A wireless camera identification method and system based on WiFi signal

Abstract

The present invention relates to the field of WiFi data processing technology, and in particular to a method for identifying a wireless camera based on a WiFi signal, which first obtains a WiFi round-scan signal, decodes the WiFi round-scan signal, obtains scan data, then extracts features from the scan data, obtains scan feature data, and then compares the scan feature data with a plurality of preset thresholds, obtains a target wireless camera based on the comparison result, and can also upload the scan feature data to a server for data analysis, and obtains the target wireless camera based on the data analysis result. Compared with the prior art, the present invention obtains the scan feature data of the WiFi round-scan signal by decoding and extracting features from the WiFi round-scan signal, so that the data transmission features reflected by the scan feature data can be used to identify hidden wireless cameras.

Description

Technical Field

[0001] The present invention relates to the technical field of WiFi data processing, and in particular to a method for identifying a wireless camera based on WiFi signals. Background Art

[0002] Wireless cameras are a modern technology product that allows users to remotely monitor and record video via a wireless network connection. Compared to traditional wired cameras, wireless cameras offer greater flexibility and convenience because they do not need to be connected to a power source or recording device via physical cables. Users can easily install wireless cameras anywhere they need to monitor and view images or receive notifications in real time through devices such as smartphones, tablets, or computers.

[0003] However, the convenience of wireless cameras is also accompanied by privacy and security issues. If wireless cameras are used improperly or installed in a hidden way, personal privacy may be seriously violated. Obviously, in some special scenarios, the importance of wireless camera identification is self-evident.

[0004] Therefore, people need a solution that can identify wireless cameras. Summary of the invention

[0005] Therefore, the present invention provides a method for identifying a wireless camera based on a WiFi signal, so as to solve the problem of how to identify a wireless camera in the prior art.

[0006] The present invention provides a method for identifying a wireless camera based on a WiFi signal, comprising:

[0007] Obtain WiFi round-scanning signals, and decode the WiFi round-scanning signals to obtain scanning data;

[0008] Extracting features from the scan data to obtain scan feature data;

[0009] Compare the scan feature data with a plurality of preset thresholds, and obtain the target wireless camera according to the comparison result;

[0010] The scan feature data is uploaded to the server for data analysis, and the target wireless camera is obtained based on the data analysis results.

[0011] The present invention also provides a preferred solution: extracting features from the scan data to obtain scan feature data, including:

[0012] Preprocess the scanned data to obtain multiple basic transmission information, each of which includes a MAC address;

[0013] Based on the MAC address and the statistical characteristics of basic transmission information, the scanning feature data corresponding to each MAC address is established;

[0014] Among them, the scanning characteristic data includes: traffic statistical characteristic data, transmission statistical characteristic data, data volume statistical characteristic data and signal characteristic statistical characteristic data.

[0015] The present invention also provides a preferred solution: the traffic statistical feature data includes traffic balance; based on the MAC address, according to the statistical characteristics of the basic transmission information, the scanning feature data corresponding to each MAC address is established, including:

[0016] Count the uplink and downlink traffic corresponding to the target MAC address, where the target MAC address is the MAC address to be extracted.

[0017] According to the upstream and downstream traffic corresponding to the target MAC address, the traffic balance is calculated by the following formula: ;

[0018] in, is the flow balance, is the upstream traffic, For downstream traffic.

[0019] The present invention also provides a preferred solution: the traffic statistical characteristic data also includes the average traffic; the transmission statistical characteristic data includes the number of data uploads, the standard deviation of the upload time interval, the ratio of the standard deviation of the upload time interval to the mean of the upload time interval, and the maximum upload time interval; the data volume statistical characteristic data includes the mean of each data packet size, the standard deviation of each data packet size, and the ratio of the standard deviation of each data packet size to the mean of each data packet size; the signal characteristic statistical characteristic data includes the signal strength mean and the signal strength variance.

[0020] The present invention also provides a preferred solution: uploading the scan feature data to a server for data analysis, and obtaining the target wireless camera according to the data analysis result, including:

[0021] Upload the target mac address and the corresponding target scan feature data to the server, wherein the server stores multiple reference mac addresses and corresponding reference scan feature data, the target mac address is a mac address to be analyzed currently, the target scan feature data is the scan feature data corresponding to the target mac address, the reference mac address is the mac address uploaded to the server by other devices running the identification method of wireless cameras based on WiFi signals, and the reference scan feature data is the scan feature data corresponding to the reference mac address;

[0022] The target MAC address and the corresponding target scanning feature data, multiple reference MAC addresses and the corresponding reference scanning feature data are clustered, and the target wireless camera is obtained according to the clustering result.

[0023] The present invention also provides a preferred solution: clustering the target MAC address and the corresponding target scanning feature data, multiple reference MAC addresses and the corresponding reference scanning feature data, and obtaining the target wireless camera according to the clustering result, including:

[0024] Performing a first clustering on the target MAC address and multiple reference MAC addresses to obtain multiple first clustering clusters;

[0025] Performing a second clustering on the target scanning feature data and the plurality of reference scanning feature data to obtain a plurality of second clustering clusters;

[0026] According to the difference between the number of elements in the first cluster where the target MAC address is located and the number of elements in the second cluster where the target scan feature data is located, it is determined whether the device corresponding to the target MAC address is the target wireless camera.

[0027] The present invention also provides a preferred solution: performing a second clustering on the target scanning feature data and the plurality of reference scanning feature data to obtain a plurality of second clustering clusters, including:

[0028] Establishing a plurality of feature vectors based on the target scanning feature data and the plurality of reference scanning feature data;

[0029] According to the first cluster, the number of elements in the first cluster where the MAC address corresponding to each feature vector is located is counted as a distance metric correction coefficient for each feature vector;

[0030] A second clustering is performed on the plurality of feature vectors, and in the second clustering process, the metric distance between two feature vectors is corrected based on the distance metric correction coefficient to obtain a plurality of second clustering clusters.

[0031] The present invention also provides a preferred solution: correcting the metric distance between two feature vectors based on the distance metric correction coefficient, including:

[0032] The modified metric distance between two eigenvectors is obtained by the following formula: ;

[0033] in, is the modified metric distance between two eigenvectors, is the metric distance between two eigenvectors before correction, and are the distance metric correction coefficients of the two eigenvectors, Adjust the factor for the preset units.

[0034] The present invention also provides a preferred solution: judging whether the device corresponding to the target MAC address is a target wireless camera according to the difference between the number of elements in the first cluster where the target MAC address is located and the number of elements in the second cluster where the target scanning feature data is located, including:

[0035] Calculate the difference between the number of elements in the first cluster where the target MAC address is located and the number of elements in the second cluster where the target scan feature data is located to obtain a feature difference;

[0036] According to the feature difference, the probability that the device corresponding to the target MAC address is the target wireless camera is obtained, wherein the smaller the feature difference, the higher the probability that the device corresponding to the target MAC address is the target wireless camera;

[0037] According to the probability that the device corresponding to the target MAC address is the target wireless camera, determine whether the device corresponding to the target MAC address is the target wireless camera.

[0038] The present invention also provides a wireless camera recognition system based on WiFi signals, comprising:

[0039] The signal scanning module is used to obtain the WiFi round scanning signal and decode the WiFi round scanning signal to obtain the scanning data;

[0040] A feature extraction module is used to extract features from the scan data to obtain scan feature data;

[0041] A primary recognition module is used to compare the scan feature data with a plurality of preset thresholds and obtain the target wireless camera according to the comparison result;

[0042] The advanced recognition module is used to upload the scan feature data to the server for data analysis, and obtain the target wireless camera based on the data analysis results.

[0043] The beneficial effects of adopting the above embodiment are:

[0044] The present invention provides a method and system for identifying a wireless camera based on a WiFi signal. The method first obtains a WiFi round-scan signal, decodes the WiFi round-scan signal, obtains scan data, extracts features from the scan data, obtains scan feature data, compares the scan feature data with a plurality of preset thresholds, obtains a target wireless camera based on the comparison result, and can also upload the scan feature data to a server for data analysis, and obtains a target wireless camera based on the data analysis result. Compared with the prior art, the present invention obtains the scan feature data of the WiFi round-scan signal by decoding and extracting features from the WiFi round-scan signal, so that the data transmission features reflected by the scan feature data can be used to identify the hidden wireless camera. In addition, the present invention completes the rough identification of the wireless camera by comparing the scan feature data with a plurality of preset thresholds to meet daily needs, so that devices with relatively low computing power such as personal terminals can also run the method, thereby realizing the identification of the wireless camera. At the same time, the present invention further uses the server data analysis method to analyze the scan feature data in a more detailed manner to achieve accurate identification, so as to meet the identification needs in more stringent scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A method flow chart of an embodiment of a method for identifying a wireless camera based on WiFi signals provided by the present invention;

[0046] Figure 2 for Figure 1 Specific step diagram of step S104;

[0047] Figure 3 for Figure 2 Specific step diagram of step S202;

[0048] Figure 4 A system architecture diagram of an embodiment of a wireless camera recognition system based on WiFi signals provided by the present invention;

[0049] Figure 5 A system framework diagram of a specific embodiment provided by the present invention;

[0050] Figure 6 A business process diagram of a specific embodiment provided by the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] Combination Figure 1 As shown, a specific embodiment of the present invention discloses a method for identifying a wireless camera based on a WiFi signal, comprising:

[0053] S101, obtaining a WiFi round-scan signal, and decoding the WiFi round-scan signal to obtain scan data;

[0054] S102, extracting features from the scan data to obtain scan feature data;

[0055] S103, comparing the scan feature data with a plurality of preset thresholds, and obtaining a target wireless camera according to the comparison result;

[0056] S104, uploading the scan feature data to the server for data analysis, and obtaining the target wireless camera according to the data analysis result.

[0057] Compared with the prior art, the present invention obtains the scanning feature data of the WiFi round-scanning signal by decoding and extracting features from the WiFi round-scanning signal, so that the data transmission characteristics reflected by the scanning feature data can be used to identify hidden wireless cameras. In addition, the present invention performs a relatively rough identification of wireless cameras by comparing the scanning feature data with multiple preset thresholds to meet daily needs, so that devices with relatively low computing power such as personal terminals can also run this method to achieve the identification of wireless cameras. At the same time, the present invention further uses server data analysis to more carefully analyze the scanning feature data to achieve accurate identification to meet the identification needs in more stringent scenarios.

[0058] In addition, it should be noted that there is actually no sequential dependency between the above processes S103 and S104, and the serial numbers in this article are only for the convenience of reading. In practice, S103 and S104 can be sequential or parallel, depending on the specific situation. For example, in practice, step S103 can be executed first. If the returned result does not satisfy the user (for example, there are too many returned results), the result of step S103 can be used to further perform in-depth analysis of step S104, or the process of steps S101-S102 can be repeated, and the new result obtained can be used for step S104. For another example, in practice, step S103 and step S104 can be continued at the same time. Obviously, the execution speed of S103 is faster, and the result can be returned faster to allow the user to preliminarily obtain the identification situation on the scene. When the server completes the operation of step S104, the results returned by steps S103 and S104 are combined to make a more accurate judgment of the wireless camera.

[0059] Specifically, in a preferred embodiment, the above step S102, extracting features from the scan data to obtain scan feature data, specifically includes:

[0060] Preprocess the scanned data to obtain multiple basic transmission information, each of which includes a MAC address;

[0061] Based on the MAC address and the statistical characteristics of basic transmission information, the scanning feature data corresponding to each MAC address is established;

[0062] Among them, the scanning characteristic data includes: traffic statistical characteristic data, transmission statistical characteristic data, data volume statistical characteristic data and signal characteristic statistical characteristic data.

[0063] Specifically, in a preferred embodiment, the traffic statistical feature data includes traffic balance; and the above step: taking the mac address as a reference, according to the statistical characteristics of the basic transmission information, establishing the scanning feature data corresponding to each mac address, includes:

[0064] Count the uplink and downlink traffic corresponding to the target MAC address, where the target MAC address is the MAC address to be extracted.

[0065] According to the upstream and downstream traffic corresponding to the target MAC address, the traffic balance is calculated by the following formula: ;

[0066] in, is the flow balance, For upstream traffic, For downstream traffic.

[0067] In the above process, traffic balance is selected as a traffic statistical feature data. The balance of the device corresponding to the target MAC address is measured by subtracting the upstream traffic from the downstream traffic and then comparing it with the upstream traffic, or uncertainty. For a wireless camera, its upstream traffic and downstream traffic are obviously different. For the above formula, it measures traffic balance from two aspects at the same time to obtain a more accurate result. On the one hand, the larger the upstream traffic, the greater the possibility that the device corresponding to the MAC address is a wireless camera. On the other hand, the greater the difference between the upstream traffic and the downstream traffic, the greater the possibility that the device corresponding to the MAC address is a wireless camera.

[0068] Furthermore, in a preferred embodiment, the traffic statistical characteristic data also includes average traffic; the transmission statistical characteristic data includes the number of data uploads, the standard deviation of the upload time interval, the ratio of the standard deviation of the upload time interval to the mean of the upload time interval, and the maximum upload time interval; the data volume statistical characteristic data includes the mean of each data packet size, the standard deviation of each data packet size, and the ratio of the standard deviation of each data packet size to the mean of each data packet size; the signal characteristic statistical characteristic data includes the signal strength mean and the signal strength variance.

[0069] For the above-mentioned scan data, by comparing it with some preset thresholds, the similarity between a certain scan data and the data transmission characteristics that a wireless camera should have can be analyzed based on the comparison results, thereby preliminarily identifying the wireless camera. However, for some special occasions, such as when some devices are processing abnormal situations and their data transmission and reception are abnormal, or when there are many wireless devices in the environment and there are many suspected wireless camera devices, the above method is difficult to identify the wireless camera more accurately. Therefore, the present invention further provides a method for analyzing the scan feature data, that is, sending it to a server, using a more intelligent model to analyze it, and using hardware with higher computing power to process it, so as to complete the identification task more accurately.

[0070] Specifically, combined Figure 2 As shown, in a preferred embodiment, the above step S104, uploading the scan feature data to the server for data analysis, and obtaining the target wireless camera according to the data analysis results, specifically includes:

[0071] S201, upload the target mac address and the corresponding target scan feature data to the server, wherein the server stores multiple reference mac addresses and corresponding reference scan feature data, the target mac address is a mac address to be currently analyzed, the target scan feature data is the scan feature data corresponding to the target mac address, the reference mac address is the mac address uploaded to the server by other devices running the identification method of wireless cameras based on WiFi signals, and the reference scan feature data is the scan feature data corresponding to the reference mac address;

[0072] S202, clustering the target MAC address and the corresponding target scanning feature data, multiple reference MAC addresses and the corresponding reference scanning feature data, and obtaining the target wireless camera according to the clustering result.

[0073] The reference MAC address and the corresponding reference scanning feature data in the above process are the data uploaded by other similar devices collected by the server. This embodiment uses the powerful computing function of the server to analyze all the data sets, that is, to use the clustering algorithm to analyze the differences between the scanning feature data and identify the more abnormal feature scanning data.

[0074] Specifically, combined Figure 3 As shown, in a preferred embodiment, the above step S202, clustering the target mac address and the corresponding target scan feature data, multiple reference mac addresses and the corresponding reference scan feature data, and obtaining the target wireless camera according to the clustering result, specifically includes:

[0075] S301, performing a first clustering on the target MAC address and multiple reference MAC addresses to obtain multiple first clustering clusters;

[0076] S302, performing a second clustering on the target scan feature data and the plurality of reference scan feature data to obtain a plurality of second clustering clusters;

[0077] S303: Determine whether the device corresponding to the target MAC address is the target wireless camera based on the difference between the number of elements in the first cluster where the target MAC address is located and the number of elements in the second cluster where the target scan feature data is located.

[0078] Because users may rarely perform feedback in practice, such as feeding back to the system whether the identified wireless camera is correct, it is unknown whether the base mac address stored in the server and the corresponding base scan feature data are the corresponding data of the wireless camera. In order to eliminate this defect and improve the usability of the entire system, the present embodiment adopts a secondary clustering method for data analysis. Because the mac address is unique, especially the first 24 bits are called OUI (Organizationally Unique Identifier), which is a unique identifier assigned by IEEE to network hardware manufacturers. Then, if there is enough data, there must be an OUI corresponding to the same manufacturer in the base mac address, so by clustering the mac address for the first time, the possibility of whether a mac address is a wireless camera can be analyzed from a hardware perspective. In addition, the second clustering analysis of the features of the scan feature data itself is also used in the present embodiment. The combination of the above two aspects of analysis can greatly improve the recognition accuracy. It can be understood that the first clustering and the second clustering can both adopt clustering methods such as DBSCAN that do not require the pre-specified number of clusters. The specific process is a prior art, so this article will not make too much explanation.

[0079] Furthermore, in a preferred embodiment, the above step S302, performing a second clustering on the target scan feature data and the plurality of reference scan feature data to obtain a plurality of second clustering clusters, includes:

[0080] Establishing a plurality of feature vectors based on the target scanning feature data and the plurality of reference scanning feature data;

[0081] According to the first cluster, the number of elements in the first cluster where the MAC address corresponding to each feature vector is located is counted as a distance metric correction coefficient for each feature vector;

[0082] A second clustering is performed on the plurality of feature vectors, and in the second clustering process, the metric distance between two feature vectors is corrected based on the distance metric correction coefficient to obtain a plurality of second clustering clusters.

[0083] The advantage of the above process is that the distance measurement result of clustering is further corrected by using the result of the first clustering in the second clustering, and clustering is performed based on the corrected measurement distance to specifically optimize the clustering effect of the second clustering.

[0084] Specifically, in a preferred embodiment, the above step of correcting the metric distance between two feature vectors based on the distance metric correction coefficient includes:

[0085] The modified metric distance between two eigenvectors is obtained by the following formula: ;

[0086] in, is the modified metric distance between two eigenvectors, is the metric distance between two eigenvectors before correction, and are the distance metric correction coefficients of the two eigenvectors, Adjust the factor for the preset units.

[0087] In the above process, the metric distance between the two feature vectors before correction is the metric method of the conventional clustering algorithm in the prior art, such as the Euclidean distance of the two feature vectors. The significance of the above formula is that the clustering result of the first clustering is further corrected. For example, when the difference in the distance metric correction coefficients of the two feature vectors is smaller, the probability that the two are devices of the same specification (the OUI in the mac address is the same) is greater. At this time, more attention should be paid to the difference in the working conditions expressed by the scanning feature data of the two, and the clustering should be more detailed. Therefore, the corrected metric distance obtained at this time should also be larger, so that more clusters can be obtained for more accurate analysis. On the contrary, when the difference in the distance metric correction coefficients of the two feature vectors is larger, the probability that the two are devices of the same specification (the OUI in the mac address is the same) is smaller, and the probability that the meanings expressed by the scanning feature data of the two are different is also greater. Because the final judgment of whether it is a wireless sensor is also combined with the first clustering result, even if the two are divided into the same cluster at this time, it will not have a significant impact on the final result, and this behavior will also simplify the complexity of the data and improve the clustering speed.

[0088] Further, in a preferred embodiment, the above step S303, based on the difference between the number of elements in the first cluster where the target mac address is located and the number of elements in the second cluster where the target scan feature data is located, determines whether the device corresponding to the target mac address is the target wireless camera, specifically including:

[0089] Calculate the difference between the number of elements in the first cluster where the target MAC address is located and the number of elements in the second cluster where the target scan feature data is located to obtain a feature difference;

[0090] According to the feature difference, the probability that the device corresponding to the target MAC address is the target wireless camera is obtained, wherein the smaller the feature difference, the higher the probability that the device corresponding to the target MAC address is the target wireless camera;

[0091] According to the probability that the device corresponding to the target MAC address is the target wireless camera, determine whether the device corresponding to the target MAC address is the target wireless camera.

[0092] The significance of the above process is to comprehensively consider the two aspects of mac address and scan feature data to obtain accurate clustering results. For example, if the number of elements in the first cluster where the target mac address is located is small, and the number of elements in the second cluster where the target scan feature data is located is large, then it means that there are fewer target mac addresses in the base mac address, but there are a large number of similar target scan features in the base scan feature data. At this time, it can be considered that the target mac address is unlikely to be a mass-produced device, and the target scan feature data is also likely to be just a feature of some abnormal data that different devices will encounter. Similarly, if the number of elements in the first cluster where the target mac address is located is large, and the number of elements in the second cluster where the target scan feature data is located is small, then it means that there are more target mac addresses in the base mac address, but there are only a small number of similar target scan features in the base scan feature data. At this time, it can be considered that although the target mac address may be a mass-produced device, the target scan feature data is likely to be a data feature of a common working condition of this type of device. Only when the difference between the number of elements in the first cluster where the target MAC address is located and the number of elements in the second cluster where the target scan feature data is located is small, that is, the proportion of the target MAC address in the benchmark MAC address and the proportion of the target scan feature data in the benchmark scan feature data are relatively consistent, the possibility that the device corresponding to the target MAC address is a wireless sensor device is greater.

[0093] Combination Figure 4 As shown, the present invention also provides a wireless camera recognition system based on WiFi signals, comprising:

[0094] The signal scanning module 410 is used to obtain the WiFi round scanning signal and decode the WiFi round scanning signal to obtain scanning data;

[0095] The feature extraction module 420 is used to extract features from the scan data to obtain scan feature data;

[0096] The primary identification module 430 is used to compare the scan feature data with a plurality of preset thresholds and obtain the target wireless camera according to the comparison result;

[0097] The advanced recognition module 440 is used to upload the scan feature data to the server for data analysis, and obtain the target wireless camera according to the data analysis results.

[0098] The present invention also provides a more specific embodiment to more clearly illustrate the above steps S101-S104 and the identification system of the wireless camera based on the WiFi signal:

[0099] This embodiment is applied to a hardware system related to a smart phone. The architecture of the system is as follows: Figure 5 shown.

[0100] Overall business process:

[0101] 1. After the probe device is powered on and self-checked, it is connected to the mobile phone via Bluetooth to establish communication;

[0102] 2. After the mobile applet or app is opened, the probe device can be controlled to start detection (time is 120s);

[0103] 3. The probe scans the 2.4g WiFi channel, obtains relevant information (i.e. scanning data) and sends it to the mobile phone via Bluetooth;

[0104] 4. The mobile phone receives the relevant information and decodes it to obtain the basic transmission information;

[0105] 5. Perform feature extraction on basic transmission information and make judgments based on preset thresholds;

[0106] 6. The determined data is compared with the MAC address library and the data is labeled;

[0107] 7. The mobile app or app displays a list of suspected wireless camera devices;

[0108] 8. Users can select the information displayed in the mobile phone list to select a single device and perform location query function;

[0109] 9. Users can also choose to perform a deep scan to rescan the control probe device and upload the data to the server for further identification, and then send it back to the phone for display.

[0110] In the above process, before proceeding to step 4, preprocessing is required, that is, data that does not meet the following requirements is excluded:

[0111] 1. When the upstream traffic is greater than 0, the downstream traffic must be equal to 0;

[0112] 2. When the downlink traffic is greater than 0, the uplink traffic must be equal to 0;

[0113] 3. Uplink traffic and downlink traffic cannot coexist at the same time, that is, the traffic of both cannot be greater than 0 at the same time;

[0114] 4. The uplink MAC address is not equal to the downlink MAC address;

[0115] 5. The signal strength must be less than 0;

[0116] 6. The channel number is greater than or equal to 1 and less than or equal to 13.

[0117] The basic transmission information obtained after parsing is as follows:

[0118] Based on the above basic information, the traffic data is calculated as follows: based on the statistics of the uplink MAC within the detection period, the relevant information is counted. The following is the relevant scanning feature data:

[0119] a. (Uplink traffic - Downlink traffic) / Uplink traffic;

[0120] b. Number of data uploads / minute;

[0121] c. Standard deviation of upload time interval;

[0122] d. Standard deviation of upload time interval / mean of upload time interval;

[0123] e. Maximum upload time interval;

[0124] f. The mean size of each data packet;

[0125] g, standard deviation of each packet size;

[0126] h, standard deviation of each packet size / mean of each packet size;

[0127] i. Average flow rate / s;

[0128] j, mean signal strength;

[0129] k, signal strength variance.

[0130] Based on the above scan feature data, if the following preset threshold conditions are met at the same time, it will be judged as a camera:

[0131] 1. Average signal strength > -70db;

[0132] 2. Upstream traffic > x / min;

[0133] 3. (Uplink traffic - Downlink traffic) / Uplink traffic > y;

[0134] 4. Maximum upload time interval <s;

[0135] 5. The mean size of each data packet > k × the number of data uploads + b;

[0136] Among them, -70db, x, y, and s are all different preset thresholds, and k and b are preset linear transformation coefficients.

[0137] A more specific business process diagram of this embodiment is as follows Figure 6 shown.

[0138] The present invention provides a method and system for identifying a wireless camera based on a WiFi signal. The method first obtains a WiFi round-scan signal, decodes the WiFi round-scan signal, obtains scan data, extracts features from the scan data, obtains scan feature data, compares the scan feature data with a plurality of preset thresholds, obtains a target wireless camera based on the comparison result, and can also upload the scan feature data to a server for data analysis, and obtains a target wireless camera based on the data analysis result. Compared with the prior art, the present invention obtains the scan feature data of the WiFi round-scan signal by decoding and extracting features from the WiFi round-scan signal, so that the data transmission features reflected by the scan feature data can be used to identify the hidden wireless camera. In addition, the present invention completes the rough identification of the wireless camera by comparing the scan feature data with a plurality of preset thresholds to meet daily needs, so that devices with relatively low computing power such as personal terminals can also run the method, thereby realizing the identification of the wireless camera. At the same time, the present invention further uses the server data analysis method to analyze the scan feature data in a more detailed manner to achieve accurate identification, so as to meet the identification needs in more stringent scenarios.

[0139] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0140] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying a wireless camera based on WiFi signals, characterized in that: include: Obtain WiFi round-scanning signals, and decode the WiFi round-scanning signals to obtain scanning data; Extracting features from the scan data to obtain scan feature data; Compare the scan feature data with a plurality of preset thresholds, and obtain the target wireless camera according to the comparison result; The scan feature data is uploaded to the server for data analysis, and the target wireless camera is obtained according to the data analysis results; Among them, feature extraction is performed on the scan data to obtain scan feature data, including: Preprocess the scanned data to obtain multiple basic transmission information, each of which includes a MAC address; Based on the MAC address and the statistical characteristics of basic transmission information, the scanning feature data corresponding to each MAC address is established; The scanning characteristic data includes: flow statistical characteristic data, transmission statistical characteristic data, data volume statistical characteristic data and signal characteristic statistical characteristic data; Among them, the scanning feature data is uploaded to the server for data analysis, and the target wireless camera is obtained according to the data analysis results, including: Upload the target mac address and the corresponding target scan feature data to the server, wherein the server stores multiple reference mac addresses and corresponding reference scan feature data, the target mac address is a mac address to be analyzed currently, the target scan feature data is the scan feature data corresponding to the target mac address, the reference mac address is the mac address uploaded to the server by other devices running the identification method of wireless cameras based on WiFi signals, and the reference scan feature data is the scan feature data corresponding to the reference mac address; Clustering the target MAC address and the corresponding target scanning feature data, multiple reference MAC addresses and the corresponding reference scanning feature data, and obtaining the target wireless camera according to the clustering result; Among them, the target MAC address and the corresponding target scanning feature data, multiple reference MAC addresses and the corresponding reference scanning feature data are clustered, and the target wireless camera is obtained according to the clustering result, including: Performing a first clustering on the target MAC address and multiple reference MAC addresses to obtain multiple first clustering clusters; Performing a second clustering on the target scanning feature data and the plurality of reference scanning feature data to obtain a plurality of second clustering clusters; According to the difference between the number of elements in the first cluster where the target MAC address is located and the number of elements in the second cluster where the target scan feature data is located, it is determined whether the device corresponding to the target MAC address is the target wireless camera.

2. The method for identifying a wireless camera based on WiFi signals according to claim 1, characterized in that: Traffic statistics feature data include traffic balance; Based on the MAC address and the statistical characteristics of basic transmission information, the scanning feature data corresponding to each MAC address is established, including: Count the uplink and downlink traffic corresponding to the target MAC address, where the target MAC address is the MAC address to be extracted. According to the upstream and downstream traffic corresponding to the target MAC address, the traffic balance is calculated by the following formula: ; in, is the flow balance, is the upstream traffic, For downstream traffic.

3. The method for identifying a wireless camera based on WiFi signals according to claim 2, characterized in that: The traffic statistical characteristic data also includes the average traffic; the transmission statistical characteristic data includes the number of data uploads, the standard deviation of the upload time interval, the ratio of the standard deviation of the upload time interval to the mean of the upload time interval, and the maximum upload time interval; the data volume statistical characteristic data includes the mean of each data packet size, the standard deviation of each data packet size, and the ratio of the standard deviation of each data packet size to the mean of each data packet size; the signal characteristic statistical characteristic data includes the signal strength mean and the signal strength variance.

4. The method for identifying a wireless camera based on WiFi signals according to claim 1, characterized in that: The target scanning feature data and the plurality of reference scanning feature data are clustered for the second time to obtain a plurality of second clustering clusters, including: Establishing a plurality of feature vectors based on the target scanning feature data and the plurality of reference scanning feature data; According to the first cluster, the number of elements in the first cluster where the MAC address corresponding to each feature vector is located is counted as a distance metric correction coefficient for each feature vector; A second clustering is performed on the plurality of feature vectors, and in the second clustering process, the metric distance between two feature vectors is corrected based on the distance metric correction coefficient to obtain a plurality of second clustering clusters.

5. The method for identifying a wireless camera based on WiFi signals according to claim 4, characterized in that: Correcting the metric distance between two feature vectors based on the distance metric correction coefficient includes: The modified metric distance between two eigenvectors is obtained by the following formula: ; in, is the modified metric distance between two eigenvectors, is the metric distance between two eigenvectors before correction, and are the distance metric correction coefficients of the two eigenvectors, Adjust the factor for the preset units.

6. The method for identifying a wireless camera based on WiFi signals according to claim 5, characterized in that: According to the difference between the number of elements in the first cluster where the target MAC address is located and the number of elements in the second cluster where the target scanning feature data is located, determining whether the device corresponding to the target MAC address is the target wireless camera includes: Calculate the difference between the number of elements in the first cluster where the target MAC address is located and the number of elements in the second cluster where the target scan feature data is located to obtain a feature difference; According to the feature difference, the probability that the device corresponding to the target MAC address is the target wireless camera is obtained, wherein the smaller the feature difference, the higher the probability that the device corresponding to the target MAC address is the target wireless camera; According to the probability that the device corresponding to the target MAC address is the target wireless camera, determine whether the device corresponding to the target MAC address is the target wireless camera.

7. A wireless camera recognition system based on WiFi signals, characterized in that: include: The signal scanning module is used to obtain the WiFi round scanning signal and decode the WiFi round scanning signal to obtain the scanning data; A feature extraction module is used to extract features from the scan data to obtain scan feature data; A primary recognition module is used to compare the scan feature data with a plurality of preset thresholds and obtain the target wireless camera according to the comparison result; The advanced recognition module is used to upload the scan feature data to the server for data analysis, and obtain the target wireless camera based on the data analysis results; Among them, feature extraction is performed on the scan data to obtain scan feature data, including: Preprocess the scanned data to obtain multiple basic transmission information, each of which includes a MAC address; Based on the MAC address and the statistical characteristics of basic transmission information, the scanning feature data corresponding to each MAC address is established; The scanning characteristic data includes: flow statistical characteristic data, transmission statistical characteristic data, data volume statistical characteristic data and signal characteristic statistical characteristic data; Among them, the scanning feature data is uploaded to the server for data analysis, and the target wireless camera is obtained according to the data analysis results, including: Upload the target mac address and the corresponding target scan feature data to the server, wherein the server stores multiple reference mac addresses and corresponding reference scan feature data, the target mac address is a mac address to be analyzed currently, the target scan feature data is the scan feature data corresponding to the target mac address, the reference mac address is the mac address uploaded to the server by other devices running the identification method of wireless cameras based on WiFi signals, and the reference scan feature data is the scan feature data corresponding to the reference mac address; Clustering the target MAC address and the corresponding target scanning feature data, multiple reference MAC addresses and the corresponding reference scanning feature data, and obtaining the target wireless camera according to the clustering result; Among them, the target MAC address and the corresponding target scanning feature data, multiple reference MAC addresses and the corresponding reference scanning feature data are clustered, and the target wireless camera is obtained according to the clustering result, including: Performing a first clustering on the target MAC address and multiple reference MAC addresses to obtain multiple first clustering clusters; Performing a second clustering on the target scanning feature data and the plurality of reference scanning feature data to obtain a plurality of second clustering clusters; According to the difference between the number of elements in the first cluster where the target MAC address is located and the number of elements in the second cluster where the target scan feature data is located, it is determined whether the device corresponding to the target MAC address is the target wireless camera.

Images