Smart city wireless network intelligent recommendation and scheduling method and system, and storage medium

Abstract

The invention discloses a smart city wireless network intelligent recommendation and scheduling method and system, and a storage medium. The method comprises the following steps: collecting public place Internet surfing code scanning data, analyzing Internet surfing user behaviors, personal basic information and features, and carrying out data preprocessing and feature extraction; obtaining an optimal solution network transmission protocol of the current scene and a scene user matching relation according to the extracted features and user behavior data analysis, and mapping the optimal solution network transmission protocol and the scene user matching relation into an analysis table; inputting clustering data corresponding to the analysis table into a random forest machine learning model, and predicting a probability value of an optimal solution of a scene application network protocol; the network transmission protocol can be intelligently analyzed and dispatched according to different places, normal transmission of network information under the condition of different safety degrees is guaranteed, the internet surfing experience of a user is enhanced, and reference is provided for network upgrading and optimization.

Description

Technical Field

[0001] The present invention relates to the field of wireless network technology, and in particular to a method, system and storage medium for intelligent recommendation and scheduling of wireless networks in a smart city. Background Art

[0002] In recent years, with the rapid development of the economy, people's business trips, tourism, learning exchanges and other activities between cities and counties have been increasing, which has put forward higher requirements for urban management issues such as urban infrastructure resource allocation, urban security, and crowd control, which has also brought a series of management problems. How to use big data resources and accurately and timely grasp the regional characteristics of different groups of people based on machine learning technology and analyze the travel patterns of people is of great significance to improving the management capabilities of smart cities, and has become one of the hot spots of current research and application.

[0003] With the promotion of venue Internet access codes, venues such as catering and accommodation, shopping, hospitals, schools, communities, cultural tourism, transportation, office buildings, etc. have deployed venue codes to access the Internet. Due to the daily life needs of residents, venue code data contains a large number of residents' travel records. It has the advantages of high sample size, low cost and wide coverage, and can record users' daily travel information more completely. Therefore, mining the user information contained in the venue code data and analyzing the user's travel patterns to identify the user's population type is a low-cost and high-efficiency means with broad application prospects. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the shortcomings of the prior art, the present invention provides a method, system and storage medium for intelligent recommendation and scheduling of wireless networks in a smart city, which can flexibly configure the scanning data of Internet access in places, analyze and match different scenarios and population characteristics and allocate different network protocols, complete network scheduling, improve user Internet experience, and solve the above-mentioned technical problems.

[0006] (II) Technical solution

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] On the one hand, the present application proposes a method for intelligent recommendation and scheduling of wireless networks in a smart city, comprising the following steps:

[0009] S1. Analyze the online scanning log data in public places to obtain the online user behavior, personal basic information and characteristics, and perform data preprocessing and feature extraction;

[0010] S2. Extract user behavior data from the log using data features as search conditions, analyze and obtain the optimal network transmission protocol for the current scenario and the matching relationship between the scenario user, map the matching relationship into an analysis table, and calculate the scenario and sub-scenario transmission protocol indicator usage data in the analysis table;

[0011] S3. Input the calculated scenario and sub-scenario transmission protocol indicator usage data into the random forest machine learning model to predict the final evaluation value of the transmission protocol.

[0012] Preferably, in S3, the indicator scoring items in the analysis table include: main scene, sub-scene, optimal solution protocol for the frequency characteristics of scanning code entry and exit, main scene crowd characteristics and final evaluation value. The transmission protocol with the largest final evaluation value is taken as the optimal solution probability value, and then the network protocol score is obtained by accumulating the indicator scoring items in the analysis table.

[0013] Preferably, the main scene crowd characteristics include optimal residence state and crowd type.

[0014] Preferably, the K-means++ algorithm is used to classify the scene and sub-scene log data in the order of the analysis table scores.

[0015] Preferably, the calculation process of the K-means++ algorithm is:

[0016] (1) Randomly select a center u among the data points i ;

[0017] (2) For each data point x that has not yet been selected, calculate the distance between z and the closest center that has already been selected

[0018] (3) Use the weighted probability distribution to randomly select a new data point as the new center, where the probability of selecting point z is is proportional to;

[0019] (4) Repeat steps (2) and (3) until K centers are selected;

[0020] (5) For the selected initial centers, use standard K as the mean clustering.

[0021] Preferably, the clustering data generation network protocol scheduling uses configuration file management and the emergency handling process for possible scene abnormal factors is:

[0022] (1) To address the problem of poor user experience caused by using only the Wi-Fi emergency transmission protocol due to possible abnormal factors in the scene, the cluster data is first generated into a configuration file through command execution and temporarily marked;

[0023] (2) Temporarily execute the traffic protocol scheduling instruction according to the mark and send it to the routing server for switching. At the same time, all transmission data during the abnormal time is stored separately in the data block of the server. When the abnormality is restored, the temporary mark is cancelled and the normal configuration file is read and executed;

[0024] (3) If part of the data block of the server is lost, erasure coding fault tolerance technology is used to recover the data.

[0025] Preferably, in the erasure code fault tolerance technology, the erasure code encoding algorithm is MDS encoding.

[0026] Preferably, the random forest machine learning model is specifically:

[0027]

[0028] Where J represents the total number of key scoring indicators, including the most used protocols in the scenario, the type of resident personnel, and the status of the resident personnel. j represents the index of J. represents the sampling probability of the jth key scoring indicator, |D| represents the total number of historical data of all key scoring indicators, |D j | represents the total number of data in |D| that meet the jth key scoring indicator, H(j) represents the evaluation value of the jth key scoring indicator for the corresponding scenario or sub-scenario scoring process link, and H represents the weighted average of all H(j).

[0029] On the other hand, the present application proposes a system for implementing the method based on wireless network intelligent recommendation and scheduling, including:

[0030] The data collection and preprocessing module is used to collect the QR code scanning data in public places, analyze the Internet user behavior, personal basic information and characteristics, and perform preprocessing;

[0031] The feature extraction module is used to extract features from the data processed by the data collection and preprocessing modules;

[0032] The mapping analysis module is used to extract user behavior data in the log using data features as search conditions, and analyze to obtain the optimal solution for the current scenario; map the network transmission protocol and scenario user matching relationship into an analysis table, and calculate the scenario and sub-scenario transmission protocol indicator usage data in the analysis table;

[0033] The random forest model prediction module is used to input the calculated scene and sub-scenario transmission protocol indicator usage data into the random forest machine learning model to predict the final evaluation value of the transmission protocol.

[0034] On the other hand, the present application further proposes a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the method based on wireless network intelligent recommendation and scheduling as described above.

[0035] Compared with the prior art, the present invention provides a method, system and storage medium for intelligent recommendation and scheduling of wireless networks in a smart city, which has the following beneficial effects:

[0036] The present invention comprises: collecting the data of scanning QR codes in public places to analyze the behaviors, basic personal information and characteristics of Internet users, and performing data preprocessing and feature extraction; obtaining the optimal solution network transmission protocol and the matching relationship between scene users according to the extracted characteristics and in combination with the analysis of user behavior data, and mapping the matching relationship into an analysis table; inputting the clustering data corresponding to the analysis table into a machine learning model of a random forest, predicting the probability value of the optimal solution of the scene application network protocol, and then obtaining the network protocol score according to the accumulation of the scores of each indicator in the analysis table; orderly classifying the scenes by using the K-means++ algorithm according to the optimal solution protocol score; generating a configuration file management for the use of the clustering data for network protocol scheduling and an emergency processing flow for possible abnormal factors in the scene; backing up the configuration file by using the erasure code fault-tolerant technology, and when part of the data blocks stored in the server are lost, the lost data can be restored by the remaining data blocks and the check blocks, thereby ensuring the integrity and reliability of the data.

[0037] The present invention can intelligently analyze and schedule network transmission protocols according to different locations, ensure the normal transmission of network information under different security levels, enhance the user's Internet experience, and provide a reference for network upgrade and optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments, features, and aspects of the disclosure and, together with the description, serve to explain the principles of the disclosure.

[0039] Figure 1 A flow chart of a method for intelligent recommendation and scheduling of wireless networks in smart cities according to the present invention;

[0040] Figure 2 Schematic diagram of data block encoding and decoding process in the present invention;

[0041] Figure 3 Schematic diagram of the calculation principle of the K-means++ algorithm in the present invention. DETAILED DESCRIPTION

[0042] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0043] In the description of the present invention, unless otherwise specified, "plurality" means two or more than two; the orientations or positional relationships indicated by the terms "upper", "lower", "left", "right", "inner", "outer", "front end", "rear end", "head", "tail", etc. are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0044] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0045] First embodiment

[0046] See also Figure 1 , a method for intelligent recommendation and scheduling of wireless networks in a smart city, comprising the following steps:

[0047] S1. Analyze the online scanning log data in public places to obtain the online user behavior, personal basic information and characteristics, and perform data preprocessing and feature extraction. The data preprocessing methods include but are not limited to sorting, cleaning and conversion;

[0048] S2. Extract user behavior data from the log using data features as search conditions, analyze and obtain the optimal solution network transmission protocol for the current scenario and the matching relationship between the scenario user, and map the matching relationship into an analysis table, visualize the matching relationship for easy viewing and calculation, and calculate the scenario and sub-scenario transmission protocol indicator usage data of the analysis table;

[0049] S3, input the calculated scenario and sub-scenario transmission protocol indicator usage data into the random forest machine learning model to predict the final evaluation value of the transmission protocol;

[0050] The random forest machine learning model is a common and very powerful machine learning algorithm used for classification and regression tasks. It belongs to the ensemble learning method and works mainly by building multiple decision trees and combining the prediction results of these decision trees. It can be applied to various data types and problems.

[0051] In this embodiment, the random forest machine learning model is specifically:

[0052]

[0053] Where J represents the total number of key scoring indicators, including the most used protocols in the scenario, the type of resident personnel, and the status of the resident personnel. j represents the index of J. represents the sampling probability of the jth key scoring indicator, |D| represents the total number of historical data of all key scoring indicators, |D j | represents the total number of data in |D| that meet the jth key scoring indicator, H(j) represents the evaluation value of the jth key scoring indicator for the corresponding scenario or sub-scenario scoring process link, and H represents the weighted average of all H(j).

[0054] Second embodiment

[0055] The difference from the first embodiment is that:

[0056] In this embodiment, in S3, the indicator scoring items in the analysis table include: main scene, sub-scene, optimal solution protocol for the frequency characteristics of scanning code entry and exit, main scene crowd characteristics and final evaluation value. The transmission protocol with the largest final evaluation value is used as the optimal solution probability value, and then the network protocol score is obtained by accumulating the indicator scoring items in the analysis table. The main scene crowd characteristics include the optimal residence state and crowd type.

[0057] In the analysis table, the scene and sub-scene log data are sorted using the K-means++ algorithm. Figure 3 As shown, the calculation process of the K-means++ algorithm is:

[0058] (1) Randomly select a center ui among the data points;

[0059] (2) For each data point x that has not yet been selected, calculate the distance between z and the closest center that has already been selected

[0060] (3) Use the weighted probability distribution to randomly select a new data point as the new center, where the probability of selecting point z is is proportional to;

[0061] (4) Repeat steps (2) and (3) until K centers are selected;

[0062] (5) For the selected initial centers, use standard K as the mean clustering.

[0063] Third embodiment

[0064] The difference from the first embodiment is that:

[0065] In this embodiment, the clustering data generation network protocol scheduling uses configuration file management and the emergency handling process for possible scene abnormal factors is as follows:

[0066] (1) To address the problem of poor user experience caused by using only the Wi-Fi emergency transmission protocol due to possible abnormal factors in the scene, the cluster data is first generated into a configuration file through command execution and temporarily marked;

[0067] (2) Temporarily execute the traffic protocol scheduling instruction according to the mark and send it to the routing server for switching. At the same time, all transmission data during the abnormal time is stored separately in the data block of the server. When the abnormality is restored, the temporary mark is cancelled and the normal configuration file is read and executed;

[0068] (3) If part of the data block of the server is lost, the erasure code fault tolerance technology is used to recover the data. In the erasure code fault tolerance technology, the erasure code encoding algorithm is MDS encoding.

[0069] like Figure 2 As shown in the figure, the principle and implementation process of erasure code fault tolerance technology:

[0070] Among them, original data (OriginalData): the real original data information stored by the user;

[0071] Erasure Code: According to the erasure code algorithm, the remaining data is calculated using the original data information. These data are derived from the original data, ensuring that the original data can be restored in time when errors occur.

[0072] Configuration parameters (k, m, w): where k is the number of data blocks, m is the number of erasure code blocks, and w is the code word length. Element: In the process of correction code encoding, the element is the atomic unit that carries the original data information or redundant data information, and is also the basic unit of erasure code calculation;

[0073] Encode: The process of calculating the erasure code data using the original data according to the erasure code algorithm is called encoding;

[0074] Decoding: When the original data is wrong, the process of calculating and restoring the original data using the erasure code data corresponding to the original data according to the erasure code encoding mechanism is called decoding;

[0075] Update: If the original data is modified appropriately according to the user's needs, the corresponding erasure code also needs to be recalculated. This calculation process is called update.

[0076] Fault Tolerance Rate: A set of correction code calculations contains k original data blocks and m erasure code blocks. The original data blocks and erasure code blocks are stored on different disks. After that, this set of original data blocks and erasure code blocks can allow no more than m blocks to fail. m / k+m is the fault tolerance rate.

[0077] MDS (Maximum Distance Separable) coding: In the case of any number of original data blocks and m correction blocks, the coding method that can recover any m errors in these km blocks is MDS coding. MDS coding achieves the theoretical optimality in terms of storage utilization.

[0078] The working process of erasure coding is introduced by taking three original data blocks and two erasure code blocks as examples. L0, L1, and L2 are the original data blocks of the user. After being encoded with erasure code, two erasure code blocks C0 and C1 are constructed. When L1 and C0 are wrong, the remaining blocks L0, L2, and C1 are used, and the decoding algorithm corresponding to the erasure code is adopted to obtain the original data corresponding to the wrong data.

[0079] On the other hand, the present application proposes a system for implementing the method based on wireless network intelligent recommendation and scheduling, including:

[0080] The data collection and preprocessing module is used to collect the QR code scanning data in public places, analyze the Internet user behavior, personal basic information and characteristics, and perform preprocessing;

[0081] The feature extraction module is used to extract features from the data processed by the data collection and preprocessing modules;

[0082] The mapping analysis module is used to extract user behavior data in the log using data features as search conditions, and analyze to obtain the optimal solution for the current scenario; map the network transmission protocol and scenario user matching relationship into an analysis table, and calculate the scenario and sub-scenario transmission protocol indicator usage data in the analysis table;

[0083] The random forest model prediction module is used to input the calculated scene and sub-scenario transmission protocol indicator usage data into the random forest machine learning model to predict the final evaluation value of the transmission protocol.

[0084] On the other hand, the present application further proposes a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the method based on wireless network intelligent recommendation and scheduling as described above.

[0085] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent recommendation and scheduling of wireless networks in smart cities, characterized by: The steps include: S1. Analyze the online scanning log data in public places to obtain the online user behavior, personal basic information and characteristics, and perform data preprocessing and feature extraction; S2. Extract user behavior data from the log using data features as search conditions, analyze and obtain the optimal network transmission protocol for the current scenario and the matching relationship between the scenario user, map the matching relationship into an analysis table, and calculate the scenario and sub-scenario transmission protocol indicator usage data in the analysis table; S3. Input the calculated scenario and sub-scenario transmission protocol indicator usage data into the random forest machine learning model to predict the final evaluation value of the transmission protocol.

2. The method for intelligent recommendation and scheduling of wireless networks in a smart city according to claim 1, characterized in that: In S3, the indicator scoring items in the analysis table include: main scene, sub-scene, optimal solution protocol for the frequency characteristics of scanning code entry and exit, main scene crowd characteristics and final evaluation value. The transmission protocol with the largest final evaluation value is taken as the optimal solution probability value, and then the network protocol score is obtained by accumulating the indicator scoring items in the analysis table.

3. The method for intelligent recommendation and scheduling of wireless networks in a smart city according to claim 2, characterized in that: The main scene crowd characteristics include the optimal residence state and crowd type.

4. The method for intelligent recommendation and scheduling of wireless networks in a smart city according to claim 2, characterized in that: The K-means++ algorithm is used to classify the scene and sub-scene log data in the analysis table.

5. The method for intelligent recommendation and scheduling of wireless networks in a smart city according to claim 4, characterized in that: The calculation process of the K-means++ algorithm is: (1) Randomly select a center u among the data points i ; (2) For each data point x that has not yet been selected, calculate the distance between z and the closest center that has already been selected (3) Use the weighted probability distribution to randomly select a new data point as the new center, where the probability of selecting point z is is proportional to; (4) Repeat steps (2) and (3) until K centers are selected; (5) For the selected initial centers, use standard K as the mean clustering.

6. The method for intelligent recommendation and scheduling of wireless networks in a smart city according to claim 5, characterized in that: The clustering data generation network protocol scheduling uses configuration file management and the emergency response process for possible scene abnormal factors is as follows: (1) To address the problem of poor user experience caused by using only the Wi-Fi emergency transmission protocol due to possible abnormal factors in the scene, the cluster data is first generated into a configuration file through command execution and temporarily marked; (2) Temporarily execute the traffic protocol scheduling instruction according to the mark and send it to the routing server for switching. At the same time, all transmission data during the abnormal time is stored separately in the data block of the server. When the abnormality is restored, the temporary mark is cancelled and the normal configuration file is read and executed; (3) If part of the data block of the server is lost, erasure coding fault tolerance technology is used to recover the data.

7. The method for intelligent recommendation and scheduling of wireless networks in a smart city according to claim 6, characterized in that: In the erasure code fault tolerance technology, the erasure code encoding algorithm is MDS coding.

8. The method for intelligent recommendation and scheduling of wireless networks in a smart city according to claim 1, characterized in that: The random forest machine learning model is specifically: Where J represents the total number of key scoring indicators, including the most used protocols in the scenario, the type of resident personnel, and the status of resident personnel, and j represents the index of J. represents the sampling probability of the jth key scoring indicator, |D| represents the total number of historical data of all key scoring indicators, |D j | represents the total number of data in |D| that meet the jth key scoring indicator, H(j) represents the evaluation value of the jth key scoring indicator for the corresponding scenario or sub-scenario scoring process link, and H represents the weighted average of all H(j).

9. A system for implementing the method based on wireless network intelligent recommendation and scheduling as described in any one of claims 1 to 8, characterized in that: include: The data collection and preprocessing module is used to collect the QR code scanning data in public places, analyze the Internet user behavior, personal basic information and characteristics, and perform preprocessing; The feature extraction module is used to extract features from the data processed by the data collection and preprocessing modules; The mapping analysis module is used to extract user behavior data in the log using data features as search conditions, and analyze to obtain the optimal solution for the current scenario; map the network transmission protocol and scenario user matching relationship into an analysis table, and calculate the scenario and sub-scenario transmission protocol indicator usage data in the analysis table; The random forest model prediction module is used to input the calculated scene and sub-scenario transmission protocol indicator usage data into the random forest machine learning model to predict the final evaluation value of the transmission protocol.

10. A computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the method based on wireless network intelligent recommendation and scheduling according to any one of claims 1 to 8.

Images