Method for realizing scene wireless network quality evaluation based on analytic hierarchy process

Abstract

The invention particularly relates to a method for realizing scenarized wireless network quality evaluation based on an analytic hierarchy process. According to the method for realizing the scene wireless network quality evaluation based on the analytic hierarchy process, a scene health degree evaluation index system is constructed, and a judgment matrix is constructed for each layer of indexes by performing assignment on the relative importance of each layer of indexes; calculating a maximum characteristic value lambda max of the judgment matrix and a corresponding characteristic vector thereof to obtain a weight vector W; after the judgment matrix passes the consistency verification, calculating the comprehensive weight of the final index; scoring each index according to the formulated rule, obtaining the final score of each index in combination with the index weight, adding to obtain the health degree coefficient of the whole scene, and further obtaining the health degree level of the whole scene. According to the method for realizing the scenarized wireless network quality evaluation based on the analytic hierarchy process, comprehensive monitoring on the wireless network and comprehensive evaluation on the scenarized wireless network are realized.

Description

Technical Field

[0001] The present invention relates to the field of communication information technology, and in particular to a method for implementing scenario-based wireless network quality assessment based on hierarchical analysis method. Background Art

[0002] As human society's demand for information acquisition and transmission continues to increase, the scale of wireless networks has shown explosive growth, and the quality monitoring and evaluation of wireless networks has become particularly important.

[0003] In order to further strengthen the perception-oriented analysis capability, ensure the rapid and fundamental solution of perception problems, implement the customer-centricity of the network, build a perception-based wireless network quality optimization system, realize the production traction transformation from "performance-centric" to "perception-centric", the analysis object transformation from "network" to "user and business", and the solution granularity transformation from "single problem cell, single complaint" to key scenarios, and strengthen the effective transmission of real customer experience to network operation, the present invention proposes a method for scenario-based wireless network quality assessment based on hierarchical analysis method. Summary of the invention

[0004] In order to overcome the defects of the prior art, the present invention provides a simple and efficient method for implementing scenario-based wireless network quality assessment based on hierarchical analysis method.

[0005] The present invention is achieved through the following technical solutions:

[0006] A method for implementing scenario-based wireless network quality assessment based on hierarchical analysis method, characterized in that it includes the following steps:

[0007] Step S1, constructing a scenario health evaluation indicator system, where the first-level indicators include 4G network B1 and 5G network B2;

[0008] The second-layer indicators of 4G network B1 include quality analysis C11, complaint molecules C12, perception analysis C13 and collaborative analysis C14;

[0009] The third-layer indicators of quality analysis C11 include no direct coverage D111 and poor network quality D112. The third-layer indicators of complaint analysis C12 include complaint hotspots D121 and repeated complaints D122. The third-layer indicators of perception analysis C13 include users with poor perceived quality D131, poor data service perception D132, and poor voice service perception D133. The third-layer indicators of collaborative analysis C14 include backflow D141, diversion D142, terminal development D143, and residence D144.

[0010] The second-layer indicators of 5G network B2 all include quality analysis C21, complaint element C22, perception analysis C23 and collaborative analysis C24;

[0011] The third-layer indicators of quality analysis C21 include no direct coverage D211 and poor network quality D212. The third-layer indicators of complaint analysis C22 include complaint hotspots D221 and repeated complaints D222. The third-layer indicators of perception analysis C23 include users with poor perceived quality D231, poor data service perception D232, and poor voice service perception D233. The third-layer indicators of collaborative analysis C24 include backflow D241, diversion D242, terminal development D243, and residence D244D.

[0012] By assigning the relative importance of each of the above indicators, a judgment matrix is ​​constructed for each indicator layer;

[0013] Step S2, calculate the eigenvalue λ of the judgment matrix and the eigenvector corresponding to the eigenvalue, and take the maximum eigenvalue λ max and its corresponding eigenvector, and normalize the eigenvector to obtain the weight vector W;

[0014] Step S3, performing consistency check on the judgment matrix. If the consistency check passes, it is considered that the ranking of the indicator weights represented by the weight vector W is reasonable and effective;

[0015] If the consistency check fails, the judgment matrix is ​​adjusted until it passes the consistency check;

[0016] Step S4, calculating the comprehensive weight of the final indicator;

[0017] The comprehensive weight of the final indicator is the ratio of the weight of the corresponding third-layer indicator to the sum of the weights of the third-layer indicators in the 4G network and 5G network scenarios;

[0018] Step S5, formulate the measurement rules of the indicator health, divide the health into 5 levels, and the corresponding numerical values ​​are 1 point for low health, 2 points for lower health, 3 points for medium health, 4 points for higher health, and 5 points for high health;

[0019] According to the measurement rules of indicator health, each indicator except the no direct coverage indicator is scored. When scoring the no direct coverage indicator, it is determined whether the scene has cell coverage. If there is cell coverage, it is 1, otherwise it is 0;

[0020] Multiply the score of each indicator by the corresponding weight to get the final score of each indicator; add up the final scores of each indicator to get the health coefficient of the entire scene;

[0021] According to the measurement rules of indicator health, the health level of the entire scene is obtained; the measurement rules of indicator health are as follows:

[0022] The health coefficient is not less than 0.5 and less than 1.5, which is the first level of health. The health coefficient is not less than 1.5 and less than 2.5, which is the second level of health. The health coefficient is not less than 2.5 and less than 3.5, which is the third level of health. The health coefficient is not less than 3.5 and less than 4.5, which is the fourth level of health. The health coefficient is not less than 4.5 and less than 5, which is the fifth level of health.

[0023] In the step S1, the no direct coverage indicator is used to determine whether the scenario has cell coverage, the poor network quality indicator is used to evaluate the proportion of poor quality cells in the scenario, the complaint hotspot indicator is used to evaluate the number of hotspot complaints in the scenario, the complaint repetition indicator is used to evaluate the number of repeated complaints in the scenario, the perceived poor quality user indicator is used to evaluate the proportion of perceived poor quality users in the scenario, the data service perception poor indicator is used to evaluate the proportion of data service perception poor quality in the scenario, the voice service perception poor indicator is used to evaluate the proportion of voice service perception poor quality in the scenario, the backflow indicator is used to evaluate the proportion of high backflow cells, the diversion indicator is used to evaluate the diversion ratio in the scenario, the terminal development indicator is used to evaluate the proportion of new terminal users in the scenario, and the residence indicator is used to evaluate the residence ratio.

[0024] In step S2, for the matrix M, the vector X and the real number λ, if MX=λX, where X≠0, then λ is called the eigenvalue of the matrix M, and the vector X is called the eigenvector of the matrix M corresponding to the eigenvalue λ;

[0025] The eigenvalues ​​and eigenvectors are calculated using the R language. The syntax is eigen(M), where M is the matrix for which the eigenvalues ​​and eigenvectors need to be calculated.

[0026] In step S3, for an n-order reciprocal matrix M, where m ij >0, m ii =1, the calculation method of consistency index CI is as follows:

[0027]

[0028] Consistency test coefficient:

[0029] The consistency test coefficient CR is the ratio of the consistency index CI to the random consistency index RI, and the calculation formula is as follows:

[0030]

[0031] Among them, the value of the random consistency index RI is related to the dimension of the judgment matrix; the values ​​of the random consistency index RI corresponding to matrix dimensions 1, 2, 3 and 4 are 0, 0, 0.58 and 0.9 respectively; if the consistency test coefficient CR is less than or equal to 0.1, the judgment matrix is ​​considered to be consistent, and the order of the indicator weights represented by its weight vector is reasonable and effective, otherwise the judgment matrix is ​​considered to be inconsistent and the judgment matrix needs to be adjusted until the consistency test coefficient is less than or equal to 0.1.

[0032] In the first-layer indicators, 5G network B2 is more important than 4G network B1. Therefore, in the judgment matrix AB, the value assigned to 5G network B2 relative to 4G network B1 is 3, and the value assigned to 4G network B1 relative to 5G network B2 is 1 / 3. The judgment matrix AB and weight vector are shown in the following table:

[0033] Table 3 Judgment matrix AB and weight vector

[0034] AB B1 B2 W B1 1 1 / 3 0.25 B2 3 1 0.75

[0035] Its maximum eigenvalue = 2, and the random consistency index CI of the two-dimensional matrix is ​​zero. There is no need to calculate CR, and it can be directly judged that the matrix AB has passed the consistency check;

[0036] The second layer indicators of the 4G network B1, the constructed judgment matrix B1-C1 and the weight vector are shown in the following table:

[0037] Table 4 Judgment matrix B1-C1 and weight vector

[0038] B1-C1 C11 C12 C13 C14 W C11 1 1 / 3 1 / 3 1 / 5 0.07353 C12 3 1 1 1 / 7 0.14652 C13 3 1 1 1 / 5 0.15578 C14 5 7 5 1 0.62416

[0039] Its maximum eigenvalue λ max =4.218, CI=0.218 / 3=0.073, CR=0.083; CR<0.1, so the judgment matrix B1-C1 passed the consistency check;

[0040] The third level indicators of the quality analysis C11, the constructed judgment matrix C11-D11 and the weight vector are shown in the following table:

[0041] Table 5 Judgment matrix C11-D11 and weight vector

[0042] C11-D11 D111 D112 W D111 1 1 / 7 0.125 D112 7 1 0.875

[0043] Its maximum eigenvalue λ max =2, CI = 0, the random consistency index of the two-dimensional matrix is ​​zero, there is no need to calculate CR, and it is directly judged that the matrix AB has passed the consistency check;

[0044] The third-level indicators of the complainant C12, the constructed judgment matrix C12-D12 and the weight vector are shown in the following table:

[0045] Table 6 Judgment matrix C12-D12 and weight vector

[0046] C12-D12 D121 D122 W D121 1 1 / 3 0.25 D122 3 1 0.75

[0047] Its maximum eigenvalue λ max =2, the random consistency index CI of the two-dimensional matrix is ​​zero, there is no need to calculate CR, and it is directly judged that the matrix AB has passed the consistency check;

[0048] The third-level indicators of the perception analysis C13, the constructed judgment matrix C13-D13 and the weight vector are shown in the following table:

[0049] Table 7 Judgment matrix C13-D13 and weight vector

[0050]

[0051]

[0052] Its maximum eigenvalue λ max =3.039, CI = 0.039 / 2 = 0.019, CR = 0.037; CR < 0.1, so the judgment matrix C13-D13 passed the consistency check

[0053] The third-level indicators of the collaborative analysis C14, the constructed judgment matrix C14-D14 and the weight vector are shown in the following table:

[0054] Table 8 Judgment matrix C14-D14 and weight vector

[0055] C14-D14 D141 D142 D143 D144 W D141 1 1 / 2 1 / 5 1 / 5 0.07537 D142 2 1 1 / 3 1 / 3 0.13758 D143 5 3 1 1 0.39352 D144 5 3 1 1 0.39352

[0056] Its maximum eigenvalue λ max =4.004, CI=0.004 / 3=0.001, CR=0.002; CR<0.1, so the judgment matrix C14-D14 passed the consistency check;

[0057] The second layer indicators of the 5G network B2, the constructed judgment matrix B2-C2 and the weight vector are shown in the following table:

[0058] Table 9 Judgment matrix B2-C2 and weight vector

[0059] B2-C2 C21 C22 C23 C24 W C21 1 1 / 3 1 / 5 1 / 5 0.068 C22 3 1 1 / 3 1 / 4 0.14198 C23 5 3 1 1 0.38001 C24 5 4 1 1 0.41001

[0060] Its maximum eigenvalue λ max =4.076, CI=0.076 / 3=0.025, CR=0.029; CR<0.1, so the judgment matrix B2-C2 passed the consistency check;

[0061] The third level indicators of the quality analysis C21, the constructed judgment matrix C21-D21 and the weight vector are shown in the following table:

[0062] Table 10 Judgment matrix C21-D21 and weight vector

[0063]

[0064]

[0065] Its maximum eigenvalue λ max =2, the random consistency index CI of the two-dimensional matrix is ​​zero, there is no need to calculate CR, and it is directly judged that the matrix C21-D21 has passed the consistency check;

[0066] The third-level indicators of the complainant C22, the constructed judgment matrix C22-D22 and the weight vector are shown in the following table:

[0067] Table 11 Judgment matrix C22-D22 and weight vector

[0068] C22-D22 D221 D222 W D221 1 1 / 7 0.125 D222 7 1 0.875

[0069] Its maximum eigenvalue λ max =2, the random consistency index CI of the two-dimensional matrix is ​​zero, there is no need to calculate CR, and it is directly judged that the matrix C22-D22 has passed the consistency check;

[0070] The third-level indicators of the perception analysis C23, the constructed judgment matrix C23-D23 and the weight vector are shown in the following table:

[0071] Table 12 Judgment matrix C23-D23 and weight vector

[0072] C23-D23 D231 D232 D233 W D231 1 1 / 3 1 / 5 0.10616 D232 3 1 1 / 3 0.2605 D233 5 3 1 0.63335

[0073] Its maximum eigenvalue λ max =4.076, CI=0.076 / 3=0.025, CR=0.029; CR<0.1, so the judgment matrix C23-D23 passed the consistency check;

[0074] The third-level indicators of the collaborative analysis C24, the constructed judgment matrix C24-D24 and the weight vector are shown in the following table:

[0075] Table 13 Judgment matrix C24-D24 and weight vector

[0076] C24-D24 D241 D242 D243 D244 W D241 1 1 / 3 1 / 3 1 / 3 0.09832 D242 3 1 1 / 3 1 0.21803 D243 3 3 1 2 0.44639 D244 3 1 1 / 2 1 0.23726

[0077] Its maximum eigenvalue λ max=4.118, CI=0.118 / 3=0.039, CR=0.045; CR<0.1, so the judgment matrix C24-D24 passed the consistency check.

[0078] In step S4, the comprehensive weights of the final indicators are shown in the following table:

[0079] Table 14 Comprehensive indicators and corresponding weights

[0080]

[0081]

[0082] In step S5, the corresponding relationship between the index score and the health degree is shown in the following table:

[0083] Table 16 Correspondence between indicator scores and health level

[0084]

[0085] A system for implementing scenario-based wireless network quality assessment based on hierarchical analysis method, comprising an assessment index system construction module, a judgment matrix processing module, a comprehensive weight calculation module, a measurement rule module and a measurement module;

[0086] The evaluation index system construction module is used to construct the scenario health evaluation index system. By assigning the relative importance of each layer of indicators mentioned above, a judgment matrix is ​​constructed for each layer of indicators.

[0087] The first-tier indicators include 4G network B1 and 5G network B2;

[0088] The second-layer indicators of 4G network B1 include quality analysis C11, complaint molecules C12, perception analysis C13 and collaborative analysis C14;

[0089] The third-layer indicators of quality analysis C11 include no direct coverage D111 and poor network quality D112. The third-layer indicators of complaint analysis C12 include complaint hotspots D121 and repeated complaints D122. The third-layer indicators of perception analysis C13 include users with poor perceived quality D131, poor data service perception D132, and poor voice service perception D133. The third-layer indicators of collaborative analysis C14 include backflow D141, diversion D142, terminal development D143, and residence D144.

[0090] The second-layer indicators of 5G network B2 all include quality analysis C21, complaint element C22, perception analysis C23 and collaborative analysis C24;

[0091] The third-layer indicators of quality analysis C21 include no direct coverage D211 and poor network quality D212. The third-layer indicators of complaint analysis C22 include complaint hotspots D221 and repeated complaints D222. The third-layer indicators of perception analysis C23 include users with poor perceived quality D231, poor data service perception D232, and poor voice service perception D233. The third-layer indicators of collaborative analysis C24 include backflow D241, diversion D242, terminal development D243, and residence D244D.

[0092] The judgment matrix processing module is responsible for calculating the eigenvalue λ of the judgment matrix and the eigenvector corresponding to the eigenvalue, and taking the maximum eigenvalue λ max and its corresponding eigenvector, and normalize the eigenvector to obtain the weight vector W;

[0093] The judgment matrix is ​​checked for consistency. If it passes the consistency check, it is considered that the ranking of the indicator weights represented by the weight vector W is reasonable and effective;

[0094] If the consistency check fails, the judgment matrix is ​​adjusted until it passes the consistency check;

[0095] Comprehensive weight calculation module, used to calculate the comprehensive weight of the final indicator;

[0096] The comprehensive weight of the final indicator is the ratio of the weight of the corresponding third-layer indicator to the sum of the weights of the third-layer indicators in the 4G network and 5G network scenarios;

[0097] The measurement rule module is used to formulate the measurement rules for indicator health and scenario health;

[0098] In the measurement rules of the indicator health, the health is divided into 5 levels, and the corresponding numerical values ​​are 1 point for low health, 2 points for lower health, 3 points for medium health, 4 points for higher health, and 5 points for high health;

[0099] In the measurement rules of scenario health, the health coefficient is not less than 0.5 and less than 1.5 for the first level of health, the health coefficient is not less than 1.5 and less than 2.5 for the second level of health, the health coefficient is not less than 2.5 and less than 3.5 for the third level of health, the health coefficient is not less than 3.5 and less than 4.5 for the fourth level of health, and the health coefficient is not less than 4.5 and less than 5 for the fifth level of health;

[0100] The measurement module is used to score each indicator except the no direct coverage indicator according to the measurement rules of the indicator health. When scoring the no direct coverage indicator, it is determined whether the scene has cell coverage. If there is cell coverage, it is 1, otherwise it is 0;

[0101] Multiply the score of each indicator by the corresponding weight to obtain the final score of each indicator; add up the final scores of each indicator to obtain the health coefficient of the entire scene; according to the measurement rules of scene health, obtain the health level of the entire scene.

[0102] A device for implementing scenario-based wireless network quality assessment based on hierarchical analysis method, characterized by comprising a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the above method steps when executing the computer program.

[0103] A readable storage medium, characterized in that: a computer program is stored on the readable storage medium, and the computer program implements the above method steps when executed by a processor.

[0104] The beneficial effects of the present invention are as follows: the method for implementing scenario-based wireless network quality assessment based on the hierarchical analysis method realizes comprehensive monitoring of the wireless network by constructing a set of health assessment indicator systems for the wireless network, and realizes comprehensive assessment of the scenario-based wireless network by analyzing the correlation between various network indicators, thus solving the problem of lack of systematic assessment results in the traditional network optimization process and inability to support daily optimization work of the wireless network. BRIEF DESCRIPTION OF THE DRAWINGS

[0105] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0106] Attached Figure 1 Schematic diagram of the scenario health assessment index system of the present invention. DETAILED DESCRIPTION

[0107] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only a 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 should fall within the scope of protection of the present invention.

[0108] The analytic hierarchy process is a multi-criteria decision analysis method that assists decision makers in solving complex problems by decomposing decision-related elements into different levels, including goals, criteria, and plans, and performing qualitative and quantitative analysis. This method is suitable for target systems with hierarchical and staggered evaluation indicators, especially when the target value is difficult to describe quantitatively.

[0109] The method for implementing scenario-based wireless network quality assessment based on hierarchical analysis method includes the following steps:

[0110] Step S1, constructing a scenario health evaluation indicator system, where the first-level indicators include 4G network B1 and 5G network B2;

[0111] The second-layer indicators of 4G network B1 include quality analysis C11, complaint molecules C12, perception analysis C13 and collaborative analysis C14;

[0112] The third-layer indicators of quality analysis C11 include no direct coverage D111 and poor network quality D112. The third-layer indicators of complaint analysis C12 include complaint hotspots D121 and repeated complaints D122. The third-layer indicators of perception analysis C13 include users with poor perceived quality D131, poor data service perception D132, and poor voice service perception D133. The third-layer indicators of collaborative analysis C14 include backflow D141, diversion D142, terminal development D143, and residence D144.

[0113] The second-layer indicators of 5G network B2 all include quality analysis C21, complaint element C22, perception analysis C23 and collaborative analysis C24;

[0114] The third-layer indicators of quality analysis C21 include no direct coverage D211 and poor network quality D212. The third-layer indicators of complaint analysis C22 include complaint hotspots D221 and repeated complaints D222. The third-layer indicators of perception analysis C23 include users with poor perceived quality D231, poor data service perception D232, and poor voice service perception D233. The third-layer indicators of collaborative analysis C24 include backflow D241, diversion D242, terminal development D243, and residence D244D.

[0115] By assigning the relative importance of each of the above indicators, a judgment matrix is ​​constructed for each indicator layer;

[0116] Step S2, calculate the eigenvalue λ of the judgment matrix and the eigenvector corresponding to the eigenvalue, and take the maximum eigenvalue λ max and its corresponding eigenvector, and normalize the eigenvector to obtain the weight vector W;

[0117] Step S3, performing consistency check on the judgment matrix. If the consistency check passes, it is considered that the ranking of the indicator weights represented by the weight vector W is reasonable and effective;

[0118] If the consistency check fails, the judgment matrix is ​​adjusted until it passes the consistency check;

[0119] Step S4, calculating the comprehensive weight of the final indicator;

[0120] The comprehensive weight of the final indicator is the ratio of the weight of the corresponding third-layer indicator to the sum of the weights of the third-layer indicators in the 4G network and 5G network scenarios;

[0121] Step S5, formulate the measurement rules of the indicator health, divide the health into 5 levels, and the corresponding numerical values ​​are 1 point for low health, 2 points for lower health, 3 points for medium health, 4 points for higher health, and 5 points for high health;

[0122] According to the measurement rules of indicator health, each indicator except the no direct coverage indicator is scored. When scoring the no direct coverage indicator, it is determined whether the scene has cell coverage. If there is cell coverage, it is 1, otherwise it is 0;

[0123] Multiply the score of each indicator by the corresponding weight to get the final score of each indicator; add up the final scores of each indicator to get the health coefficient of the entire scene;

[0124] According to the measurement rules of indicator health, the health level of the entire scene is obtained; the measurement rules of indicator health are as follows:

[0125] The health coefficient is not less than 0.5 and less than 1.5, which is the first level of health. The health coefficient is not less than 1.5 and less than 2.5, which is the second level of health. The health coefficient is not less than 2.5 and less than 3.5, which is the third level of health. The health coefficient is not less than 3.5 and less than 4.5, which is the fourth level of health. The health coefficient is not less than 4.5 and less than 5, which is the fifth level of health.

[0126] In the step S1, the no direct coverage indicator is used to determine whether the scenario has cell coverage, the poor network quality indicator is used to evaluate the proportion of poor quality cells in the scenario, the complaint hotspot indicator is used to evaluate the number of hotspot complaints in the scenario, the complaint repetition indicator is used to evaluate the number of repeated complaints in the scenario, the perceived poor quality user indicator is used to evaluate the proportion of perceived poor quality users in the scenario, the data service perception poor indicator is used to evaluate the proportion of data service perception poor quality in the scenario, the voice service perception poor indicator is used to evaluate the proportion of voice service perception poor quality in the scenario, the backflow indicator is used to evaluate the proportion of high backflow cells, the diversion indicator is used to evaluate the diversion ratio in the scenario, the terminal development indicator is used to evaluate the proportion of new terminal users in the scenario, and the residence indicator is used to evaluate the residence ratio.

[0127] In step S2, for the matrix M, the vector X and the real number λ, if MX=λX, where X≠0, then λ is called the eigenvalue of the matrix M, and the vector X is called the eigenvector of the matrix M corresponding to the eigenvalue λ;

[0128] The eigenvalues ​​and eigenvectors are calculated using the R language. The syntax is eigen(M), where M is the matrix for which the eigenvalues ​​and eigenvectors need to be calculated.

[0129] In step S3, for an n-order reciprocal matrix M, where m ij >0, m ii =1, the matrix M is consistent if and only if the eigenvalue λ=n;

[0130] Since it is difficult for the matrix to have complete consistency when the dimension n is large, it is determined by testing whether it approximately satisfies the consistency. The calculation method of the consistency index CI is as follows:

[0131]

[0132] Consistency test coefficient:

[0133] The consistency test coefficient CR is the ratio of the consistency index CI to the random consistency index RI, and the calculation formula is as follows:

[0134]

[0135] Among them, the value of the random consistency index RI is related to the dimension of the judgment matrix; the values ​​of the random consistency index RI corresponding to matrix dimensions 1, 2, 3 and 4 are 0, 0, 0.58 and 0.9 respectively; see Table 1 for details.

[0136] Table 1 Random consistency index values

[0137] Matrix dimensions 1 2 3 4 RI 0 0 0.58 0.9

[0138] If the consistency test coefficient CR is less than or equal to 0.1, the judgment matrix is ​​considered to be consistent, and the order of the indicator weights represented by its weight vector is reasonable and effective. Otherwise, the judgment matrix is ​​considered to be inconsistent and needs to be adjusted until the consistency test coefficient is less than or equal to 0.1.

[0139] Table 2 Description of relative importance assignment

[0140]

[0141] In the first-layer indicators, 5G network B2 is more important than 4G network B1. Therefore, in the judgment matrix AB, the value assigned to 5G network B2 relative to 4G network B1 is 3, and the value assigned to 4G network B1 relative to 5G network B2 is 1 / 3. The judgment matrix AB and weight vector are shown in the following table:

[0142] Table 3 Judgment matrix AB and weight vector

[0143]

[0144]

[0145] Its maximum eigenvalue = 2, and the random consistency index CI of the two-dimensional matrix is ​​zero. There is no need to calculate CR, and it can be directly judged that the matrix AB has passed the consistency check;

[0146] The second layer indicators of the 4G network B1, the constructed judgment matrix B1-C1 and the weight vector are shown in the following table:

[0147] Table 4 Judgment matrix B1-C1 and weight vector

[0148] B1-C1 C11 C12 C13 C14 W C11 1 1 / 3 1 / 3 1 / 5 0.07353 C12 3 1 1 1 / 7 0.14652 C13 3 1 1 1 / 5 0.15578 C14 5 7 5 1 0.62416

[0149] Its maximum eigenvalue λ max =4.218, CI=0.218 / 3=0.073, CR=0.083; CR<0.1, so the judgment matrix B1-C1 passed the consistency check;

[0150] The third level indicators of the quality analysis C11, the constructed judgment matrix C11-D11 and the weight vector are shown in the following table:

[0151] Table 5 Judgment matrix C11-D11 and weight vector

[0152] C11-D11 D111 D112 W D111 1 1 / 7 0.125 D112 7 1 0.875

[0153] Its maximum eigenvalue λ max =2, CI = 0, the random consistency index of the two-dimensional matrix is ​​zero, there is no need to calculate CR, and it is directly judged that the matrix AB has passed the consistency check;

[0154] The third-level indicators of the complainant C12, the constructed judgment matrix C12-D12 and the weight vector are shown in the following table:

[0155] Table 6 Judgment matrix C12-D12 and weight vector

[0156] C12-D12 D121 D122 W D121 1 1 / 3 0.25 D122 3 1 0.75

[0157] Its maximum eigenvalue λ max=2, the random consistency index CI of the two-dimensional matrix is ​​zero, there is no need to calculate CR, and it is directly judged that the matrix AB has passed the consistency check;

[0158] The third-level indicators of the perception analysis C13, the constructed judgment matrix C13-D13 and the weight vector are shown in the following table:

[0159] Table 7 Judgment matrix C13-D13 and weight vector

[0160] C13-D13 D131 D132 D133 W D131 1 1 / 3 1 / 5 0.10616 D132 3 1 1 / 3 0.2605 D133 5 3 1 0.63335

[0161] Its maximum eigenvalue λ max =3.039, CI = 0.039 / 2 = 0.019, CR = 0.037; CR < 0.1, so the judgment matrix C13-D13 passed the consistency check

[0162] The third-level indicators of the collaborative analysis C14, the constructed judgment matrix C14-D14 and the weight vector are shown in the following table:

[0163] Table 8 Judgment matrix C14-D14 and weight vector

[0164] C14-D14 D141 D142 D143 D144 W D141 1 1 / 2 1 / 5 1 / 5 0.07537 D142 2 1 1 / 3 1 / 3 0.13758 D143 5 3 1 1 0.39352 D144 5 3 1 1 0.39352

[0165] Its maximum eigenvalue λ max =4.004, CI=0.004 / 3=0.001, CR=0.002; CR<0.1, so the judgment matrix C14-D14 passed the consistency check;

[0166] The second layer indicators of the 5G network B2, the constructed judgment matrix B2-C2 and the weight vector are shown in the following table:

[0167] Table 9 Judgment matrix B2-C2 and weight vector

[0168] B2-C2 C21 C22 C23 C24 W C21 1 1 / 3 1 / 5 1 / 5 0.068 C22 3 1 1 / 3 1 / 4 0.14198 C23 5 3 1 1 0.38001 C24 5 4 1 1 0.41001

[0169] Its maximum eigenvalue λ max =4.076, CI=0.076 / 3=0.025, CR=0.029; CR<0.1, so the judgment matrix B2-C2 passed the consistency check;

[0170] The third level indicators of the quality analysis C21, the constructed judgment matrix C21-D21 and the weight vector are shown in the following table:

[0171] Table 10 Judgment matrix C21-D21 and weight vector

[0172] C21-D21 D211 D212 W D211 1 1 / 3 0.25 D212 3 1 0.75

[0173] Its maximum eigenvalue λ max=2, the random consistency index CI of the two-dimensional matrix is ​​zero, there is no need to calculate CR, and it is directly judged that the matrix C21-D21 has passed the consistency check;

[0174] The third-level indicators of the complainant C22, the constructed judgment matrix C22-D22 and the weight vector are shown in the following table:

[0175] Table 11 Judgment matrix C22-D22 and weight vector

[0176] C22-D22 D221 D222 W D221 1 1 / 7 0.125 D222 7 1 0.875

[0177] Its maximum eigenvalue λ max =2, the random consistency index CI of the two-dimensional matrix is ​​zero, there is no need to calculate CR, and it is directly judged that the matrix C22-D22 has passed the consistency check;

[0178] The third-level indicators of the perception analysis C23, the constructed judgment matrix C23-D23 and the weight vector are shown in the following table:

[0179] Table 12 Judgment matrix C23-D23 and weight vector

[0180] C23-D23 D231 D232 D233 W D231 1 1 / 3 1 / 5 0.10616 D232 3 1 1 / 3 0.2605 D233 5 3 1 0.63335

[0181] Its maximum eigenvalue λ max =4.076, CI=0.076 / 3=0.025, CR=0.029; CR<0.1, so the judgment matrix C23-D23 passed the consistency check;

[0182] The third-level indicators of the collaborative analysis C24, the constructed judgment matrix C24-D24 and the weight vector are shown in the following table:

[0183] Table 13 Judgment matrix C24-D24 and weight vector

[0184]

[0185]

[0186] Its maximum eigenvalue λ max =4.118, CI=0.118 / 3=0.039, CR=0.045; CR<0.1, so the judgment matrix C24-D24 passed the consistency check.

[0187] In step S4, the comprehensive weights of the final indicators are shown in the following table:

[0188] Table 14 Comprehensive indicators and corresponding weights

[0189]

[0190]

[0191] Based on the analysis of the actual situation of the scenario and the macro environment, the health of each indicator is assigned a value and combined with the comprehensive weight to obtain the health score of each indicator:

[0192] The mapping relationship between the original values ​​of various indicators and health is as follows (the threshold value can be modified based on the actual situation of the existing network and the experience of experts).

[0193] Table 15 Index definition level value range

[0194]

[0195] In step S5, the corresponding relationship between the index score and the health degree is shown in the following table:

[0196] Table 16 Correspondence between indicator scores and health level

[0197]

[0198]

[0199] According to the corresponding relationship in the above table, the original values ​​of the 4G and 5G indicators are input to obtain the health level corresponding to the 4G and 5G indicators, and input into the evaluation system. Example 1 is shown in the following table:

[0200] Table 17 Comprehensive indicator health evaluation

[0201]

[0202]

[0203] According to the comprehensive health evaluation of each indicator, the total health coefficient of the entire scene is 3.523, which is within the numerical range of the fourth level of health. Therefore, the health of the scene is determined to be the fourth level of health.

[0204] At this point, the overall health of the scenario has been evaluated through the hierarchical analysis method. This evaluation comprehensively considers the two types of network standards, 4G and 5G, in the scenario, and specifically involves quality analysis, complaint analysis, perception analysis, and collaborative analysis, thereby comprehensively measuring the health of the scenario with a combination of qualitative and quantitative methods.

[0205] The system for implementing scenario-based wireless network quality assessment based on hierarchical analysis method includes an assessment index system construction module, a judgment matrix processing module, a comprehensive weight calculation module, a measurement rule module and a measurement module;

[0206] The evaluation index system construction module is used to construct the scenario health evaluation index system. By assigning the relative importance of each layer of indicators mentioned above, a judgment matrix is ​​constructed for each layer of indicators.

[0207] The first-tier indicators include 4G network B1 and 5G network B2;

[0208] The second-layer indicators of 4G network B1 include quality analysis C11, complaint molecules C12, perception analysis C13 and collaborative analysis C14;

[0209] The third-layer indicators of quality analysis C11 include no direct coverage D111 and poor network quality D112. The third-layer indicators of complaint analysis C12 include complaint hotspots D121 and repeated complaints D122. The third-layer indicators of perception analysis C13 include users with poor perceived quality D131, poor data service perception D132, and poor voice service perception D133. The third-layer indicators of collaborative analysis C14 include backflow D141, diversion D142, terminal development D143, and residence D144.

[0210] The second-layer indicators of 5G network B2 all include quality analysis C21, complaint element C22, perception analysis C23 and collaborative analysis C24;

[0211] The third-layer indicators of quality analysis C21 include no direct coverage D211 and poor network quality D212. The third-layer indicators of complaint analysis C22 include complaint hotspots D221 and repeated complaints D222. The third-layer indicators of perception analysis C23 include users with poor perceived quality D231, poor data service perception D232, and poor voice service perception D233. The third-layer indicators of collaborative analysis C24 include backflow D241, diversion D242, terminal development D243, and residence D244D.

[0212] The judgment matrix processing module is responsible for calculating the eigenvalue λ of the judgment matrix and the eigenvector corresponding to the eigenvalue, and taking the maximum eigenvalue λ max and its corresponding eigenvector, and normalize the eigenvector to obtain the weight vector W;

[0213] The judgment matrix is ​​checked for consistency. If it passes the consistency check, it is considered that the ranking of the indicator weights represented by the weight vector W is reasonable and effective;

[0214] If the consistency check fails, the judgment matrix is ​​adjusted until it passes the consistency check;

[0215] Comprehensive weight calculation module, used to calculate the comprehensive weight of the final indicator;

[0216] The comprehensive weight of the final indicator is the ratio of the weight of the corresponding third-layer indicator to the sum of the weights of the third-layer indicators in the 4G network and 5G network scenarios;

[0217] The measurement rule module is used to formulate the measurement rules for indicator health and scenario health;

[0218] In the measurement rules of the indicator health, the health is divided into 5 levels, and the corresponding numerical values ​​are 1 point for low health, 2 points for lower health, 3 points for medium health, 4 points for higher health, and 5 points for high health;

[0219] In the measurement rules of scenario health, the health coefficient is not less than 0.5 and less than 1.5 for the first level of health, the health coefficient is not less than 1.5 and less than 2.5 for the second level of health, the health coefficient is not less than 2.5 and less than 3.5 for the third level of health, the health coefficient is not less than 3.5 and less than 4.5 for the fourth level of health, and the health coefficient is not less than 4.5 and less than 5 for the fifth level of health;

[0220] The measurement module is used to score each indicator except the no direct coverage indicator according to the measurement rules of the indicator health. When scoring the no direct coverage indicator, it is determined whether the scene has cell coverage. If there is cell coverage, it is 1, otherwise it is 0;

[0221] Multiply the score of each indicator by the corresponding weight to obtain the final score of each indicator; add up the final scores of each indicator to obtain the health coefficient of the entire scene; according to the measurement rules of scene health, obtain the health level of the entire scene.

[0222] The device for implementing scenario-based wireless network quality assessment based on hierarchical analysis method includes a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the above method steps when executing the computer program.

[0223] The readable storage medium stores a computer program, and when the computer program is executed by a processor, the above method steps are implemented.

[0224] The embodiment described above is only one specific implementation of the present invention. Common changes and substitutions made by those skilled in the art within the scope of the technical solution of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for implementing scenario-based wireless network quality assessment based on hierarchical analysis method, characterized in that: The following steps are involved: Step S1, constructing a scenario health evaluation indicator system, where the first-level indicators include 4G network B1 and 5G network B2; The second-layer indicators of 4G network B1 include quality analysis C11, complaint molecules C12, perception analysis C13 and collaborative analysis C14; The third-layer indicators of quality analysis C11 include no direct coverage D111 and poor network quality D112. The third-layer indicators of complaint analysis C12 include complaint hotspots D121 and repeated complaints D122. The third-layer indicators of perception analysis C13 include users with poor perceived quality D131, poor data service perception D132, and poor voice service perception D133. The third-layer indicators of collaborative analysis C14 include backflow D141, diversion D142, terminal development D143, and residence D144. The second-layer indicators of 5G network B2 all include quality analysis C21, complaint element C22, perception analysis C23 and collaborative analysis C24; The third-layer indicators of quality analysis C21 include no direct coverage D211 and poor network quality D212. The third-layer indicators of complaint analysis C22 include complaint hotspots D221 and repeated complaints D222. The third-layer indicators of perception analysis C23 include users with poor perceived quality D231, poor data service perception D232, and poor voice service perception D233. The third-layer indicators of collaborative analysis C24 include backflow D241, diversion D242, terminal development D243, and residence D244D. By assigning the relative importance of each of the above indicators, a judgment matrix is ​​constructed for each indicator layer; Step S2, calculate the eigenvalue λ of the judgment matrix and the eigenvector corresponding to the eigenvalue, and take the maximum eigenvalue λ max and its corresponding eigenvector, and normalize the eigenvector to obtain the weight vector W; Step S3, performing consistency check on the judgment matrix. If the consistency check passes, it is considered that the ranking of the indicator weights represented by the weight vector W is reasonable and effective; If the consistency check fails, the judgment matrix is ​​adjusted until it passes the consistency check; Step S4, calculating the comprehensive weight of the final indicator; The comprehensive weight of the final indicator is the ratio of the weight of the corresponding third-layer indicator to the sum of the weights of the third-layer indicators in the 4G network and 5G network scenarios; Step S5, formulate the measurement rules of the indicator health, divide the health into 5 levels, and the corresponding numerical values ​​are 1 point for low health, 2 points for lower health, 3 points for medium health, 4 points for higher health, and 5 points for high health; According to the measurement rules of indicator health, each indicator except the no direct coverage indicator is scored. When scoring the no direct coverage indicator, it is determined whether the scene has cell coverage. If there is cell coverage, it is 1, otherwise it is 0; Multiply the score of each indicator by the corresponding weight to get the final score of each indicator; add the final scores of each indicator to get the health coefficient of the entire scene; According to the measurement rules of scene health, the health level of the entire scene is obtained; the measurement rules of scene health are as follows: The health coefficient is not less than 0.5 and less than 1.5, which is the first level of health. The health coefficient is not less than 1.5 and less than 2.5, which is the second level of health. The health coefficient is not less than 2.5 and less than 3.5, which is the third level of health. The health coefficient is not less than 3.5 and less than 4.5, which is the fourth level of health. The health coefficient is not less than 4.5 and less than 5, which is the fifth level of health.

2. The method for implementing scenario-based wireless network quality assessment based on hierarchical analysis method according to claim 1 is characterized in that: In the step S1, the no direct coverage indicator is used to determine whether the scenario has cell coverage, the poor network quality indicator is used to evaluate the proportion of poor quality cells in the scenario, the complaint hotspot indicator is used to evaluate the number of hotspot complaints in the scenario, the complaint repetition indicator is used to evaluate the number of repeated complaints in the scenario, the perceived poor quality user indicator is used to evaluate the proportion of perceived poor quality users in the scenario, the data service perception poor indicator is used to evaluate the proportion of data service perception poor quality in the scenario, the voice service perception poor indicator is used to evaluate the proportion of voice service perception poor quality in the scenario, the backflow indicator is used to evaluate the proportion of high backflow cells, the diversion indicator is used to evaluate the diversion ratio in the scenario, the terminal development indicator is used to evaluate the proportion of new terminal users in the scenario, and the residence indicator is used to evaluate the residence ratio.

3. The method for implementing scenario-based wireless network quality assessment based on hierarchical analysis method according to claim 1 is characterized in that: In step S2, for the matrix M, the vector X and the real number λ, if MX=λX, where X≠0, then λ is called the eigenvalue of the matrix M, and the vector X is called the eigenvector of the matrix M corresponding to the eigenvalue λ; The eigenvalues ​​and eigenvectors are calculated using the R language. The syntax is eigen(M), where M is the matrix for which the eigenvalues ​​and eigenvectors need to be calculated.

4. The method for implementing scenario-based wireless network quality assessment based on hierarchical analysis method according to claim 3 is characterized in that: In step S3, for an n-order reciprocal matrix M, where m ij >0, m ii =1, the calculation method of consistency index CI is as follows: Consistency test coefficient: The consistency test coefficient CR is the ratio of the consistency index CI to the random consistency index RI, and the calculation formula is as follows: Among them, the value of the random consistency index RI is related to the dimension of the judgment matrix; the values ​​of the random consistency index RI corresponding to matrix dimensions 1, 2, 3 and 4 are 0, 0, 0.58 and 0.9 respectively; if the consistency test coefficient CR is less than or equal to 0.1, the judgment matrix is ​​considered to be consistent, and the order of the indicator weights represented by its weight vector is reasonable and effective, otherwise the judgment matrix is ​​considered to be inconsistent and the judgment matrix needs to be adjusted until the consistency test coefficient is less than or equal to 0.

1.

5. The method for implementing scenario-based wireless network quality assessment based on hierarchical analysis method according to claim 4 is characterized in that: In the first-layer indicators, 5G network B2 is more important than 4G network B1. Therefore, in the judgment matrix AB, the value assigned to 5G network B2 relative to 4G network B1 is 3, and the value assigned to 4G network B1 relative to 5G network B2 is 1 / 3. The judgment matrix AB and weight vector are shown in the following table: Its maximum eigenvalue = 2, and the random consistency index CI of the two-dimensional matrix is ​​zero. There is no need to calculate CR, and it can be directly judged that the matrix AB has passed the consistency check; The second layer indicators of the 4G network B1, the constructed judgment matrix B1-C1 and the weight vector are shown in the following table: Its maximum eigenvalue λ max =4.218, CI=0.218 / 3=0.073, CR=0.083; CR<0.1, so the judgment matrix B1-C1 passed the consistency check; The third level indicators of the quality analysis C11, the constructed judgment matrix C11-D11 and the weight vector are shown in the following table: Its maximum eigenvalue λ max =2, CI = 0, the random consistency index of the two-dimensional matrix is ​​zero, there is no need to calculate CR, and it is directly judged that the matrix AB has passed the consistency check; The third-level indicators of the complainant C12, the constructed judgment matrix C12-D12 and the weight vector are shown in the following table: Its maximum eigenvalue λ max =2, the random consistency index CI of the two-dimensional matrix is ​​zero, there is no need to calculate CR, and it is directly judged that the matrix AB has passed the consistency check; The third-level indicators of the perception analysis C13, the constructed judgment matrix C13-D13 and the weight vector are shown in the following table: Its maximum eigenvalue λ max =3.039, CI = 0.039 / 2 = 0.019, CR = 0.037; CR < 0.1, so the judgment matrix C13-D13 passed the consistency check The third-level indicators of the collaborative analysis C14, the constructed judgment matrix C14-D14 and the weight vector are shown in the following table: Its maximum eigenvalue λ max =4.004, CI=0.004 / 3=0.001, CR=0.002; CR<0.1, so the judgment matrix C14-D14 passed the consistency check; The second layer indicators of the 5G network B2, the constructed judgment matrix B2-C2 and the weight vector are shown in the following table: Its maximum eigenvalue λ max =4.076, CI=0.076 / 3=0.025, CR=0.029; CR<0.1, so the judgment matrix B2-C2 passed the consistency check; The third level indicators of the quality analysis C21, the constructed judgment matrix C21-D21 and the weight vector are shown in the following table: Its maximum eigenvalue λ max =2, the random consistency index CI of the two-dimensional matrix is ​​zero, there is no need to calculate CR, and it is directly judged that the matrix C21-D21 has passed the consistency check; The third-level indicators of the complainant C22, the constructed judgment matrix C22-D22 and the weight vector are shown in the following table: Its maximum eigenvalue λ max =2, the random consistency index CI of the two-dimensional matrix is ​​zero, there is no need to calculate CR, and it is directly judged that the matrix C22-D22 has passed the consistency check; The third-level indicators of the perception analysis C23, the constructed judgment matrix C23-D23 and the weight vector are shown in the following table: Its maximum eigenvalue λ max =4.076, CI=0.076 / 3=0.025, CR=0.029; CR<0.1, so the judgment matrix C23-D23 passed the consistency check; The third-level indicators of the collaborative analysis C24, the constructed judgment matrix C24-D24 and the weight vector are shown in the following table: Its maximum eigenvalue λ max =4.118, CI=0.118 / 3=0.039, CR=0.045; CR<0.1, so the judgment matrix C24-D24 passed the consistency check; In step S4, the comprehensive weight of the final indicator is shown in the above table.

6. The method for implementing scenario-based wireless network quality assessment based on hierarchical analysis method according to claim 2 is characterized in that: In step S5, the corresponding relationship between the indicator score and the health degree is shown in the above table.

7. A system for implementing scenario-based wireless network quality assessment based on hierarchical analysis method, characterized in that: It includes evaluation index system construction module, judgment matrix processing module, comprehensive weight calculation module, measurement rule module and measurement module; The evaluation index system construction module is used to construct the scenario health evaluation index system. By assigning the relative importance of each layer of indicators mentioned above, a judgment matrix is ​​constructed for each layer of indicators. Among them, the first-tier indicators include 4G network B1 and 5G network B2; The second-layer indicators of 4G network B1 include quality analysis C11, complaint molecules C12, perception analysis C13 and collaborative analysis C14; The third-layer indicators of quality analysis C11 include no direct coverage D111 and poor network quality D112. The third-layer indicators of complaint analysis C12 include complaint hotspots D121 and repeated complaints D122. The third-layer indicators of perception analysis C13 include users with poor perceived quality D131, poor data service perception D132, and poor voice service perception D133. The third-layer indicators of collaborative analysis C14 include backflow D141, diversion D142, terminal development D143, and residence D144. The second-layer indicators of 5G network B2 all include quality analysis C21, complaint element C22, perception analysis C23 and collaborative analysis C24; The third-layer indicators of quality analysis C21 include no direct coverage D211 and poor network quality D212. The third-layer indicators of complaint analysis C22 include complaint hotspots D221 and repeated complaints D222. The third-layer indicators of perception analysis C23 include users with poor perceived quality D231, poor data service perception D232, and poor voice service perception D233. The third-layer indicators of collaborative analysis C24 include backflow D241, diversion D242, terminal development D243, and residence D244D. The judgment matrix processing module is responsible for calculating the eigenvalue λ of the judgment matrix and the eigenvector corresponding to the eigenvalue, and taking the maximum eigenvalue λ max and its corresponding eigenvector, and normalize the eigenvector to obtain the weight vector W; The judgment matrix is ​​checked for consistency. If it passes the consistency check, it is considered that the ranking of the indicator weights represented by the weight vector W is reasonable and effective; If the consistency check fails, the judgment matrix is ​​adjusted until it passes the consistency check; Comprehensive weight calculation module, used to calculate the comprehensive weight of the final indicator; The comprehensive weight of the final indicator is the ratio of the weight of the corresponding third-layer indicator to the sum of the weights of the third-layer indicators in the 4G network and 5G network scenarios; The measurement rule module is used to formulate the measurement rules for indicator health and scenario health; In the measurement rules of the indicator health, the health is divided into 5 levels, and the corresponding numerical values ​​are 1 point for low health, 2 points for lower health, 3 points for medium health, 4 points for higher health, and 5 points for high health; In the measurement rules of scenario health, the health coefficient is not less than 0.5 and less than 1.5 for the first level of health, the health coefficient is not less than 1.5 and less than 2.5 for the second level of health, the health coefficient is not less than 2.5 and less than 3.5 for the third level of health, the health coefficient is not less than 3.5 and less than 4.5 for the fourth level of health, and the health coefficient is not less than 4.5 and less than 5 for the fifth level of health; The measurement module is used to score each indicator except the no direct coverage indicator according to the measurement rules of the indicator health. When scoring the no direct coverage indicator, it is determined whether the scene has cell coverage. If there is cell coverage, it is 1, otherwise it is 0; Multiply the score of each indicator by the corresponding weight to obtain the final score of each indicator; add up the final scores of each indicator to obtain the health coefficient of the entire scene; according to the measurement rules of scene health, obtain the health level of the entire scene.

8. A device for implementing scenario-based wireless network quality assessment based on hierarchical analysis method, characterized in that: It comprises a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the method steps as claimed in any one of claims 1 to 6 when executing the computer program.

9. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps according to any one of claims 1 to 6 are implemented.

Images