A method and system for allocating bandwidth in a wireless communication network

By dividing service levels in wireless communication networks and using SVM models and artificial bee colony algorithms to predict and allocate bandwidth, the problem of bandwidth resource waste is solved, and efficient bandwidth management of wireless communication networks is realized.

CN118632372BActive Publication Date: 2024-12-03BEIJING CHINACOMM DESIGN & CONSULTING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411097140.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-12
Publication Date
2024-12-03
Estimated Expiration
2044-08-12

AI Technical Summary

Technical Problem

The bandwidth allocation method in existing wireless communication networks is difficult to effectively allocate and recover unused bandwidth resources, resulting in waste of resources.

Method used

By dividing wireless communication network users according to service levels, using SVM models to predict real-time bandwidth requirements, and using artificial bee colony algorithm to set the optimal threshold bandwidth, and allocating bandwidth in sequence according to service levels.

Benefits of technology

It realizes the effective allocation of bandwidth resources of wireless communication networks, meets user needs, avoids additional consumption of bandwidth resources, and improves allocation efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118632372B_ABST
    Figure CN118632372B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of wireless communication technology, and in particular to a method and system for allocating bandwidth in a wireless communication network. The system comprises a data acquisition module, a final SVM model construction module, a wireless communication network user service level classification module, an optimal threshold bandwidth setting module, a wireless communication network user real-time required bandwidth prediction module, and a wireless communication user bandwidth allocation module. The wireless communication network users are classified according to the corresponding wireless communication network user service levels, the optimal threshold bandwidth is allocated to each service level, and the final SVM model obtained by training is used to predict the real-time required bandwidth of the wireless communication network users. Bandwidth is allocated to all wireless communication network users according to the wireless communication network user service levels, thereby ensuring the network needs of users and avoiding unnecessary consumption of bandwidth resources.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a bandwidth allocation method and system in a wireless communication network. Background Art

[0002] With the rapid development of Internet technology, the number of wireless communication network users is also growing rapidly. Therefore, it is particularly important to ensure user experience while allocating wireless communication network resources reasonably and effectively.

[0003] The existing technology often receives network usage information fed back to the system by wireless communication network users, calculates the bandwidth required by the wireless communication network users based on the network usage information and the wireless communication resource allocation technology, and allocates the required bandwidth to each wireless communication network user based on the bandwidth requirements of all wireless communication network users, which easily causes unnecessary consumption of bandwidth resources and makes it impossible to recover unused bandwidth resources in a timely manner. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] To address the deficiencies in the background technology, the present invention proposes a bandwidth allocation method and system in a wireless communication network, which divides the wireless communication network users according to the corresponding wireless communication network user service levels, allocates the optimal threshold bandwidth to each service level, and predicts the real-time bandwidth required by the wireless communication network users through the final SVM model obtained through training, and allocates bandwidth to all wireless communication network users according to the wireless communication network user service levels.

[0006] (II) Technical solution

[0007] A method for allocating bandwidth in a wireless communication network, the method comprising the following steps:

[0008] S1. Setting a wireless communication network user set and a transmission data volume collection period set, collecting the transmission data volume of each wireless communication network user in the wireless communication network user set in each transmission data volume collection period in the transmission data volume collection period set, and obtaining a transmission data volume matrix;

[0009] S2, constructing an initial SVM model, a transmission data volume training matrix, and a transmission data volume test matrix, training and testing the initial SVM model through the transmission data volume training matrix and the transmission data volume test matrix to obtain a final SVM model;

[0010] S3, setting a wireless communication network user service level set, dividing each wireless communication network user in the wireless communication network user set according to the corresponding wireless communication network user service level, and obtaining a wireless communication network user classification matrix;

[0011] S4. Setting an optimal threshold bandwidth for each wireless communication network user service level in the wireless communication network user service level set by using an intelligent optimization algorithm to obtain an optimal threshold bandwidth set;

[0012] S5, collecting the transmission data volume of each wireless communication network user in the wireless communication network user classification matrix in real time, obtaining a real-time transmission data volume matrix, setting a basic energy consumption transmission data volume threshold to correct the real-time transmission data volume matrix, obtaining a real-time transmission data volume correction matrix, and predicting the real-time required bandwidth of the wireless communication network user through the final SVM model;

[0013] S6. Allocate bandwidth to each wireless communication user in descending order according to the service level of the wireless communication network user.

[0014] The initial SVM model is trained and tested by collecting a large amount of data to obtain the final SVM model, thereby ensuring the accuracy of the final SVM model and providing a prediction tool for subsequent operations; the wireless communication network users are divided according to the corresponding wireless communication network user service levels, and the optimal threshold bandwidth is allocated to each service level to meet the basic bandwidth requirements, and the order of bandwidth allocation is determined at the same time; according to the predicted user real-time required bandwidth matrix, bandwidth is allocated to all wireless communication network users according to the wireless communication network user service level, thereby realizing the effective allocation of wireless communication network bandwidth resources.

[0015] Preferably, the specific steps of setting a wireless communication network user set and a transmission data volume collection period set, collecting the transmission data volume of each wireless communication network user in the wireless communication network user set according to the transmission data volume collection period set, and obtaining the transmission data volume matrix are as follows:

[0016] S11. Setting the wireless communication network user set ,in, Indicates Wireless communication network users, Indicates the total number of wireless communication network users;

[0017] S12, set the transmission data volume collection cycle set ,in, Indicates The transmission data volume collection cycle, Indicates the total number of transmission data volume collection cycles;

[0018] S13. Collecting wireless communication network user data Each wireless communication network user in the transmission data volume collection cycle set The transmission data volume of each transmission data volume collection cycle in the transmission data volume matrix is ​​obtained as follows:

[0019]

[0020] in, Indicates Wireless communication network users in The amount of data transmitted during a transmission data collection cycle.

[0021] By setting a transmission data volume collection cycle to collect the transmission data volume of each wireless communication network user, the universality of the collected data is ensured, providing a data basis for subsequent operations.

[0022] Preferably, the specific steps of constructing an initial SVM model, a transmission data volume training matrix, and a transmission data volume test matrix, training and testing the initial SVM model through the transmission data volume training matrix and the transmission data volume test matrix, and obtaining the final SVM model are as follows:

[0023] S21. Build the initial SVM model and set the penalty factor parameter to , the radial basis kernel parameters are And the kernel function of the initial parameter space during the operation of the initial SVM model is a Gaussian radial basis kernel function; the formula of the Gaussian radial basis kernel function is as follows:

[0024]

[0025] in, represents the Gaussian radial basis kernel function, represents the support vector corresponding to two training samples, express and The Euclidean distance between

[0026] S22. Set the training data ratio And test data ratio , dividing the transmission data volume matrix by the training data ratio and the test data ratio, and obtaining a transmission data volume training matrix and a transmission data volume test matrix respectively;

[0027] S23, set the training error threshold to , input the transmission data volume training matrix into the initial SVM model to train the initial SVM model until the training error of the initial SVM model is less than or equal to When , stop training and get the trained SVM model;

[0028] S24. Setting the test accuracy threshold , input the transmission data volume test matrix into the trained SVM model for testing, obtain the real-time bandwidth required by the wireless communication network user, and calculate the accuracy of the test result , when testing accuracy Greater than or equal to the test accuracy threshold If the training SVM model is , the trained SVM model is used as the final SVM model; otherwise, the grid search algorithm is used to search for the optimal penalty factor parameters and the optimal radial basis kernel parameters, and the trained SVM model is retested according to the optimal penalty factor parameters and the optimal radial basis kernel parameters until the test accuracy is Greater than or equal to the test accuracy threshold ;

[0029] S241, setting the search range of the penalty factor parameter and the radial basis kernel parameter, and constructing a two-dimensional search grid;

[0030] S242, performing a rough search with a large step size on the two-dimensional search grid, selecting a set of penalty factor parameters and radial basis kernel parameters with the highest test accuracy, and obtaining a two-dimensional optimal parameter search grid;

[0031] S243. Calculate the test accuracy of each set of penalty factor parameters and radial basis kernel parameters in the two-dimensional optimal parameter search grid according to the cross-validation method, and select a pair of penalty factor parameters and radial basis kernel parameters with the highest test accuracy as the optimal penalty factor parameters and optimal radial basis kernel parameters.

[0032] The initial SVM model is trained and tested by collecting a large amount of data to obtain the final SVM model. At the same time, the grid search algorithm is used to search for the optimal penalty factor parameters and the optimal radial basis kernel parameters, which ensures the accuracy of the final SVM model prediction results and provides a prediction tool for subsequent operations.

[0033] Preferably, the specific steps of setting a wireless communication network user service level set, dividing each wireless communication network user in the wireless communication network user set according to the corresponding wireless communication network user service level, and obtaining the wireless communication network user classification matrix are as follows:

[0034] S31, arranging the service levels of wireless communication network users from high to low, and constructing a wireless communication network user service level set ,in, Indicates The service level of each wireless communication network user, Indicates the total number of service levels of wireless communication network users;

[0035] S32, collecting the service levels of all wireless communication network users, and classifying the service levels of each wireless communication network user according to the wireless communication network user service level, to obtain a wireless communication network user classification matrix as follows:

[0036]

[0037] in, Indicates The first level of service for users of a wireless communication network Wireless communication network users, Indicates The total number of wireless communication network users in the wireless communication network user service level.

[0038] Wireless communication network users are divided according to corresponding wireless communication network user service levels, the order of bandwidth allocation is determined, and a data basis is provided for the subsequent reasonable allocation of bandwidth.

[0039] Preferably, the specific steps of setting the optimal threshold bandwidth for each wireless communication network user service level in the wireless communication network user service level set by using an intelligent optimization algorithm to obtain the optimal threshold bandwidth set are as follows:

[0040] S41. Build an artificial bee population and set the maximum number of nectar source searches , maximum number of iterations , current iteration number And the population ;

[0041] S42, randomly generate in the search space of the artificial bee population nectar sources are used as the initial solution; the formula for randomly generating nectar sources is as follows:

[0042]

[0043] in, Indicates The honey source is The position of the dimension, express A random number between and Respectively represent The upper and lower bounds of the honey source search space;

[0044] S43, calculating the fitness function values ​​of all randomly generated nectar sources; the fitness function formula is as follows:

[0045]

[0046] in, represents the fitness function value, Indicates the bandwidth ratio that meets the basic network usage of wireless communication network users. Indicates the correction value;

[0047] S44, start iteration, select from the artificial bee population There are hired bees, and one hired bee is assigned to each nectar source. Each hired bee searches for a new nectar source near the current nectar source. After finding the new nectar source, the fitness function value of the new nectar source is calculated. If the fitness function value of the new nectar source is higher than the fitness function value of the original nectar source, the new nectar source is used to replace the original nectar source and the nectar source position is updated; otherwise, the original nectar source is retained, and the current nectar source increases the number of nectar source searches by one; the formula for hired bees to search for nectar sources is as follows:

[0048]

[0049] in, Indicates the updated honey source location, express A random number between express Not equal to A random integer, Indicates The honey source is The location of the dimension;

[0050] S45. The remaining artificial bees in the artificial bee population are used as follower bees. The follower bees select the nectar source location of the employed bees through the roulette strategy. When the follower bees arrive at the selected nectar source location, they search for a new nectar source near the current nectar source according to the formula for searching nectar sources by employed bees. After finding the new nectar source, the fitness function value of the new nectar source is calculated. If the fitness function value of the new nectar source is higher than the fitness function value of the original nectar source, the new nectar source is used to replace the original nectar source and the nectar source location is updated; otherwise, the original nectar source is retained, and the current nectar source increases the number of nectar source searches by one; the probability calculation formula for employed bees being selected is as follows:

[0051]

[0052] in, Indicates The fitness function value of each nectar source;

[0053] S46, record the number of searches for each nectar source, and when the number of searches for a nectar source reaches the maximum number of searches for a nectar source , the nectar source is abandoned, and the employed bees at the nectar source position randomly generate a new nectar source in the search space according to the steps in S44;

[0054] S47, determine the current number of iterations Whether the maximum number of iterations has been reached If it is not reached, return to S44; if it is reached, calculate the fitness function values ​​of all current honey sources, select the honey source with the highest fitness function value as the current global optimal solution, output the optimal threshold bandwidth of each wireless communication network user service level, and obtain the optimal threshold bandwidth set.

[0055] The artificial bee colony algorithm is used to set the optimal threshold bandwidth for each user service level of the wireless communication network. The artificial bee colony algorithm simulates the foraging behavior of bees in nature and establishes a mathematical model to ensure the speed and accuracy of the optimization process while improving the efficiency of the convergence process.

[0056] Preferably, the transmission data volume of each wireless communication network user in the wireless communication network user classification matrix is ​​collected in real time to obtain a real-time transmission data volume matrix, a basic energy consumption transmission data volume threshold is set to correct the real-time transmission data volume matrix to obtain a real-time transmission data volume correction matrix, and the specific steps of predicting the bandwidth required by the wireless communication network user in real time through the final SVM model are as follows:

[0057] S51. Real-time collection of wireless communication network user classification matrix The transmission data volume of each wireless communication network user in the real-time transmission data volume matrix is ​​obtained as follows:

[0058]

[0059] in, Indicates The first in the service level of wireless communication network users The real-time transmission data volume of each wireless communication network user;

[0060] S52. Setting basic energy consumption data transmission volume threshold , the real-time transmission data volume in the real-time transmission data volume matrix is ​​corrected one by one through the basic energy consumption transmission data volume threshold, and the real-time transmission data volume correction matrix is ​​obtained , the correction process is as follows:

[0061] S521, if , then The first in the service level of wireless communication network users The real-time transmission data volume of wireless communication network users remain unchanged;

[0062] like , then the The first in the service level of wireless communication network users The real-time transmission data volume of wireless communication network users Corrected to 0;

[0063] S53, correcting the real-time transmission data volume matrix The corrected real-time transmission data volume in the final SVM model is used to predict the real-time bandwidth required by each wireless communication network user, and the user real-time required bandwidth matrix is ​​obtained. as follows:

[0064]

[0065] in, Indicates The first level of service for users of a wireless communication network The bandwidth required in real time by users of a wireless communication network.

[0066] By setting the basic energy consumption transmission data volume threshold to correct the real-time collected wireless communication network user transmission data volume, and then using the final SVM model to predict the real-time required bandwidth of the wireless communication network users, the user's network needs are guaranteed while avoiding the extra consumption of bandwidth resources.

[0067] Preferably, the specific steps of allocating bandwidth to each wireless communication user in descending order according to the service level of the wireless communication network user are as follows:

[0068] S61, allocating the optimal threshold bandwidth corresponding to the optimal threshold bandwidth set to each wireless communication network user whose corrected real-time transmission data volume is not 0;

[0069] S62, allocating the total remaining bandwidth to all wireless communication network users in order of the service levels of the wireless communication network users, from high to low, to form a matrix of bandwidth required by the users in real time. The real-time bandwidth required by the corresponding user;

[0070] S621, if the remaining total bandwidth can allocate the user's real-time required bandwidth matrix to all wireless communication network users The bandwidth allocation operation is terminated after allocating the real-time bandwidth required by all wireless communication network users in the lowest wireless communication network user service level.

[0071] S622: If the remaining bandwidth is allocated to When there is a wireless communication network user service level, it is impossible to allocate the user real-time required bandwidth matrix to all wireless communication network users in the wireless communication network user service level. The corresponding user in real time requires bandwidth, then all the remaining bandwidth is evenly distributed to the first After all wireless communication network users in the wireless communication network user service level are reached, the bandwidth allocation operation is ended.

[0072] According to the predicted first user's real-time required bandwidth matrix, bandwidth is allocated to all wireless communication network users according to the wireless communication network user service level, thereby achieving effective allocation of wireless communication network bandwidth resources.

[0073] The present invention also discloses a bandwidth allocation system in a wireless communication network, including a data acquisition module, a final SVM model construction module, a wireless communication network user service level classification module, an optimal threshold bandwidth setting module, a wireless communication network user real-time required bandwidth prediction module and a wireless communication user bandwidth allocation module;

[0074] The data collection module collects the transmission data volume of each wireless communication network user by setting a transmission data volume collection period to obtain a transmission data volume matrix;

[0075] The final SVM model construction module trains and tests the constructed initial SVM model by using the transmission data volume training matrix and the transmission data volume test matrix obtained from the transmission data volume matrix to obtain the final SVM model;

[0076] The wireless communication network user service level classification module arranges the service levels of the wireless communication network users from high to low, constructs a wireless communication network user service level set, and classifies the wireless communication network users according to the corresponding wireless communication network user service levels to obtain a wireless communication network user classification matrix;

[0077] The optimal threshold bandwidth setting module sets the optimal threshold bandwidth of each wireless communication network user service level through an artificial bee colony algorithm to obtain an optimal threshold bandwidth set;

[0078] The wireless communication network user real-time required bandwidth prediction module corrects the real-time collected wireless communication network user transmission data volume by setting the basic energy consumption transmission data volume threshold to obtain a real-time transmission data volume correction matrix, and predicts the real-time required bandwidth of the wireless communication network user through the final SVM model to obtain the user's real-time required bandwidth matrix;

[0079] The wireless communication user bandwidth allocation module first allocates the optimal threshold bandwidth corresponding to the wireless communication network user service level to each wireless communication network user whose corrected real-time transmission data volume is not 0, and then allocates the real-time required bandwidth to all wireless communication network users in order from high to low according to the order of the wireless communication network user service levels.

[0080] (III) Beneficial effects

[0081] The present invention provides a method and system for allocating bandwidth in a wireless communication network, which has the following beneficial effects:

[0082] 1. The present invention is provided with a data acquisition module, a final SVM model construction module, a wireless communication network user service level classification module, an optimal threshold bandwidth setting module, a wireless communication network user real-time required bandwidth prediction module and a wireless communication user bandwidth allocation module; a large amount of data is collected to train and test the initial SVM model to obtain a final SVM model, thereby ensuring the accuracy of the final SVM model and providing a prediction tool for subsequent operations; wireless communication network users are classified according to the corresponding wireless communication network user service levels, and the optimal threshold bandwidth is allocated to each service level to meet the basic bandwidth requirements and determine the order of bandwidth allocation; bandwidth is allocated to all wireless communication network users according to the predicted user real-time required bandwidth matrix according to the wireless communication network user service level, thereby realizing the effective allocation of wireless communication network bandwidth resources;

[0083] Second, by setting the basic energy consumption transmission data volume threshold to correct the real-time collected wireless communication network user transmission data volume, and then using the final SVM model to predict the real-time bandwidth required by the wireless communication network users, the user's network needs are guaranteed while avoiding the extra consumption of bandwidth resources;

[0084] 3. Use the artificial bee colony algorithm to set the optimal threshold bandwidth for each wireless communication network user service level. The artificial bee colony algorithm simulates the foraging behavior of bees in nature and establishes a mathematical model to ensure the speed and accuracy of the optimization process while improving the efficiency of the convergence process. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, they can also obtain drawings based on these drawings without paying creative work.

[0086] Figure 1 A flow chart of a bandwidth allocation method in a wireless communication network provided by the present invention;

[0087] Figure 2 A module schematic diagram of a bandwidth allocation system in a wireless communication network provided by the present invention. DETAILED DESCRIPTION

[0088] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0089] In the description of the present invention, it is necessary to understand that the terms "opening", "upper", "lower", "top", "middle", "inside" and the like indicating orientation or positional relationship are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the invention.

[0090] Embodiment 1 of the bandwidth allocation method and system in a wireless communication network is as follows:

[0091] See also Figure 1 , a bandwidth allocation method in a wireless communication network, the method comprising the following steps:

[0092] S1. Setting a wireless communication network user set and a transmission data volume collection period set, collecting the transmission data volume of each wireless communication network user in the wireless communication network user set in each transmission data volume collection period in the transmission data volume collection period set, and obtaining a transmission data volume matrix;

[0093] S11. Setting the wireless communication network user set ,in, Indicates Wireless communication network users, Indicates the total number of wireless communication network users;

[0094] S12, set the transmission data volume collection cycle set ,in, Indicates The transmission data volume collection cycle, Indicates the total number of transmission data volume collection cycles;

[0095] S13. Collecting wireless communication network user data Each wireless communication network user in the transmission data volume collection cycle set The transmission data volume of each transmission data volume collection cycle in the transmission data volume matrix is ​​obtained as follows:

[0096]

[0097] in, Indicates Wireless communication network users in The amount of data transmitted during a transmission data collection cycle.

[0098] S2, constructing an initial SVM model, a transmission data volume training matrix, and a transmission data volume test matrix, training and testing the initial SVM model through the transmission data volume training matrix and the transmission data volume test matrix to obtain a final SVM model;

[0099] S21. Build the initial SVM model and set the penalty factor parameter to , the radial basis kernel parameters are And the kernel function of the initial parameter space during the operation of the initial SVM model is a Gaussian radial basis kernel function; the formula of the Gaussian radial basis kernel function is as follows:

[0100]

[0101] in, represents the Gaussian radial basis kernel function, represents the support vector corresponding to two training samples, express and The Euclidean distance between

[0102] S22. Set the training data ratio And test data ratio , dividing the transmission data volume matrix by the training data ratio and the test data ratio, and obtaining a transmission data volume training matrix and a transmission data volume test matrix respectively;

[0103] S23, set the training error threshold to , input the transmission data volume training matrix into the initial SVM model to train the initial SVM model until the training error of the initial SVM model is less than or equal to When , stop training and get the trained SVM model;

[0104] S24. Setting the test accuracy threshold , input the transmission data volume test matrix into the trained SVM model for testing, obtain the real-time bandwidth required by the wireless communication network user, and calculate the accuracy of the test result , when testing accuracy Greater than or equal to the test accuracy threshold If the training SVM model is , the trained SVM model is used as the final SVM model; otherwise, the grid search algorithm is used to search for the optimal penalty factor parameters and the optimal radial basis kernel parameters, and the trained SVM model is retested according to the optimal penalty factor parameters and the optimal radial basis kernel parameters until the test accuracy is Greater than or equal to the test accuracy threshold ;

[0105] S241, setting the search range of the penalty factor parameter and the radial basis kernel parameter, and constructing a two-dimensional search grid;

[0106] S242, performing a rough search with a large step size on the two-dimensional search grid, selecting a set of penalty factor parameters and radial basis kernel parameters with the highest test accuracy, and obtaining a two-dimensional optimal parameter search grid;

[0107] S243. Calculate the test accuracy of each set of penalty factor parameters and radial basis kernel parameters in the two-dimensional optimal parameter search grid according to the cross-validation method, and select the set of penalty factor parameters and radial basis kernel parameters with the highest test accuracy as the optimal penalty factor parameters and optimal radial basis kernel parameters.

[0108] S3, setting a wireless communication network user service level set, dividing each wireless communication network user in the wireless communication network user set according to the corresponding wireless communication network user service level, and obtaining a wireless communication network user classification matrix;

[0109] S31, arranging the service levels of wireless communication network users from high to low, and constructing a wireless communication network user service level set ,in, Indicates The service level of each wireless communication network user, Indicates the total number of service levels of wireless communication network users;

[0110] S32, collecting the service levels of all wireless communication network users, and classifying the service levels of each wireless communication network user according to the wireless communication network user service level, to obtain a wireless communication network user classification matrix as follows:

[0111]

[0112] in, Indicates The first level of service for users of a wireless communication network Wireless communication network users, Indicates The total number of wireless communication network users in the wireless communication network user service level.

[0113] S4. Setting an optimal threshold bandwidth for each wireless communication network user service level in the wireless communication network user service level set by using an intelligent optimization algorithm to obtain an optimal threshold bandwidth set;

[0114] S41. Build an artificial bee population and set the maximum number of nectar source searches , maximum number of iterations , current iteration number And the population ;

[0115] S42, randomly generate in the search space of the artificial bee population nectar sources are used as the initial solution; the formula for randomly generating nectar sources is as follows:

[0116]

[0117] in, Indicates The honey source is The position of the dimension, express A random number between and Respectively represent The upper and lower bounds of the honey source search space;

[0118] S43, calculating the fitness function values ​​of all randomly generated nectar sources; the fitness function formula is as follows:

[0119]

[0120] in, represents the fitness function value, Indicates the bandwidth ratio that meets the basic network usage of wireless communication network users. Indicates the correction value;

[0121] S44, start iteration, select from the artificial bee population There are hired bees, and one hired bee is assigned to each nectar source. Each hired bee searches for a new nectar source near the current nectar source. After finding the new nectar source, the fitness function value of the new nectar source is calculated. If the fitness function value of the new nectar source is higher than the fitness function value of the original nectar source, the new nectar source is used to replace the original nectar source and the nectar source position is updated; otherwise, the original nectar source is retained, and the current nectar source increases the number of nectar source searches by one; the formula for hired bees to search for nectar sources is as follows:

[0122]

[0123] in, Indicates the updated honey source location, express A random number between express Not equal to A random integer, Indicates The honey source is The location of the dimension;

[0124] S45. The remaining artificial bees in the artificial bee population are used as follower bees. The follower bees select the nectar source location of the employed bees through the roulette strategy. When the follower bees arrive at the selected nectar source location, they search for a new nectar source near the current nectar source according to the formula for searching nectar sources by employed bees. After finding the new nectar source, the fitness function value of the new nectar source is calculated. If the fitness function value of the new nectar source is higher than the fitness function value of the original nectar source, the new nectar source is used to replace the original nectar source and the nectar source location is updated; otherwise, the original nectar source is retained, and the current nectar source increases the number of nectar source searches by one; the probability calculation formula for employed bees being selected is as follows:

[0125]

[0126] in, Indicates The fitness function value of each nectar source;

[0127] S46, record the number of searches for each nectar source, and when the number of searches for a nectar source reaches the maximum number of searches for a nectar source , the nectar source is abandoned, and the employed bees at the nectar source position randomly generate a new nectar source in the search space according to the steps in S44;

[0128] S47, determine the current number of iterations Whether the maximum number of iterations has been reached If it is not reached, return to S44; if it is reached, calculate the fitness function values ​​of all current honey sources, select the honey source with the highest fitness function value as the current global optimal solution, output the optimal threshold bandwidth of each wireless communication network user service level, and obtain the optimal threshold bandwidth set.

[0129] S5, collecting the transmission data volume of each wireless communication network user in the wireless communication network user classification matrix in real time, obtaining a real-time transmission data volume matrix, setting a basic energy consumption transmission data volume threshold to correct the real-time transmission data volume matrix, obtaining a real-time transmission data volume correction matrix, and predicting the real-time required bandwidth of the wireless communication network user through the final SVM model;

[0130] S51. Real-time collection of wireless communication network user classification matrix The real-time transmission data volume of each wireless communication network user in the real-time transmission data volume matrix is ​​obtained as follows:

[0131]

[0132] in, Indicates The first in the service level of wireless communication network users The real-time transmission data volume of each wireless communication network user;

[0133] S52. Setting basic energy consumption data transmission volume threshold , the real-time transmission data volume in the real-time transmission data volume matrix is ​​corrected one by one through the basic energy consumption transmission data volume threshold, and the real-time transmission data volume correction matrix is ​​obtained , the correction process is as follows:

[0134] S521, if , then The first in the service level of wireless communication network users The real-time transmission data volume of wireless communication network users remain unchanged;

[0135] like , then the The first in the service level of wireless communication network users The real-time transmission data volume of wireless communication network users Corrected to 0;

[0136] S53, correcting the real-time transmission data volume matrix The corrected real-time transmission data volume in the final SVM model is used to predict the real-time bandwidth required by each wireless communication network user, and the user real-time required bandwidth matrix is ​​obtained. as follows:

[0137]

[0138] in, Indicates The first level of service for users of a wireless communication network The bandwidth required in real time by users of a wireless communication network.

[0139] S6. Allocate bandwidth to each wireless communication user in descending order according to the service level of the wireless communication network user;

[0140] S61, allocating the optimal threshold bandwidth corresponding to the optimal threshold bandwidth set to each wireless communication network user whose corrected real-time transmission data volume is not 0;

[0141] S62, allocating the total remaining bandwidth to all wireless communication network users in order of the service levels of the wireless communication network users, from high to low, to form a matrix of bandwidth required by the users in real time. The real-time bandwidth required by the corresponding user;

[0142] S621, if the remaining total bandwidth can allocate the user's real-time required bandwidth matrix to all wireless communication network users The bandwidth allocation operation is terminated after allocating the real-time bandwidth required by all wireless communication network users in the lowest wireless communication network user service level.

[0143] S622: If the remaining bandwidth is allocated to When there is a wireless communication network user service level, it is impossible to allocate the user real-time required bandwidth matrix to all wireless communication network users in the wireless communication network user service level. The corresponding user in real time requires bandwidth, then all the remaining bandwidth is evenly distributed to the first After all wireless communication network users in the wireless communication network user service level are reached, the bandwidth allocation operation is ended.

[0144] Embodiment 2 of the method and system for allocating bandwidth in a wireless communication network is as follows:

[0145] See also Figure 2 , a bandwidth allocation system in a wireless communication network includes a data acquisition module, a final SVM model building module, a wireless communication network user service level classification module, an optimal threshold bandwidth setting module, a wireless communication network user real-time required bandwidth prediction module and a wireless communication user bandwidth allocation module;

[0146] The data collection module collects the transmission data volume of each wireless communication network user by setting a transmission data volume collection period to obtain a transmission data volume matrix;

[0147] The final SVM model construction module trains and tests the constructed initial SVM model by using the transmission data volume training matrix and the transmission data volume test matrix obtained from the transmission data volume matrix to obtain the final SVM model;

[0148] The wireless communication network user service level classification module arranges the service levels of the wireless communication network users from high to low, constructs a wireless communication network user service level set, and classifies the wireless communication network users according to the corresponding wireless communication network user service levels to obtain a wireless communication network user classification matrix;

[0149] The optimal threshold bandwidth setting module sets the optimal threshold bandwidth of each wireless communication network user service level through an artificial bee colony algorithm to obtain an optimal threshold bandwidth set;

[0150] The wireless communication network user real-time required bandwidth prediction module corrects the real-time collected wireless communication network user transmission data volume by setting the basic energy consumption transmission data volume threshold to obtain a real-time transmission data volume correction matrix, and predicts the real-time required bandwidth of the wireless communication network user through the final SVM model to obtain the user's real-time required bandwidth matrix;

[0151] The wireless communication user bandwidth allocation module first allocates the optimal threshold bandwidth corresponding to the wireless communication network user service level to each wireless communication network user whose corrected real-time transmission data volume is not 0, and then allocates the real-time required bandwidth to all wireless communication network users in order from high to low according to the order of the wireless communication network user service levels.

[0152] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0153] The preferred embodiments of the invention disclosed above are only used to help explain the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the invention, so that those skilled in the art can understand and use the invention well.

Claims

1. A method for allocating bandwidth in a wireless communication network, characterized in that: The steps include: S1. Setting a wireless communication network user set and a transmission data volume collection period set, collecting the transmission data volume of each wireless communication network user in the wireless communication network user set in each transmission data volume collection period in the transmission data volume collection period set, and obtaining a transmission data volume matrix; S2, constructing an initial SVM model, a transmission data volume training matrix, and a transmission data volume test matrix, training and testing the initial SVM model through the transmission data volume training matrix and the transmission data volume test matrix to obtain a final SVM model; S3, setting a wireless communication network user service level set, dividing each wireless communication network user in the wireless communication network user set according to the corresponding wireless communication network user service level, and obtaining a wireless communication network user classification matrix; S4. Setting an optimal threshold bandwidth for each wireless communication network user service level in the wireless communication network user service level set by using an intelligent optimization algorithm to obtain an optimal threshold bandwidth set; S5, collecting the transmission data volume of each wireless communication network user in the wireless communication network user classification matrix in real time, obtaining a real-time transmission data volume matrix, setting a basic energy consumption transmission data volume threshold to correct the real-time transmission data volume matrix, obtaining a real-time transmission data volume correction matrix, and predicting the real-time required bandwidth of the wireless communication network user through the final SVM model; S6. Allocate bandwidth to each wireless communication user in descending order according to the service level of the wireless communication network user; The S1 comprises the following steps: S11, set the wireless communication network user set A = {a1, a2, ..., a i ,…,a k }, a i represents the i-th wireless communication network user, k represents the total number of wireless communication network users; S12, set the transmission data volume collection cycle set B = {b1, b2, ..., b i ,…,b l }, b i represents the i-th transmission data volume collection cycle, l represents the total number of transmission data volume collection cycles; S13, collect wireless communication network user set A = {a1, a2, ..., a i ,…,a k Each wireless communication network user in the transmission data volume collection period set B = {b1, b2, ..., b i ,…,b l }, and obtain the transmission data volume matrix C; The S2 comprises the following steps: S21, constructing an initial SVM model, setting the penalty factor parameter to V, the radial basis kernel parameter to γ, and the kernel function of the initial parameter space during the operation of the initial SVM model to a Gaussian radial basis kernel function; S22, setting a training data ratio α and a test data ratio 1-α, dividing the transmission data volume matrix by the training data ratio and the test data ratio, and obtaining a transmission data volume training matrix and a transmission data volume test matrix respectively; S23, setting the training error threshold to η1, inputting the transmission data volume training matrix into the initial SVM model to train the initial SVM model until the training error of the initial SVM model is less than or equal to η1, stopping the training, and obtaining a trained SVM model; S24, set the test accuracy threshold η2, input the transmission data volume test matrix into the trained SVM model for testing, obtain the real-time bandwidth required by the wireless communication network user, calculate the accuracy m of the test result, and when m≥η2, use the trained SVM model as the final SVM model; otherwise, use the grid search algorithm to search for the optimal penalty factor parameters and the optimal radial basis kernel parameters, and retest the trained SVM model according to the optimal penalty factor parameters and the optimal radial basis kernel parameters until m≥η2; The grid search for the optimal penalty factor parameters and the optimal radial basis kernel parameters in S24 includes the following steps: S241, setting the search range of the penalty factor parameter and the radial basis kernel parameter, and constructing a two-dimensional search grid; S242, performing a rough search with a large step size on the two-dimensional search grid, selecting a set of penalty factor parameters and radial basis kernel parameters with the highest test accuracy, and obtaining a two-dimensional optimal parameter search grid; S243, calculating the test accuracy of each set of penalty factor parameters and radial basis kernel parameters in the two-dimensional optimal parameter search grid according to the cross-validation method, and selecting a set of penalty factor parameters and radial basis kernel parameters with the highest test accuracy as the optimal penalty factor parameters and optimal radial basis kernel parameters; The S3 comprises the following steps: S31, arranging the service levels of wireless communication network users from high to low, and constructing a wireless communication network user service level set A'={a'1, a'2, ..., a' i ,…,a' p }, a' i represents the service level of the i-th wireless communication network user, and p represents the total number of service levels of the wireless communication network users; S32, collecting the service levels of all wireless communication network users, and classifying the service levels of each wireless communication network user according to the wireless communication network user service level set, to obtain a wireless communication network user classification matrix D; The S4 comprises the following steps: S41. Build an artificial bee population and set the maximum number of nectar source searches, limit, and the maximum number of iterations, t max , the current number of iterations t and the population size N; S42, randomly generating N nectar sources in the search space of the artificial bee population as initial solutions; S43, calculating the fitness of all randomly generated nectar sources; the fitness function formula is as follows: Wherein, fit(x) represents the fitness function value, x represents the bandwidth ratio that satisfies the basic network usage of the wireless communication network users, and λ represents the correction value; S44, start iteration, select N employed bees from the artificial bee population, assign one employed bee to each nectar source, each employed bee searches for a new nectar source near the current nectar source, calculates the fitness of the new nectar source after finding the new nectar source, and if the fitness of the new nectar source is higher than that of the original nectar source, replaces the original nectar source with the new nectar source; otherwise, the original nectar source increases the number of nectar source searches by one; S45, the remaining artificial bees in the artificial bee population are used as follower bees, and the follower bees select the nectar source location of the employed bees through a roulette strategy. When the follower bees arrive at the selected nectar source location, they search for a new nectar source near the current nectar source according to the formula for searching nectar sources by employed bees, and calculate the fitness of the new nectar source after finding it. If the fitness of the new nectar source is higher than that of the original nectar source, the new nectar source replaces the original nectar source; otherwise, the number of nectar source searches for the original nectar source increases once; S46, record the number of searches for each nectar source, and when the number of searches for a nectar source reaches the maximum nectar source search number limit, the nectar source is abandoned, and the employed bees at the nectar source position randomly generate a new nectar source in the search space according to the steps in S44; S47, determine whether the current number of iterations t reaches the maximum number of iterations t max If not, return to S44; if reached, calculate the fitness of all current honey sources, select the honey source with the highest fitness as the current global optimal solution, output the optimal threshold bandwidth of each wireless communication network user service level, and obtain the optimal threshold bandwidth set; The S5 comprises the following steps: S51, real-time collection of the real-time transmission data volume of each wireless communication network user in the wireless communication network user classification matrix D, and obtaining the real-time transmission data volume matrix D' as follows: Among them, d' ij represents the real-time transmission data volume of the jth wireless communication network user in the i-th wireless communication network user service level, d i represents the total number of wireless communication network users in the i-th wireless communication network user service level; S52, correcting each real-time transmission data amount in the real-time transmission data amount matrix one by one to obtain a real-time transmission data amount correction matrix D"; S53, input each corrected real-time transmission data volume in the real-time transmission data volume correction matrix D" into the final SVM model to predict the real-time bandwidth required by each wireless communication network user, and obtain the real-time required bandwidth matrix of the user 2. A method for allocating bandwidth in a wireless communication network according to claim 1, characterized in that: The step S52 of correcting each real-time transmission data amount in the real-time transmission data amount matrix one by one comprises the following steps: A basic energy consumption data transmission volume threshold n is set, and the real-time transmission data volumes in the real-time transmission data volume matrix are corrected one by one by the basic energy consumption data transmission volume threshold; If d' ij >n, then the real-time transmission data volume d' of the jth wireless communication network user in the i-th wireless communication network user service level ij remain unchanged; If d' ij ≤n, then the real-time transmission data volume d' of the jth wireless communication network user in the i-th wireless communication network user service level ij Corrected to 0.

3. A method for allocating bandwidth in a wireless communication network according to claim 2, characterized in that: The S6 comprises the following steps: S61, allocating the optimal threshold bandwidth corresponding to the optimal threshold bandwidth set to each wireless communication network user whose corrected real-time transmission data volume is not 0; S62, allocating the total remaining bandwidth to all wireless communication network users in order of the service levels of the wireless communication network users, from high to low, to form a matrix of bandwidth required by the users in real time. The real-time bandwidth required by the corresponding user; S621, if the remaining total bandwidth can allocate the user's real-time required bandwidth matrix to all wireless communication network users The bandwidth allocation operation is terminated after allocating the real-time bandwidth required by all wireless communication network users in the lowest wireless communication network user service level. S622: If the remaining total bandwidth cannot be allocated to all wireless communication network users in the wireless communication network user service level when it is allocated to the i-th wireless communication network user service level, the user real-time required bandwidth matrix The bandwidth required by the corresponding user in real time is calculated, and then all the remaining bandwidth is evenly distributed to all wireless communication network users in the i-th wireless communication network user service level, and the bandwidth allocation operation is ended.

4. A system for implementing the bandwidth allocation method in a wireless communication network as described in any one of claims 1 to 3, comprising a data acquisition module, a final SVM model construction module, a wireless communication network user service level classification module, an optimal threshold bandwidth setting module, a wireless communication network user real-time required bandwidth prediction module and a wireless communication user bandwidth allocation module.