CN114205817A - Wireless local area network access method, system and electronic equipment - Google Patents

Wireless local area network access method, system and electronic equipment Download PDF

Info

Publication number
CN114205817A
CN114205817A CN202111427127.8A CN202111427127A CN114205817A CN 114205817 A CN114205817 A CN 114205817A CN 202111427127 A CN202111427127 A CN 202111427127A CN 114205817 A CN114205817 A CN 114205817A
Authority
CN
China
Prior art keywords
communication data
wireless
feature map
score
feature
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111427127.8A
Other languages
Chinese (zh)
Inventor
黄旭东
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xiamen Jinqiaoyu Information Technology Co ltd
Original Assignee
Xiamen Jinqiaoyu Information Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xiamen Jinqiaoyu Information Technology Co ltd filed Critical Xiamen Jinqiaoyu Information Technology Co ltd
Priority to CN202111427127.8A priority Critical patent/CN114205817A/en
Publication of CN114205817A publication Critical patent/CN114205817A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
    • H04W12/08—Access security
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/253—Fusion techniques of extracted features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/048—Activation functions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W84/00—Network topologies
    • H04W84/02—Hierarchically pre-organised networks, e.g. paging networks, cellular networks, WLAN [Wireless Local Area Network] or WLL [Wireless Local Loop]
    • H04W84/10—Small scale networks; Flat hierarchical networks
    • H04W84/12—WLAN [Wireless Local Area Networks]

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Software Systems (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Probability & Statistics with Applications (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The application relates to the field of network connection security, and particularly discloses a wireless local area network access method, a system and an electronic device. The method adopts a convolutional neural network model based on a deep learning technology to excavate high-dimensional association features among data so as to extract features of a topological structure constructed by the wireless APs and interactive features among the wireless APs, and also considers the communication data volume and the number of connected terminal devices, so that the finally calculated probability value of whether each AP is an illegal AP is more accurate. By the mode, the user can be prevented from accessing the illegal AP, and further personal information and money safety of the user are protected.

Description

Wireless local area network access method, system and electronic equipment
Technical Field
The present invention relates to the field of network connection security, and more particularly, to a wireless local area network access method, system and electronic device.
Background
At present, when a user uses a mobile terminal to access a wireless network through Wi-Fi, some risks in the aspect of safety are met, especially, more and more merchants provide free Wi-Fi access at present, and more risks are exposed while the use of the mobile terminal is facilitated. In all the Wireless network Access risks, the most harmful one should be that an illegal Access Point (AP) is used to provide Wireless network Access, and then a large amount of private information of a user is further obtained through a phishing website. Specifically, through an illegal AP, the same or similar Service Set Identifier (SSID) is set to provide free internet access service. Once a user accesses such an illegal AP, it is difficult to detect it. Such illegal APs can also implement Portal pages by way of redirection, but they are just a similar phishing page or website. The user continues to input the account information of the user to complete authentication, and the illegal AP can easily obtain the account information of the user such as the mobile phone number and the like. However, after the fake-decoration authentication is successful, any website visited by the user may be transferred to a designated phishing website, which includes internet banking, various electronic bank payment websites and the like, and as a result, a large amount of money of the user is lost.
Generally, it is difficult for most ordinary users to distinguish whether the users access an illegal AP. And when the user unconsciously accesses and uses the wireless network, personal information and money of the user can be leaked.
Therefore, in order to prevent a user from accessing an illegal AP, a wireless lan access method is desired.
Disclosure of Invention
The present application is proposed to solve the above-mentioned technical problems. The embodiment of the application provides a wireless local area network access method, a system and electronic equipment, wherein a convolutional neural network model based on a deep learning technology is adopted to dig out high-dimensional association features among data so as to extract features of a topological structure constructed by wireless APs and interactive features among the wireless APs, and the communication data volume and the number of connected terminal equipment are taken into consideration, so that the finally calculated probability value of whether each AP is an illegal AP is more accurate. By the mode, the user can be prevented from accessing the illegal AP, and further personal information and money safety of the user are protected.
According to an aspect of the present application, there is provided a wireless local area network access method, including:
constructing a communication data matrix for representing a plurality of wireless APs based on the topological structures of the wireless APs and the interaction characteristics among the wireless APs, wherein the characteristic value of each position in the communication data matrix is the communication data quantity between two wireless APs;
constructing a terminal degree matrix for representing the wireless AP and the terminal equipment based on the number of the terminal equipment connected with each wireless AP, wherein the terminal degree matrix and the communication data matrix have the same matrix structure;
inputting the communication data matrix and the termination degree matrix into a first convolutional neural network and a second convolutional neural network respectively to obtain a communication data characteristic diagram and a termination degree characteristic diagram, wherein the first convolutional neural network and the second convolutional neural network have the same network structure;
calculating a score function value based on node relationship scores between feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map to obtain a score feature map composed of the score function values, wherein the score function value is a result of dividing a natural exponent function value to which each node relationship score is raised by a weighted sum of each node relationship score, and the node relationship score is generated based on a comparison between a square of a difference between feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map and a feature value of a corresponding position in the communication data feature map;
generating a query vector consisting of the result of dividing the total communication data volume of each wireless AP by the number of terminal devices based on the total communication data volume between each wireless AP and other wireless APs and the number of terminal devices currently connected to each wireless AP;
converting the query vector into a query feature vector through an encoder;
mapping the query feature vector to a high-dimensional feature space in which the score feature map is located to obtain a classification feature vector;
calculating a Softmax-like function value of each position in the classification characteristic vector to serve as a probability value of whether each wireless AP is an illegal AP or not; and
and determining whether to access a certain wireless AP or not based on the Softmax-like function value.
According to another aspect of the present application, there is provided a wireless local area network access system, including:
the communication data matrix construction unit is used for constructing a communication data matrix for representing a plurality of wireless APs based on the topological structures of the wireless APs and the interaction characteristics among the wireless APs, wherein the characteristic value of each position in the communication data matrix is the communication data quantity between two wireless APs;
a terminal degree matrix constructing unit, configured to construct a terminal degree matrix used for representing the wireless AP and the terminal device based on the number of the terminal devices connected to each wireless AP, where the terminal degree matrix and the communication data matrix have the same matrix structure;
a convolutional neural network processing unit, configured to input the communication data matrix obtained by the communication data matrix constructing unit and the terminal degree matrix obtained by the terminal degree matrix constructing unit into a first convolutional neural network and a second convolutional neural network respectively to obtain a communication data feature map and a terminal degree feature map, where the first convolutional neural network and the second convolutional neural network have the same network structure;
a score feature map generation unit configured to calculate a score function value based on node relationship scores between the communication data feature map obtained by the convolutional neural network processing unit and feature values of each pair of corresponding positions in the terminal degree feature map obtained by the convolutional neural network processing unit to obtain a score feature map composed of the score function values, wherein the score function value is a result of dividing a natural exponent function value to which each node relationship score is raised by a weighted sum of each node relationship score generated based on a comparison between a square of a difference between feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map and feature values of corresponding positions in the communication data feature map;
a query vector generation unit, configured to generate a query vector composed of a result of dividing the total communication data amount of each wireless AP by the number of terminal devices, based on the total communication data amount between each wireless AP and other wireless APs and the number of terminal devices currently connected to each wireless AP;
the encoding unit is used for converting the query vector obtained by the query vector generation unit into a query feature vector through an encoder;
a mapping unit, configured to map the query feature vector obtained by the encoding unit into a high-dimensional feature space in which the score feature map obtained by the score feature map generating unit is located to obtain a classification feature vector;
a probability value calculating unit, configured to calculate a Softmax-like function value of each position in the classification feature vector obtained by the mapping unit, as a probability value of whether each wireless AP is an illegal AP; and
a result determining unit, configured to determine whether to access a certain wireless AP based on the Softmax-like function value obtained by the probability value calculating unit.
According to yet another aspect of the present application, there is provided an electronic device including: a processor; and a memory having stored therein computer program instructions which, when executed by the processor, cause the processor to perform the wireless local area network access method as described above.
According to yet another aspect of the present application, there is provided a computer readable medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform the wireless local area network access method as described above.
Compared with the prior art, the wireless local area network access method, the system and the electronic equipment adopt a convolutional neural network model based on a deep learning technology to dig out high-dimensional association features among data so as to extract features of a topological structure constructed by the wireless APs and interactive features among the wireless APs, and the method also considers the communication data quantity and the number of connected terminal equipment, so that the finally calculated probability value of whether each AP is an illegal AP is more accurate. By the mode, the user can be prevented from accessing the illegal AP, and further personal information and money safety of the user are protected.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing in more detail embodiments of the present application with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of the embodiments of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings, like reference numbers generally represent like parts or steps.
Fig. 1 is a diagram of an application scenario of a wlan access method according to an embodiment of the present application;
fig. 2 is a flowchart of a wlan access method according to an embodiment of the present application;
fig. 3 is a system architecture diagram illustrating a wlan access method according to an embodiment of the present application;
fig. 4 is a flowchart of calculating a score function value based on a node relation score between feature values of each pair of corresponding locations in the communication data feature map and the terminal degree feature map to obtain a score feature map composed of the score function values in the wireless local area network access method according to the embodiment of the present application;
fig. 5 is a block diagram of a wireless local area network access system according to an embodiment of the present application;
fig. 6 is a block diagram of a scoring profile generation unit in a wlan access system according to an embodiment of the present application;
fig. 7 is a block diagram of an electronic device according to an embodiment of the application.
Detailed Description
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be understood that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and that the present application is not limited by the example embodiments described herein.
Overview of a scene
As mentioned above, currently, when a user uses a mobile terminal to access a wireless network through Wi-Fi, some security risks are encountered, and especially, more and more merchants provide free Wi-Fi access, which also exposes more and more risks while facilitating the use of the user. In all the Wireless network Access risks, the most harmful one should be that an illegal Access Point (AP) is used to provide Wireless network Access, and then a large amount of private information of a user is further obtained through a phishing website. Specifically, through an illegal AP, the same or similar Service Set Identifier (SSID) is set to provide free internet access service. Once a user accesses such an illegal AP, it is difficult to detect it. Such illegal APs can also implement Portal pages by way of redirection, but they are just a similar phishing page or website. The user continues to input the account information of the user to complete authentication, and the illegal AP can easily obtain the account information of the user such as the mobile phone number and the like. However, after the fake-decoration authentication is successful, any website visited by the user may be transferred to a designated phishing website, which includes internet banking, various electronic bank payment websites and the like, and as a result, a large amount of money of the user is lost.
Generally, it is difficult for most ordinary users to distinguish whether the users access an illegal AP. And when the user unconsciously accesses and uses the wireless network, personal information and money of the user can be leaked. Therefore, in order to prevent a user from accessing an illegal AP, a wireless lan access method is desired.
Specifically, first, the topology structure constructed by the wireless APs and the interaction characteristics between the wireless APs are modeled. Specifically, the numbers of the plurality of wireless APs are arranged along the rows and columns of the matrix, so that the value of each position of the matrix is the communication data amount, such as the message data amount, between two corresponding wireless APs, and the values of the diagonal positions of the matrix are all 0, thereby obtaining the communication data matrix.
Also, in order to be structurally consistent with the communication data matrix, for the terminal devices to which the wireless APs are connected, a degree matrix of the wireless APs may be constructed, that is, the number of terminal devices connected to each wireless AP is taken as a diagonal line, thereby obtaining a terminal degree matrix.
Then, the communication data matrix and the terminal degree matrix are respectively input into the first convolutional neural network and the second convolutional neural network to mine high-dimensional correlation characteristics between the data, and therefore a communication data characteristic diagram and a terminal degree characteristic diagram are obtained.
Next, since the communication data feature map and the terminal degree feature map belong to different feature spaces, even if they are both converted to a probability space, if fusion is performed based on numerical values only, the accuracy of subsequent regression or classification is affected. Therefore, instead of directly using the probability values in the probability space, in the technical solution of the present application, a level theory in relation to the relational data is applied to perform feature fusion.
Specifically, assume that the probabilistic feature value of each position in the communication data feature map is fi,jAnd the probability characteristic value of each position in the terminal degree characteristic diagram is gi,jThen its node relation score sij＝a1fi,j-a2(fi,j-gi,j)2. Further, the score calculation function based on the node relation score has a value of pij＝exp(sij)/∑i,j(sij). Thus, p can be obtainedijAnd the formed score characteristic graph realizes the fusion of the communication data characteristic graph and the terminal degree characteristic graph.
Then, in the aspect of query vectors, because the communication data volume and the number of connected terminal devices are both considered, the total data volume of each wireless AP is obtained, and then divided by the number of the currently connected terminal devices to obtain the data volume of each terminal device of each wireless AP as a query vector, and the query vector is encoded into a query feature vector in a high-dimensional feature space through an encoder, and then multiplied by a score feature map to obtain a classification feature vector. Thus, by calculating the class Softmax classification function value of each position of the classification feature vector, the probability value of whether each AP is an illegal AP can be obtained.
Based on this, the present application provides a wireless local area network access method, which includes: constructing a communication data matrix for representing a plurality of wireless APs based on the topological structures of the wireless APs and the interaction characteristics among the wireless APs, wherein the characteristic value of each position in the communication data matrix is the communication data quantity between two wireless APs; constructing a terminal degree matrix for representing the wireless AP and the terminal equipment based on the number of the terminal equipment connected with each wireless AP, wherein the terminal degree matrix and the communication data matrix have the same matrix structure; inputting the communication data matrix and the termination degree matrix into a first convolutional neural network and a second convolutional neural network respectively to obtain a communication data characteristic diagram and a termination degree characteristic diagram, wherein the first convolutional neural network and the second convolutional neural network have the same network structure; calculating a score function value based on node relationship scores between feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map to obtain a score feature map composed of the score function values, wherein the score function value is a result of dividing a natural exponent function value to which each node relationship score is raised by a weighted sum of each node relationship score, and the node relationship score is generated based on a comparison between a square of a difference between feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map and a feature value of a corresponding position in the communication data feature map; generating a query vector consisting of the result of dividing the total communication data volume of each wireless AP by the number of terminal devices based on the total communication data volume between each wireless AP and other wireless APs and the number of terminal devices currently connected to each wireless AP; converting the query vector into a query feature vector through an encoder; mapping the query feature vector to a high-dimensional feature space in which the score feature map is located to obtain a classification feature vector; calculating a Softmax-like function value of each position in the classification characteristic vector to serve as a probability value of whether each wireless AP is an illegal AP or not; and determining whether to access a certain wireless AP or not based on the Softmax-like function value.
Fig. 1 illustrates an application scenario of a wireless local area network access method according to an embodiment of the present application. As shown in fig. 1, in the application scenario, first, interaction characteristics between a topology (e.g., M as illustrated in fig. 1) of a plurality of wireless APs (e.g., P as illustrated in fig. 1) and the plurality of wireless APs are obtained from a cloud (e.g., B as illustrated in fig. 1), where the interaction characteristics are communication data volumes between the plurality of wireless APs; and obtaining, from the cloud, a total communication data amount between each of the wireless APs and the other wireless APs and the number of terminal devices (e.g., T as illustrated in fig. 1) to which each of the wireless APs is currently connected.
Then, the obtained topology, the interaction feature, the total communication data amount, and the number of the terminal devices are input into a server (e.g., S as illustrated in fig. 1) deployed with a wireless local area network access algorithm, wherein the server can process the topology, the interaction feature, the total communication data amount, and the number of the terminal devices with the wireless local area network access algorithm to generate probability values representing whether each wireless AP is an illegal AP. And then, whether a certain wireless AP is accessed is determined based on the comparison between the probability value and a preset threshold value. Specifically, when the probability value is lower than a preset threshold value, the corresponding wireless AP is determined to be accessed. By the mode, the user can be prevented from accessing the illegal AP, and further personal information and money safety of the user are protected.
Having described the general principles of the present application, various non-limiting embodiments of the present application will now be described with reference to the accompanying drawings.
Exemplary method
Fig. 2 illustrates a flow chart of a wireless local area network access method. As shown in fig. 2, a wireless local area network access method according to an embodiment of the present application includes: s110, constructing a communication data matrix for representing a plurality of wireless APs based on topological structures of the wireless APs and interactive features among the wireless APs, wherein feature values of all positions in the communication data matrix are communication data quantity between two wireless APs; s120, constructing a terminal degree matrix used for representing the wireless AP and the terminal equipment based on the number of the terminal equipment connected with each wireless AP, wherein the terminal degree matrix and the communication data matrix have the same matrix structure; s130, inputting the communication data matrix and the termination degree matrix into a first convolutional neural network and a second convolutional neural network respectively to obtain a communication data characteristic diagram and a termination degree characteristic diagram, wherein the first convolutional neural network and the second convolutional neural network have the same network structure; s140, calculating a score function value based on node relationship scores between the feature values of each pair of corresponding locations in the communication data feature map and the terminal degree feature map to obtain a score feature map composed of the score function value, wherein the score function value is a result of dividing a natural exponent function value to which each node relationship score is raised by a weighted sum of each node relationship score, and the node relationship score is generated based on a comparison between a square of a difference between the feature values of each pair of corresponding locations in the communication data feature map and the feature values of the corresponding locations in the terminal degree feature map; s150, generating a query vector formed by dividing the total communication data volume of each wireless AP by the number of terminal devices based on the total communication data volume between each wireless AP and other wireless APs and the number of the terminal devices currently connected with each wireless AP; s160, converting the query vector into a query feature vector through an encoder; s170, mapping the query feature vector to a high-dimensional feature space where the score feature map is located to obtain a classification feature vector; s180, calculating a Softmax-like function value of each position in the classification characteristic vector as a probability value of whether each wireless AP is an illegal AP or not; and S190, determining whether to access a certain wireless AP or not based on the Softmax-like function value.
Fig. 3 illustrates an architecture diagram of a wlan access method according to an embodiment of the present application. As shown in fig. 3, in the network architecture of the wlan access method, first, a communication data matrix (e.g., M1 as illustrated in fig. 3) for representing a plurality of wireless APs is constructed based on a topology of the plurality of wireless APs (e.g., P1 as illustrated in fig. 3) and an interaction characteristic between the plurality of wireless APs (e.g., P2 as illustrated in fig. 3); then, constructing a terminal degree matrix (e.g., M2 as illustrated in fig. 3) for representing the wireless APs and the terminal devices based on the number of terminal devices (e.g., P3 as illustrated in fig. 3) to which each of the wireless APs is connected; then, inputting the communication data matrix and the termination degree matrix into a first convolutional neural network (e.g., CNN1 as illustrated in fig. 3) and a second convolutional neural network (e.g., CNN2 as illustrated in fig. 3), respectively, to obtain a communication data characteristic diagram (e.g., F1 as illustrated in fig. 3) and a termination degree characteristic diagram (e.g., F2 as illustrated in fig. 3); then, calculating a score function value based on the node relation score between the feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map to obtain a score feature map composed of the score function values (e.g., as illustrated in fig. 3 as F3); then, based on the total communication data amount between each of the wireless APs and the other wireless APs (e.g., Q1 as illustrated in fig. 3) and the number of terminal devices to which each of the wireless APs is currently connected (e.g., Q2 as illustrated in fig. 3), a query vector (e.g., V1 as illustrated in fig. 3) composed of the result of dividing the total communication data amount of each wireless AP by the number of terminal devices is generated; then, the query vector is converted into a query feature vector (e.g., VF1 as illustrated in fig. 3) by an encoder (e.g., as illustrated in fig. 3); then, mapping the query feature vector into a high-dimensional feature space in which the scored feature map resides to obtain a classification feature vector (e.g., VF2 as illustrated in fig. 3); next, calculating a Softmax-like function value of each position in the classification feature vector as a probability value (e.g., PV as illustrated in fig. 3) of whether the respective wireless AP is an illegal AP; and finally, determining whether to access a certain wireless AP or not based on the Softmax-like function value.
In steps S110 and S120, constructing a communication data matrix for representing a plurality of wireless APs based on the topology of the plurality of wireless APs and the interaction characteristics between the plurality of wireless APs, wherein the characteristic value of each position in the communication data matrix is the communication data amount between two wireless APs; and constructing a terminal degree matrix for representing the wireless AP and the terminal equipment based on the number of the terminal equipment connected with each wireless AP, wherein the terminal degree matrix and the communication data matrix have the same matrix structure. It should be understood that, in the technical solution of the present application, a relationship-based theory is adopted to predict whether the wireless AP is an illegal AP. Specifically, in a topology structure constructed by the wireless APs, each node in the topology structure is each wireless AP, and each node in the topology structure is an association between each wireless AP and the wireless AP. In addition, considering that if a certain AP is an illegal AP, there is a difference in the interaction mode between the AP and other APs, in the technical solution of the present application, the interaction characteristics between two wireless APs can be characterized based on the communication data, for example, the message data, between the wireless APs.
That is, in the technical solution of the present application, first, a topology structure constructed by a plurality of wireless APs and an interaction characteristic between the wireless APs are modeled to obtain a communication data matrix for representing the plurality of wireless APs. Specifically, the numbers of the plurality of wireless APs are arranged along the rows and columns of the communication data matrix, so that the characteristic value of each position of the communication data matrix is the communication data amount, such as the message data amount, between two corresponding wireless APs, and the values of the diagonal positions of the communication data matrix are both 0, thereby obtaining the communication data matrix. Then, in order to be consistent with the structure of the communication data matrix, in the technical solution of the present application, for the terminal devices connected to each of the wireless APs, a degree matrix of the wireless AP may be constructed, and in a specific example, the eigenvalue of each position of a diagonal line of the communication data matrix is replaced with the number of the terminal devices connected to each of the wireless APs, so as to obtain the terminal degree matrix.
Specifically, in the embodiment of the present application, a process for constructing a communication data matrix for representing a plurality of wireless APs based on a topology of the plurality of wireless APs and interaction characteristics between the plurality of wireless APs includes: firstly, numbering the plurality of wireless APs; then, arranging the codes of the plurality of wireless APs according to the rows and the columns of the matrix; then, filling the communication data quantity between the two corresponding wireless APs to each position of the non-diagonal line in the matrix; and finally, setting the eigenvalue of each position of the diagonal line of the matrix to 0 to obtain the communication data matrix.
In step S130, the communication data matrix and the termination degree matrix are respectively input into a first convolutional neural network and a second convolutional neural network to obtain a communication data feature map and a termination degree feature map, where the first convolutional neural network and the second convolutional neural network have the same network structure. That is, the communication data matrix is processed through a first convolutional neural network, and the terminal degree matrix is processed through a second convolutional neural network to mine high-dimensional correlation characteristics among the data, so that a communication data characteristic diagram and a terminal degree characteristic diagram are obtained. It is worth mentioning that, here, the first convolutional neural network and the second convolutional neural network have the same network structure, and the last layer of the first convolutional neural network and the last layer of the second convolutional neural network are both activated by a Sigmoid activation function, so that the feature value of each position in the communication data feature map and the termination degree feature map is in an interval of 0 to 1. It should be understood that, by performing probabilistic processing on the obtained feature values of each position in the communication data feature map and the terminal degree feature map, not only can dimensional influence between data features be eliminated, but also measurement and subsequent calculation are facilitated.
In step S140, a score function value based on node relationship scores between the feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map is calculated to obtain a score feature map composed of the score function value as a result of dividing a natural exponent function value to which each node relationship score is raised by a weighted sum of each node relationship score based on a square of a difference between the feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map and a feature value of the corresponding position in the communication data feature mapA comparison between the values is generated. It should be understood that since the communication data feature map and the terminal-degree feature map are attributed to different feature spaces, even if they are both converted to probability spaces, if fusion is performed based on numerical values only, the accuracy of subsequent regression or classification is affected. Therefore, instead of directly using the probability values in the probability space, in the technical solution of the present application, a level theory in relation to the relational data is applied to perform feature fusion. Specifically, assume that the probabilistic feature value of each location in the communication data feature map is fi,jAnd the probability characteristic value of each position in the terminal degree characteristic diagram is gi,jThen its node relation score sij＝a1fi,j-a2(fi,j-gi,j)2. Further, the score computation function based on the node relation score has a value of pij＝exp(sij)/∑i,j(sij). Thus, p can be obtainedijAnd the formed score characteristic graph realizes the fusion of the communication data characteristic graph and the terminal degree characteristic graph.
Specifically, in this embodiment of the present application, the process of calculating a score function value based on a node relationship score between feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map to obtain a score feature map composed of the score function values includes: firstly, mapping the communication data feature map and the terminal degree feature map into a probability space by a Softmax-like function to obtain a probabilistic communication data feature map and a probabilistic terminal degree feature map. It should be appreciated that the communication data feature map and the termination degree feature map are mapped into a probability space in a Softmax-like function for subsequent calculation thereof. Then, a score function value based on a node relation score between the feature values of each pair of corresponding positions in the probabilistic communication data feature map and the probabilistic terminating degree feature map is calculated to obtain the score feature map. In one specific example, first, the following formula is used to calculate the characteristic value of each pair of corresponding positions in the probabilistic communication data characteristic map and the probabilistic terminalness characteristic mapThe node relationship score between the nodes, wherein the formula is: sij＝a1fi,j-a2(fi,j-gi,j)2Wherein s isijRepresenting node relationship score, fi,jCharacteristic values, g, representing respective positions in the probabilistic communication data profilei,jRepresenting the characteristic value of each position in the probability terminal degree characteristic diagram; then, calculating a score function value based on the node relation score between the feature values of each pair of corresponding positions in the probabilistic communication data feature map and the probabilistic terminal degree feature map according to the following formula: p is a radical ofij＝exp(sij)/∑i,j(sij) Wherein p isijThe score function value is expressed.
Fig. 4 is a flowchart illustrating a node relationship score-based score function value between feature values of each pair of corresponding locations in the communication data feature map and the terminal degree feature map to obtain a score feature map composed of the score function values in the wireless local area network access method according to the embodiment of the present application. As shown in fig. 4, in the embodiment of the present application, calculating a score function value based on a node relation score between feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map to obtain a score feature map composed of the score function values includes: s210, mapping the communication data feature map and the terminal degree feature map into a probability space by a Softmax-like function to obtain a probabilistic communication data feature map and a probabilistic terminal degree feature map; and S220, calculating a score function value based on node relation scores between the feature values of each pair of corresponding positions in the probabilistic communication data feature map and the probabilistic terminal degree feature map to obtain the score feature map.
In steps S150 and S160, based on the total communication data amount between each wireless AP and other wireless APs and the number of terminal devices currently connected to each wireless AP, a query vector is generated, which is composed of the result of dividing the total communication data amount of each wireless AP by the number of terminal devices, and the query vector is converted into a query feature vector by an encoder. It should be understood that a legal wireless AP may connect to a relatively large number of terminal devices, and therefore, in terms of query vectors, the total communication data amount between each wireless AP and other wireless APs and the number of terminal devices currently connected to each wireless AP need to be considered. Therefore, in a specific example of the present application, first, the total communication data volume between each wireless AP and other wireless APs is obtained, and then divided by the number of terminal devices currently connected to the wireless AP, so as to obtain the data volume per terminal device of each wireless AP as a query vector. Then, the query vector is encoded into a query feature vector in a high-dimensional feature space through an encoder.
In step S170, the query feature vector is mapped into the high-dimensional feature space in which the score feature map is located to obtain a classification feature vector. That is, in one particular example, the query feature vector is multiplied by the scored feature map to map the query feature vector into a high-dimensional feature space in which the scored feature map resides, resulting in a classified feature vector. It is worth mentioning that, here, the classification feature vector represents a high-dimensional association feature that merges the query feature vector and each location feature value in the score feature map, that is, a correlation feature that merges the topology of the plurality of wireless APs, the interaction feature between the plurality of wireless APs, the total communication data amount between each of the wireless APs and other wireless APs, and the number of terminal devices currently connected to each of the wireless APs.
In steps S180 and S190, a Softmax-like function value of each position in the classification feature vector is calculated as a probability value of whether each wireless AP is an rogue AP, and whether a certain wireless AP is accessed is determined based on the Softmax-like function value. Specifically, in the embodiment of the present application, first, the Softmax-like function value of each position in the classification feature vector is calculated according to the following formula: q ═ exp (pi)/Σiexp (pi). Here, the Softmax-like function value of each position in the classification feature vector may be used as whether each wireless AP is an illegal AP or notA probability value. And then, determining whether to access a certain wireless AP or not based on the probability value. In one specific example, in response to the Softmax-like function value being below a preset threshold, determining to access the corresponding wireless AP; otherwise, determining not to access the corresponding wireless AP.
In summary, the wlan access method according to the embodiment of the present application is clarified, which includes mining high-dimensional association features between data by using a convolutional neural network model based on a deep learning technique to perform feature extraction on a topology structure constructed by wireless APs and interaction features between the wireless APs, and considering the amount of communication data and the number of connected terminal devices, so that a probability value of whether each of the APs is an illegal AP or not is calculated finally more accurately. By the mode, the user can be prevented from accessing the illegal AP, and further personal information and money safety of the user are protected.
Exemplary System
Fig. 5 illustrates a block diagram of a wireless local area network access system in accordance with an embodiment of the present application. As shown in fig. 5, a wlan access system 500 according to an embodiment of the present application includes: a communication data matrix constructing unit 510, configured to construct a communication data matrix for representing a plurality of wireless APs based on a topology structure of the plurality of wireless APs and interaction characteristics among the plurality of wireless APs, where a characteristic value of each position in the communication data matrix is a communication data amount between two wireless APs; a terminal degree matrix constructing unit 520, configured to construct a terminal degree matrix used for representing the wireless AP and the terminal device based on the number of terminal devices connected to each wireless AP, where the terminal degree matrix and the communication data matrix have the same matrix structure; a convolutional neural network processing unit 530, configured to input the communication data matrix obtained by the communication data matrix constructing unit 510 and the terminal degree matrix obtained by the terminal degree matrix constructing unit 520 into a first convolutional neural network and a second convolutional neural network, respectively, to obtain a communication data feature map and a terminal degree feature map, where the first convolutional neural network and the second convolutional neural network have the same network structure; a score feature map generation unit 540 configured to calculate a score function value based on node relationship scores between the communication data feature map obtained by the convolutional neural network processing unit 530 and feature values of each pair of corresponding positions in the terminal degree feature map obtained by the convolutional neural network processing unit 530 to obtain a score feature map composed of the score function values, wherein the score function value is a result of dividing a natural exponent function value to which each node relationship score is raised by a weighted sum of each node relationship score generated based on a comparison between a square of a difference between feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map and feature values of corresponding positions in the communication data feature map; a query vector generating unit 550, configured to generate a query vector composed of a result of dividing the total communication data amount of each wireless AP by the number of terminal devices, based on the total communication data amount between each wireless AP and other wireless APs and the number of terminal devices currently connected to each wireless AP; an encoding unit 560, configured to convert the query vector obtained by the query vector generation unit 550 into a query feature vector through an encoder; a mapping unit 570, configured to map the query feature vector obtained by the encoding unit 560 into a high-dimensional feature space in which the scored feature map obtained by the scored feature map generating unit 540 is located to obtain a classification feature vector; a probability value calculating unit 580 configured to calculate a class Softmax function value of each position in the classification feature vector obtained by the mapping unit 570 as a probability value of whether each wireless AP is an illegal AP; and a result determining unit 590, configured to determine whether to access a certain wireless AP based on the Softmax-like function value obtained by the probability value calculating unit 580.
In an example, in the above wireless local area network access system 500, the communication data matrix constructing unit 510 is further configured to: numbering the plurality of wireless APs; arranging the codes of the plurality of wireless APs according to the rows and the columns of the matrix; filling communication data quantity between two corresponding wireless APs to each position of non-diagonal lines in the matrix; and setting the eigenvalue of each position of the diagonal line of the matrix to 0 to obtain the communication data matrix.
In an example, in the above wireless local area network access system 500, the terminal degree matrix constructing unit 520 is further configured to: and replacing the characteristic values of each position of the diagonal line of the communication data matrix with the number of terminal devices connected with each wireless AP to obtain the terminal degree matrix.
In one example, in the above wireless local area network access system 500, the last layer of the first convolutional neural network and the second convolutional neural network is activated by a Sigmoid activation function so that the eigenvalue of each position in the communication data characteristic diagram and the terminal degree characteristic diagram is within an interval of 0 to 1.
In an example, in the above wireless local area network access system 500, as shown in fig. 6, the score feature map generating unit 540 includes: a probabilistic subunit 541, configured to map the communication data feature map and the termination degree feature map into a probability space by a Softmax-like function, so as to obtain a probabilistic communication data feature map and a probabilistic termination degree feature map; and a score function value calculation operator unit 542, configured to calculate a score function value based on a node relationship score between the probabilistic communication data feature map obtained by the probabilistic subunit 541 and a feature value of each pair of corresponding positions in the probabilistic terminating-degree feature map obtained by the probabilistic subunit 541, so as to obtain the score feature map.
In an example, in the above wireless lan access system 500, the score function value operator unit 542 includes: a relation score calculating subunit, configured to calculate a node relation score between feature values of each pair of corresponding positions in the probabilistic communication data feature map and the probabilistic terminating-degree feature map according to the following formula: sij＝a1fi,j-a2(fi,j-gi,j)2Wherein s isijRepresenting node relationship score, fi,jRepresenting a profile of said probabilistic communication dataCharacteristic value of each position in gi,jRepresenting the characteristic value of each position in the probability terminal degree characteristic diagram; and a score function value generating subunit, configured to calculate a score function value based on a node relationship score between feature values of each pair of corresponding locations in the probabilistic communication data feature map and the probabilistic terminal degree feature map according to the following formula: p is a radical ofij＝exp(sij)/∑i,j(sij) Wherein p isijThe score function value is expressed.
In an example, in the above wireless local area network access system 500, the probability value calculating unit 580 is further configured to: calculating a Softmax-like function value for each position in the classification feature vector with the following formula: q ═ exp (pi)/Σi exp(pi)。
In an example, in the above wireless local area network access system 500, the result determining unit 590 is further configured to: and in response to the Softmax-like function value being lower than a preset threshold, determining to access the corresponding wireless AP.
Here, it will be understood by those skilled in the art that the specific functions and operations of the respective units and modules in the above-described wireless local area network access system 500 have been described in detail in the above description of the wireless local area network access method with reference to fig. 1 to 4, and thus, a repetitive description thereof will be omitted.
As described above, the wireless lan access system 500 according to the embodiment of the present application may be implemented in various terminal devices, such as a server of a wireless lan access algorithm. In one example, the wlan access system 500 according to the embodiment of the present application may be integrated into the terminal device as a software module and/or a hardware module. For example, the wlan access system 500 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the wlan access system 500 can also be one of many hardware modules of the terminal device.
Alternatively, in another example, the wireless lan access system 500 and the terminal device may be separate devices, and the wireless lan access system 500 may be connected to the terminal device through a wired and/or wireless network and transmit the interaction information according to the agreed data format.
Exemplary electronic device
Next, an electronic apparatus according to an embodiment of the present application is described with reference to fig. 7. As shown in fig. 7, the electronic device 10 includes one or more processors 11 and a memory 12. The processor 11 may be a Central Processing Unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.
Memory 12 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, Random Access Memory (RAM), cache memory (cache), and/or the like. The non-volatile memory may include, for example, Read Only Memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium and executed by processor 11 to implement the functions of the wireless local area network access methods of the various embodiments of the present application described above and/or other desired functions. Various content such as a scored feature map, a classified feature vector, and the like may also be stored in the computer-readable storage medium.
In one example, the electronic device 10 may further include: an input system 13 and an output system 14, which are interconnected by a bus system and/or other form of connection mechanism (not shown).
The input system 13 may comprise, for example, a keyboard, a mouse, etc.
The output system 14 may output various information including probability values and the like to the outside. The output system 14 may include, for example, a display, speakers, a printer, and a communication network and its connected remote output devices, among others.
Of course, for simplicity, only some of the components of the electronic device 10 relevant to the present application are shown in fig. 7, and components such as buses, input/output interfaces, and the like are omitted. In addition, the electronic device 10 may include any other suitable components depending on the particular application.
Exemplary computer program product and computer-readable storage Medium
In addition to the above-described methods and apparatus, embodiments of the present application may also be a computer program product comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the functions of the wireless local area network access method according to the various embodiments of the present application described in the "exemplary methods" section of this specification above.
The computer program product may be written with program code for performing the operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the present application may also be a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, cause the processor to perform the steps in the wireless local area network access method described in the "exemplary methods" section above of this specification.
The computer-readable storage medium may take any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The foregoing describes the general principles of the present application in conjunction with specific embodiments, however, it is noted that the advantages, effects, etc. mentioned in the present application are merely examples and are not limiting, and they should not be considered essential to the various embodiments of the present application. Furthermore, the foregoing disclosure of specific details is for the purpose of illustration and description and is not intended to be limiting, since the foregoing disclosure is not intended to be exhaustive or to limit the disclosure to the precise details disclosed.
The block diagrams of devices, apparatuses, systems referred to in this application are only given as illustrative examples and are not intended to require or imply that the connections, arrangements, configurations, etc. must be made in the manner shown in the block diagrams. These devices, apparatuses, devices, systems may be connected, arranged, configured in any manner, as will be appreciated by those skilled in the art. Words such as "including," "comprising," "having," and the like are open-ended words that mean "including, but not limited to," and are used interchangeably therewith. The words "or" and "as used herein mean, and are used interchangeably with, the word" and/or, "unless the context clearly dictates otherwise. The word "such as" is used herein to mean, and is used interchangeably with, the phrase "such as but not limited to".
It should also be noted that in the devices, apparatuses, and methods of the present application, the components or steps may be decomposed and/or recombined. These decompositions and/or recombinations are to be considered as equivalents of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, the description is not intended to limit embodiments of the application to the form disclosed herein. While a number of example aspects and embodiments have been discussed above, those of skill in the art will recognize certain variations, modifications, alterations, additions and sub-combinations thereof.

Claims (10)

1. A wireless local area network access method, comprising:
constructing a communication data matrix for representing a plurality of wireless APs based on the topological structures of the wireless APs and the interaction characteristics among the wireless APs, wherein the characteristic value of each position in the communication data matrix is the communication data quantity between two wireless APs;
constructing a terminal degree matrix for representing the wireless AP and the terminal equipment based on the number of the terminal equipment connected with each wireless AP, wherein the terminal degree matrix and the communication data matrix have the same matrix structure;
inputting the communication data matrix and the termination degree matrix into a first convolutional neural network and a second convolutional neural network respectively to obtain a communication data characteristic diagram and a termination degree characteristic diagram, wherein the first convolutional neural network and the second convolutional neural network have the same network structure;
calculating a score function value based on node relationship scores between feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map to obtain a score feature map composed of the score function values, wherein the score function value is a result of dividing a natural exponent function value to which each node relationship score is raised by a weighted sum of each node relationship score, and the node relationship score is generated based on a comparison between a square of a difference between feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map and a feature value of a corresponding position in the communication data feature map;
generating a query vector consisting of the result of dividing the total communication data volume of each wireless AP by the number of terminal devices based on the total communication data volume between each wireless AP and other wireless APs and the number of terminal devices currently connected to each wireless AP;
converting the query vector into a query feature vector through an encoder;
mapping the query feature vector to a high-dimensional feature space in which the score feature map is located to obtain a classification feature vector;
calculating a Softmax-like function value of each position in the classification characteristic vector to serve as a probability value of whether each wireless AP is an illegal AP or not; and
and determining whether to access a certain wireless AP or not based on the Softmax-like function value.
2. The wlan access method according to claim 1, wherein constructing a communication data matrix for representing a plurality of wireless APs based on topology of the plurality of wireless APs and interaction characteristics among the plurality of wireless APs comprises:
numbering the plurality of wireless APs;
arranging the codes of the plurality of wireless APs according to the rows and the columns of the matrix;
filling communication data quantity between two corresponding wireless APs to each position of non-diagonal lines in the matrix; and
setting the eigenvalues of the respective positions of the diagonal of the matrix to 0 to obtain the communication data matrix.
3. The wlan access method according to claim 2, wherein constructing a terminal degree matrix for representing the APs and the terminal devices based on the number of terminal devices to which each of the APs is connected comprises:
and replacing the characteristic values of each position of the diagonal line of the communication data matrix with the number of terminal devices connected with each wireless AP to obtain the terminal degree matrix.
4. The wireless local area network access method according to claim 1, wherein the last layer of the first convolutional neural network and the second convolutional neural network is activated by a Sigmoid activation function so that the eigenvalue of each position in the communication data profile and the terminal degree profile is within an interval of 0 to 1.
5. The wireless local area network access method of claim 1, wherein calculating a score function value based on a node relationship score between feature values of each pair of corresponding locations in the communication data feature map and the terminal degree feature map to obtain a score feature map consisting of the score function values comprises:
mapping the communication data feature map and the terminal degree feature map into a probability space by a Softmax-like function to obtain a probabilistic communication data feature map and a probabilistic terminal degree feature map; and
and calculating a score function value based on node relation scores between the characteristic values of each pair of corresponding positions in the probabilistic communication data characteristic map and the probabilistic terminating degree characteristic map to obtain the score characteristic map.
6. The wireless local area network accessing method of claim 5, wherein calculating a score function value based on a node relationship score between feature values of each pair of corresponding locations in the probabilistic communication data feature map and the probabilistic terminating terminal degree feature map to obtain the score feature map further comprises:
calculating a node relation score between feature values of each pair of corresponding positions in the probabilistic communication data feature map and the probabilistic terminating degree feature map according to the following formula: sij＝a1fi,j-a2(fi,j-gi,j)2Wherein s isijRepresenting node relationship score, fi,jCharacteristic values, g, representing respective positions in the probabilistic communication data profilei,jRepresenting the characteristic value of each position in the probability terminal degree characteristic diagram; and
calculating a score function value based on a node relation score between the feature values of each pair of corresponding positions in the probabilistic communication data feature map and the probabilistic terminating degree feature map according to the following formula: p is a radical ofij＝exp(sij)/∑i,j(sij) Wherein p isijThe score function value is expressed.
7. The wireless local area network access method of claim 6, wherein calculating the Softmax-like function value for each location in the categorical feature vector as a probability value of whether the respective wireless AP is an rogue AP comprises:
calculating a Softmax-like function value for each position in the classification feature vector with the following formula: q ═ exp (pi)/Σi exp(pi)。
8. The wireless local area network access method of claim 7, wherein determining whether to access the wireless AP based on the Softmax-like function value comprises:
and in response to the Softmax-like function value being lower than a preset threshold, determining to access the corresponding wireless AP.
9. A wireless local area network access system, comprising:
the communication data matrix construction unit is used for constructing a communication data matrix for representing a plurality of wireless APs based on the topological structures of the wireless APs and the interaction characteristics among the wireless APs, wherein the characteristic value of each position in the communication data matrix is the communication data quantity between two wireless APs;
a terminal degree matrix constructing unit, configured to construct a terminal degree matrix used for representing the wireless AP and the terminal device based on the number of the terminal devices connected to each wireless AP, where the terminal degree matrix and the communication data matrix have the same matrix structure;
a convolutional neural network processing unit, configured to input the communication data matrix obtained by the communication data matrix constructing unit and the terminal degree matrix obtained by the terminal degree matrix constructing unit into a first convolutional neural network and a second convolutional neural network respectively to obtain a communication data feature map and a terminal degree feature map, where the first convolutional neural network and the second convolutional neural network have the same network structure;
a score feature map generation unit configured to calculate a score function value based on node relationship scores between the communication data feature map obtained by the convolutional neural network processing unit and feature values of each pair of corresponding positions in the terminal degree feature map obtained by the convolutional neural network processing unit to obtain a score feature map composed of the score function values, wherein the score function value is a result of dividing a natural exponent function value to which each node relationship score is raised by a weighted sum of each node relationship score generated based on a comparison between a square of a difference between feature values of each pair of corresponding positions in the communication data feature map and the terminal degree feature map and feature values of corresponding positions in the communication data feature map;
a query vector generation unit, configured to generate a query vector composed of a result of dividing the total communication data amount of each wireless AP by the number of terminal devices, based on the total communication data amount between each wireless AP and other wireless APs and the number of terminal devices currently connected to each wireless AP;
the encoding unit is used for converting the query vector obtained by the query vector generation unit into a query feature vector through an encoder;
a mapping unit, configured to map the query feature vector obtained by the encoding unit into a high-dimensional feature space in which the score feature map obtained by the score feature map generating unit is located to obtain a classification feature vector;
a probability value calculating unit, configured to calculate a Softmax-like function value of each position in the classification feature vector obtained by the mapping unit, as a probability value of whether each wireless AP is an illegal AP; and
a result determining unit, configured to determine whether to access a certain wireless AP based on the Softmax-like function value obtained by the probability value calculating unit.
10. An electronic device, comprising:
a processor; and
memory having stored therein computer program instructions which, when executed by the processor, cause the processor to perform the wireless local area network access method of any of claims 1-8.
CN202111427127.8A 2021-11-28 2021-11-28 Wireless local area network access method, system and electronic equipment Pending CN114205817A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111427127.8A CN114205817A (en) 2021-11-28 2021-11-28 Wireless local area network access method, system and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111427127.8A CN114205817A (en) 2021-11-28 2021-11-28 Wireless local area network access method, system and electronic equipment

Publications (1)

Publication Number Publication Date
CN114205817A true CN114205817A (en) 2022-03-18

Family

ID=80649299

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111427127.8A Pending CN114205817A (en) 2021-11-28 2021-11-28 Wireless local area network access method, system and electronic equipment

Country Status (1)

Country Link
CN (1) CN114205817A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116761237A (en) * 2023-05-30 2023-09-15 浙江知多多网络科技有限公司 Energy saving method and system based on wireless AP

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116761237A (en) * 2023-05-30 2023-09-15 浙江知多多网络科技有限公司 Energy saving method and system based on wireless AP

Similar Documents

Publication Publication Date Title
CN112084383B (en) Knowledge graph-based information recommendation method, device, equipment and storage medium
RU2638710C1 (en) Methods of detecting malicious elements of web pages
JP5031206B2 (en) Fit exponential model
CN111414987A (en) Training method and training device for neural network and electronic equipment
WO2021212968A1 (en) Unstructured data processing method, apparatus, and device, and medium
CN110148053B (en) User credit line evaluation method and device, electronic equipment and readable medium
US20220130496A1 (en) Method of training prediction model for determining molecular binding force
JP6532523B2 (en) Management of user identification registration using handwriting
WO2021068563A1 (en) Sample date processing method, device and computer equipment, and storage medium
CN111612635A (en) User financial risk analysis method and device and electronic equipment
CN114065693B (en) Method and system for optimizing layout of super-large-scale integrated circuit structure and electronic equipment
CN111198967A (en) User grouping method and device based on relational graph and electronic equipment
CN110855648A (en) Early warning control method and device for network attack
RU2652451C2 (en) Methods for anomalous elements detection on web pages
CN114565513A (en) Method and device for generating confrontation image, electronic equipment and storage medium
CN115878319A (en) Load balancing method and system and electronic equipment
WO2021184547A1 (en) Conversation robot intention corpus generation method and apparatus, medium, and electronic device
Murray Evolving location analytics for service coverage modeling
CN114205817A (en) Wireless local area network access method, system and electronic equipment
CN113837302B (en) City safety management monitoring method and system based on Internet of things and electronic equipment
CN116304341A (en) Fraud discrimination method and system based on user network big data
CN116167833B (en) Internet financial risk control system and method based on federal learning
CN112765330A (en) Text data processing method and device, electronic equipment and storage medium
WO2020252925A1 (en) Method and apparatus for searching user feature group for optimized user feature, electronic device, and computer nonvolatile readable storage medium
CN115409617B (en) Foreign currency exchange method and system based on block chain

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication