j Represents the overall comprehensive score of the jth node;

[0023] Step S25: Node selection is performed. Node selection is performed in the candidate computing resource node set according to the scoring results. Since the greedy algorithm may aggravate the resource competition problem and the random selection algorithm cannot meet the user preference, the roulette wheel selection algorithm is used to select computing resource nodes. The computing resource nodes with scores from high to low are randomly selected in descending order of probability to balance the load and meet the user preference. The formula used is as follows:

[0024] ;

[0025] ;

[0026] ;

[0027] Where r is a random number uniformly distributed in the interval [0, 1], which is used to determine the final selected node. Represents the total number of nodes, It represents the probability of the i-th data point being selected, which is directly proportional to the comprehensive score of the node. q represents the accumulation of the probabilities of each node. Indicates the selected computing resource node.

[0028] Furthermore, in step S3, the network parameter prediction comprises the following steps:

[0029] Step S31: data collection, collecting historical network parameter data from wireless communication network data, including network traffic data, channel quality parameters and network congestion conditions;

[0030] Step S32: construct dynamic features, extract features from data, construct dynamic features, construct historical network parameter detection results into a time series, and perform moving average feature extraction;

[0031] Step S33: constructing a dynamic Bayesian network, using the dynamic Bayesian network model to represent the dynamic change process of network parameters, including the following steps:

[0032] Step S331: determine the network structure, determine the nodes and connection relationships, each node represents a network performance related variable, the variables include directly observable variables and non-directly observable variables, determine the parent node of each node, the parent node reflects the temporal relationship between the variables;

[0033] Step S332: define the transition probability and observation probability. The transition probability represents the probability distribution of the node under the given parent node condition, and describes the degree to which the network parameter variable is affected by the parent node. The observation probability represents the probability of estimating the future observation variable according to the value of the observation variable under the given network structure and node probability. The observation probability is used to help calculate the probability of the node.

[0034] Step S333: Calculate the conditional probability of the node, using the following formula:

[0035] ;

[0036] In the formula, is the conditional probability of the node, represents the ath node, a is the index of the node, the node includes hidden nodes and observation nodes, N is the total number of nodes, Is a node The parent node of is the transition probability of the node;

[0037] Step S334: describes the process of calculating the probability distribution of time series nodes in the dynamic Bayesian network, taking into account the probability calculation method of each node at different time steps, and calculating the node probability distribution. The formula used is as follows:

[0038] ;

[0039] In the formula, represents the joint probability distribution of nodes in a dynamic Bayesian network, that is, the joint probability when there are N nodes in time steps 1 to T, represents the initial network structure, T represents the time step, represents the network structure that changes over time, Indicates that in the first time step, the probability of each node is calculated. It means that at each time step from time step 2 to T, a multiplication operation is performed on each node, and the probability of each node is calculated based on the parent node of the node;

[0040] Step S34: Parameter estimation, using historical data related to network parameters to estimate the parameters in the dynamic Bayesian network, i.e., the transition probability and conditional probability distribution between nodes;

[0041] Step S35: Predict the trend of network performance changes. Based on the current network status, predict the network status at future times, evaluate the network performance changes in advance, provide a reference for network optimization decisions, and help improve the stability and reliability of the network.

[0042] Furthermore, in step S4, the inter-region coordinated scheduling includes the following steps:

[0043] Step S41: Information interaction and sharing: nodes in different areas transmit and share information, including network load, resource usage, and user distribution;

[0044] Step S42: Inter-regional negotiation: the control entities of each region negotiate to determine resource allocation and coordination strategies, consider how to balance the load and improve the overall performance, and dynamically allocate and adjust the frequency, time, and power resources between regions based on the negotiation results and real-time conditions;

[0045] Step S43: Traffic diversion: according to the load conditions of the area, user traffic is diverted reasonably to avoid congestion in some areas while other areas are idle.

[0046] Furthermore, in step S5, the performance monitoring is specifically to continuously monitor the effect of wireless communication network scheduling, evaluate whether the performance indicators are improved, and adaptively modify the system according to the monitoring results to better adapt to the dynamically changing network environment.

[0047] The wireless communication network scheduling system based on artificial intelligence provided by the present invention includes a wireless communication network data collection module, a network scheduling module, a network parameter prediction module, an inter-regional collaborative scheduling module and a performance monitoring module;

[0048] The wireless communication network data collection module specifically collects network status parameters, network load information, user location information and historical traffic pattern information;

[0049] The network scheduling module specifically clarifies that computing resource scheduling is based on user preferences and different preferences have different scheduling strategies. It is carried out in two stages, and a comprehensive score is calculated to reflect the importance of the node in resource allocation scheduling. A roulette wheel selection algorithm is used to select computing resource nodes to balance the load and meet user preferences.

[0050] The network parameter prediction module specifically constructs a dynamic Bayesian network to perform parameter estimation, estimates the dynamic Bayesian network parameters using historical data, predicts future network performance change trends based on the current network status, and improves network stability and reliability;

[0051] The inter-regional collaborative scheduling module is specifically used to perform information interaction and sharing between nodes in different regions, control entities to negotiate, and conduct traffic diversion;

[0052] The performance monitoring module is specifically used to continuously monitor the effect of wireless communication network scheduling and evaluate performance indicators.

[0053] The beneficial results achieved by the present invention using the above scheme are as follows:

[0054] (1) In view of the technical problems that different user needs exist, the existing technology is difficult to balance load balancing and user needs, the user experience is poor, and the quality and satisfaction of network services are reduced, a clear computing resource scheduling is adopted based on user preferences, and different preferences have different scheduling strategies. It is carried out in two stages. The comprehensive score is calculated to reflect the importance of the node in resource allocation and scheduling. The roulette wheel selection algorithm is used to select computing resource nodes to balance the load and meet user preferences;

[0055] (2) In order to address the technical problem that future changes in network performance are difficult to accurately predict and that scheduling cannot be carried out in a timely manner to deal with potential emergencies, a dynamic Bayesian network is constructed to estimate parameters. The dynamic Bayesian network parameters are estimated using historical data. Based on the current network status, the future trend of network performance changes is predicted, which enhances foresight, improves network stability and reliability, and enhances the ability to respond to emergencies. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A schematic diagram of a flow chart of a wireless communication network scheduling method based on artificial intelligence provided by the present invention;

[0057] Figure 2 A schematic diagram of a wireless communication network scheduling system based on artificial intelligence provided by the present invention;

[0058] Figure 3 is a schematic flow chart of step S2;

[0059] Figure 4 It is a schematic diagram of the process of step S3.

[0060] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0061] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0062] In the description of the present invention, it should be understood that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.

[0063] Example 1, see Figure 1 The present invention provides a wireless communication network scheduling method based on artificial intelligence, which comprises the following steps:

[0064] Step S1: wireless communication network data collection, specifically collecting network status parameters, network load information, user location information and historical traffic pattern information;

[0065] Step S2: Perform network scheduling, specifically to clarify that computing resource scheduling is based on user preferences and different preferences have different scheduling strategies, which is carried out in two stages. The comprehensive score is calculated to reflect the importance of the node in resource allocation scheduling, and the roulette wheel selection algorithm is used to select computing resource nodes to balance the load and meet user preferences;

[0066] Step S3: network parameter prediction, specifically, constructing a dynamic Bayesian network, performing parameter estimation, estimating dynamic Bayesian network parameters using historical data, predicting future network performance change trends based on the current network status, and improving network stability and reliability;

[0067] Step S4: inter-regional collaborative scheduling, specifically, information interaction and sharing between nodes in different regions, control entities to negotiate, and conduct traffic diversion;

[0068] Step S5: Performance monitoring, specifically, continuously monitoring the effect of wireless communication network scheduling and evaluating performance indicators.

[0069] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S1, the wireless communication network data collection includes the following steps:

[0070] Step S11: Collect network status parameters, including signal strength, bandwidth and delay information, and perform data cleaning and integration;

[0071] Step S12: Collect network load information, including the amount of data being transmitted in the current network and the number of active users;

[0072] Step S13: Collect user location information, understand the distribution of users, and perform targeted scheduling optimization for different areas;

[0073] Step S14: Collect historical traffic pattern information, and infer future traffic trends by analyzing changes in network traffic over a period of time.

[0074] Example 3, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S2, the network scheduling includes the following steps:

[0075] Step S21: Implement the scheduling model in stages, clarify that the computing resource scheduling in the scheduling model is based on user preferences, different preferences correspond to different scheduling strategies, and the execution is divided into two stages;

[0076] Step S22: The first stage of the implementation model is used to set different priority indicators to make the evaluation more targeted and reasonable. Based on the resource status information, each computing resource node is evaluated using a node scoring algorithm with corresponding preferences, and different priority indicators are set for different preferences.

[0077] Step S23: Implement the second stage of the model to ensure that the selected nodes are more in line with the actual needs of users with different preferences. The target computing resource node is selected according to the scoring results. Users with different network resource request preferences pay attention to different status parameters of the computing resource node.

[0078] Step S24: Calculate the comprehensive score to reflect the importance of the node in resource allocation scheduling. The scheduling model designs a specific node scoring algorithm for various resource request preferences and preference combinations. The input parameters are the corresponding priority indicators. Based on the standardized data matrix, the comprehensive score is calculated. The formula used is as follows:

[0079] ;

[0080] ;

[0081] Where n represents the total number of data points, which are used to determine how to allocate and schedule resources to the corresponding nodes, i represents the index of the data point, represents the normalized data matrix of the jth node, P ijrepresents the value associated with the i-th data point at the j-th node, which is used to evaluate the relationship between the data point and the node. m represents the total number of nodes. E k represents the normalized data matrix of the kth node, w j Represents the overall comprehensive score of the jth node;

[0082] Step S25: Node selection is performed. Node selection is performed in the candidate computing resource node set according to the scoring results. Since the greedy algorithm may aggravate the resource competition problem and the random selection algorithm cannot meet the user preference, the roulette wheel selection algorithm is used to select computing resource nodes. The computing resource nodes with scores from high to low are randomly selected in descending order of probability to balance the load and meet the user preference. The formula used is as follows:

[0083] ;

[0084] ;

[0085] ;

[0086] Where r is a random number uniformly distributed in the interval [0, 1], which is used to determine the final selected node. Represents the total number of nodes, It represents the probability of the i-th data point being selected, which is directly proportional to the comprehensive score of the node. q represents the accumulation of the probabilities of each node. Indicates the selected computing resource node.

[0087] By performing the above operations, a clear computing resource scheduling is adopted based on user preferences and different preferences have different scheduling strategies. It is carried out in two stages. The comprehensive score is calculated to reflect the importance of the node in resource allocation scheduling. The roulette wheel selection algorithm is used to select computing resource nodes to balance the load and meet user preferences. The technical problems of different user needs, the difficulty of existing technologies in balancing load balancing and user needs, the lack of user experience, and the decline in the quality and satisfaction of network services are solved.

[0088] Example 4, see Figure 1 and Figure 4 This embodiment is based on the above embodiment. In step S3, the network parameter prediction includes the following steps:

[0089] Step S31: data collection, collecting historical network parameter data from wireless communication network data, including network traffic data, channel quality parameters and network congestion conditions;

[0090] Step S32: construct dynamic features, extract features from data, construct dynamic features, construct historical network parameter detection results into a time series, and perform moving average feature extraction;

[0091] Step S33: constructing a dynamic Bayesian network, using the dynamic Bayesian network model to represent the dynamic change process of network parameters, including the following steps:

[0092] Step S331: determine the network structure, determine the nodes and connection relationships, each node represents a network performance related variable, the variables include directly observable variables and non-directly observable variables, determine the parent node of each node, the parent node reflects the temporal relationship between the variables;

[0093] Step S332: define the transition probability and observation probability. The transition probability represents the probability distribution of the node under the given parent node condition, and describes the degree to which the network parameter variable is affected by the parent node. The observation probability represents the probability of estimating the future observation variable according to the value of the observation variable under the given network structure and node probability. The observation probability is used to help calculate the probability of the node.

[0094] Step S333: Calculate the conditional probability of the node, using the following formula:

[0095] ;

[0096] In the formula, is the conditional probability of the node, represents the ath node, a is the index of the node, the node includes hidden nodes and observation nodes, N is the total number of nodes, Is a node The parent node of is the transition probability of the node;

[0097] Step S334: describes the process of calculating the probability distribution of time series nodes in the dynamic Bayesian network, taking into account the probability calculation method of each node at different time steps, and calculating the node probability distribution. The formula used is as follows:

[0098] ;

[0099] In the formula, represents the joint probability distribution of nodes in a dynamic Bayesian network, that is, the joint probability when there are N nodes in time steps 1 to T, represents the initial network structure, T represents the time step, represents the network structure that changes over time, Indicates that in the first time step, the probability of each node is calculated. It means that at each time step from time step 2 to T, a multiplication operation is performed on each node, and the probability of each node is calculated based on the parent node of the node;

[0100] Step S34: Parameter estimation, using historical data related to network parameters to estimate the parameters in the dynamic Bayesian network, i.e., the transition probability and conditional probability distribution between nodes;

[0101] Step S35: Predict the trend of network performance changes. Based on the current network status, predict the network status at future times, evaluate the network performance changes in advance, provide a reference for network optimization decisions, and help improve the stability and reliability of the network.

[0102] By performing the above operations, a dynamic Bayesian network is constructed to perform parameter estimation, and historical data is used to estimate the parameters of the dynamic Bayesian network. Based on the current network status, the future trend of network performance changes is predicted, foresight is enhanced, the stability and reliability of the network are improved, and the ability to respond to emergencies is improved. This solves the technical problem that future network performance changes are difficult to accurately predict and timely scheduling cannot be carried out to deal with potential emergencies.

[0103] Example 5, see Figure 1 This embodiment is based on the above embodiment. In step S4, the inter-area coordinated scheduling includes the following steps:

[0104] Step S41: Information interaction and sharing: nodes in different areas transmit and share information, including network load, resource usage, and user distribution;

[0105] Step S42: Inter-regional negotiation: the control entities of each region negotiate to determine resource allocation and coordination strategies, consider how to balance the load and improve the overall performance, and dynamically allocate and adjust the frequency, time, and power resources between regions based on the negotiation results and real-time conditions;

[0106] Step S43: Traffic diversion: according to the load conditions of the area, user traffic is diverted reasonably to avoid congestion in some areas while other areas are idle.

[0107] Example 6, see Figure 1 This embodiment is based on the above embodiment. In step S5, the performance monitoring is specifically to continuously monitor the effect of wireless communication network scheduling, evaluate whether the performance indicators are improved, and adaptively modify the system according to the monitoring results to better adapt to the dynamically changing network environment.

[0108] Embodiment 7, see Figure 2 , this embodiment is based on the above embodiment. The wireless communication network scheduling system based on artificial intelligence provided by the present invention includes a wireless communication network data collection module, a network scheduling module, a network parameter prediction module, an inter-regional collaborative scheduling module and a performance monitoring module;

[0109] The wireless communication network data collection module specifically collects network status parameters, network load information, user location information and historical traffic pattern information;

[0110] The network scheduling module specifically clarifies that computing resource scheduling is based on user preferences and different preferences have different scheduling strategies. It is carried out in two stages, and a comprehensive score is calculated to reflect the importance of the node in resource allocation scheduling. A roulette wheel selection algorithm is used to select computing resource nodes to balance the load and meet user preferences.

[0111] The network parameter prediction module specifically constructs a dynamic Bayesian network to perform parameter estimation, estimates the dynamic Bayesian network parameters using historical data, predicts future network performance change trends based on the current network status, and improves network stability and reliability;

[0112] The inter-regional collaborative scheduling module is specifically used to perform information interaction and sharing between nodes in different regions, control entities to negotiate, and conduct traffic diversion;

[0113] The performance monitoring module is specifically used to continuously monitor the effect of wireless communication network scheduling and evaluate performance indicators.

[0114] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0115] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

[0116] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A wireless communication network scheduling method based on artificial intelligence, characterized in that: The method comprises the following steps: Step S1: wireless communication network data collection; Step S2: Perform network scheduling, specifically to clarify that computing resource scheduling is based on user preferences and different preferences have different scheduling strategies, which is carried out in two stages, and calculate the comprehensive score to reflect the importance of the node in resource allocation scheduling, and use the roulette wheel selection algorithm to select computing resource nodes; Step S3: network parameter prediction, specifically, constructing a dynamic Bayesian network, performing parameter estimation, estimating dynamic Bayesian network parameters using historical data, and predicting future network performance change trends based on the current network status; Step S4: inter-regional coordinated scheduling; Step S5: performance monitoring; In step S2, the network scheduling includes the following steps: Step S21: Implement the scheduling model in stages, clarify that the computing resource scheduling in the scheduling model is based on user preferences, different preferences correspond to different scheduling strategies, and the execution is divided into two stages; Step S22: In the first stage of the implementation model, each computing resource node is evaluated using a node scoring algorithm with a corresponding preference based on the resource status information, and different preferences are set with different priority indicators; Step S23: implementing the second stage of the model, selecting the target computing resource node according to the scoring results, and users with different resource request preferences pay attention to different status parameters of the computing resource node; Step S24: Calculate the comprehensive score to reflect the importance of the node in resource allocation scheduling. The scheduling model designs a specific node scoring algorithm for various resource request preferences and preference combinations. The input parameters are the corresponding priority indicators. Based on the standardized data matrix, the comprehensive score is calculated. The formula used is as follows: ; ; Where n represents the total number of data points, which are used to determine how to allocate and schedule resources to the corresponding nodes, i represents the index of the data point, represents the normalized data matrix of the jth node, P ij represents the value associated with the i-th data point at the j-th node, which is used to evaluate the relationship between the data point and the node. m represents the total number of nodes. E k represents the normalized data matrix of the kth node, w j Represents the overall comprehensive score of the jth node; Step S25: Node selection is performed. Node selection is performed in the candidate computing resource node set according to the scoring results. The roulette wheel selection algorithm is used to select computing resource nodes. The computing resource nodes with scores from high to low are randomly selected in descending order of probability. The formula used is as follows: ; ; ; Where r is a random number uniformly distributed in the interval [0, 1], which is used to determine the final selected node. Represents the total number of nodes, It represents the probability of the i-th data point being selected, which is directly proportional to the comprehensive score of the node. q represents the accumulation of the probabilities of each node. Indicates the selected computing resource node; In step S3, the network parameter prediction includes the following steps: Step S31: data collection, collecting historical network parameter data from wireless communication network data, including network traffic data, channel quality parameters and network congestion conditions; Step S32: construct dynamic features, extract features from data, construct dynamic features, construct historical network parameter detection results into a time series, and perform moving average feature extraction; Step S33: constructing a dynamic Bayesian network, using the dynamic Bayesian network model to represent the dynamic change process of network parameters, including the following steps: Step S331: determine the network structure, determine the nodes and connection relationships, each node represents a network performance related variable, the variables include directly observable variables and non-directly observable variables, determine the parent node of each node, the parent node reflects the temporal relationship between the variables; Step S332: define the transition probability and observation probability. The transition probability represents the probability distribution of the node under the given parent node condition, and describes the degree to which the network parameter variable is affected by the parent node. The observation probability represents the probability of estimating the future observation variable according to the value of the observation variable under the given network structure and node probability. The observation probability is used to help calculate the probability of the node. Step S333: Calculate the conditional probability of the node, using the following formula: ; In the formula, is the conditional probability of the node, represents the ath node, a is the index of the node, the node includes hidden nodes and observation nodes, N is the total number of nodes, Is a node The parent node of is the transition probability of the node; Step S334: describes the process of calculating the probability distribution of time series nodes in the dynamic Bayesian network, taking into account the probability calculation method of each node at different time steps, and calculating the node probability distribution. The formula used is as follows: ; In the formula, represents the joint probability distribution of nodes in a dynamic Bayesian network, that is, the joint probability when there are N nodes in time steps 1 to T, represents the initial network structure, T represents the time step, represents the network structure that changes over time, Indicates that in the first time step, the probability of each node is calculated. It means that at each time step from time step 2 to T, a multiplication operation is performed on each node, and the probability of each node is calculated based on the parent node of the node; Step S34: Parameter estimation, using historical data related to network parameters to estimate the parameters in the dynamic Bayesian network, i.e., the transition probability and conditional probability distribution between nodes; Step S35: predict the trend of network performance changes, based on the current network status, predict the network status at future times, evaluate the network performance changes in advance, and provide a reference for network optimization decisions.

2. The wireless communication network scheduling method based on artificial intelligence according to claim 1, characterized in that: In step S1, the wireless communication network data collection includes the following steps: Step S11: Collect network status parameters, including signal strength, bandwidth and delay information, and perform data cleaning and integration; Step S12: Collect network load information, including the amount of data being transmitted in the current network and the number of active users; Step S13: Collect user location information, understand the distribution of users, and perform targeted scheduling optimization for different areas; Step S14: Collect historical traffic pattern information, and infer future traffic trends by analyzing changes in network traffic over a period of time.

3. The wireless communication network scheduling method based on artificial intelligence according to claim 1, characterized in that: In step S4, the inter-region coordinated scheduling includes the following steps: Step S41: Information interaction and sharing: nodes in different areas transmit and share information, including network load, resource usage, and user distribution; Step S42: Inter-regional negotiation: the control entities of each region negotiate to determine resource allocation and coordination strategies, and dynamically allocate and adjust frequency, time, and power resources between regions based on the negotiation results and real-time conditions; Step S43: Traffic diversion: reasonably divert user traffic according to the load conditions of the area.

4. The wireless communication network scheduling method based on artificial intelligence according to claim 1, characterized in that: In step S5, the performance monitoring specifically includes continuously monitoring the effect of wireless communication network scheduling, and adaptively modifying the system according to the monitoring result to better adapt to the dynamically changing network environment.

5. A wireless communication network scheduling system based on artificial intelligence, used to implement the wireless communication network scheduling method based on artificial intelligence as described in any one of claims 1 to 4, characterized in that: It includes a wireless communication network data collection module, a network scheduling module, a network parameter prediction module, an inter-regional collaborative scheduling module and a performance monitoring module.

6. The wireless communication network scheduling system based on artificial intelligence according to claim 5, characterized in that: The wireless communication network data collection module specifically collects network status parameters, network load information, user location information and historical traffic pattern information; The network scheduling module specifically clarifies that computing resource scheduling is based on user preferences and different preferences have different scheduling strategies. It is carried out in two stages, and a comprehensive score is calculated to reflect the importance of the node in resource allocation scheduling. A roulette wheel selection algorithm is used to select computing resource nodes to balance the load and meet user preferences. The network parameter prediction module specifically constructs a dynamic Bayesian network to perform parameter estimation, estimates the dynamic Bayesian network parameters using historical data, and predicts future network performance change trends based on the current network status; The inter-regional collaborative scheduling module is specifically used to perform information interaction and sharing between nodes in different regions, control entities to negotiate, and conduct traffic diversion; The performance monitoring module is specifically used to continuously monitor the effect of wireless communication network scheduling and evaluate performance indicators.

Citation Information

Cited By

  • Communication scheduling network management intelligent optimization system and method based on AI dynamic decision

    CN120547039A