Radio frequency fingerprint recognition system based on wireless communication equipment

Abstract

The invention relates to the field of radio frequency identification, and discloses a radio frequency fingerprint identification system based on a wireless communication device, which is used to solve the problem that when a device identity authentication fails, it is impossible to identify that the device is not authenticated due to external factors. The method comprises an information input module, an information identification module and a problem feedback module, which judge the environment in which the device is located, collect radio frequency fingerprint feature information of the device, and enter the radio frequency fingerprint feature information of the device into a database, collect radio frequency fingerprint feature information of unknown devices, and match it with authorized device information in the database. For devices that fail the verification, the reason why the device fails the radio frequency fingerprint identification is judged according to the interference condition of the device, which effectively prevents the device identity authentication failure caused by external influencing factors of the device, and improves the accuracy of device identity identification.

Description

Technical Field

[0001] The present invention relates to the field of radio frequency identification, and more specifically to a radio frequency fingerprint identification system based on wireless communication equipment. Background Art

[0002] RF fingerprinting is a technology used to identify and recognize devices by analyzing the characteristics of the RF signals emitted by the devices to uniquely identify different devices. Each device has unique characteristics in terms of RF signals, which can be used as the "fingerprint" of the device, similar to fingerprint recognition in biometrics.

[0003] The existing radio frequency fingerprint recognition system for wireless communication devices performs device identity authentication by identifying the radio frequency fingerprint of the device to ensure that only authorized devices can access specific systems or networks. However, this authentication method cannot determine the reason why the device fails to pass the authentication, cannot identify device identity authentication failures caused by external factors of the device, and cannot take corresponding measures based on the authentication failures caused by different reasons.

[0004] In view of the above problems, the present invention proposes a solution. Summary of the invention

[0005] In order to overcome the above defects of the prior art, the present invention provides a radio frequency fingerprint recognition system based on a wireless communication device to solve the problems existing in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The radio frequency fingerprint recognition system based on wireless communication equipment includes an information entry module, an information recognition module and a problem feedback module. The modules are connected by signals. The data processing steps between the modules are as follows:

[0008] The information entry module determines the environment in which the device is located according to the calculated environmental suitability index, collects the radio frequency fingerprint feature information of the device, and enters the radio frequency fingerprint feature information of the device into the database;

[0009] The information recognition module collects the radio frequency fingerprint feature information of unknown devices and matches it with the authorized device information in the database;

[0010] The problem feedback module determines the reason why the device fails the RF fingerprint recognition based on the interference level of the device.

[0011] Preferably, the environmental suitability index calculation steps are:

[0012] Collect spectrum data through radio frequency sensors around the device;

[0013] Extract frequency information from the collected signal data by Fourier transform;

[0014] The acquired frequency information is clustered, the number of clusters of the frequency information is determined by the silhouette coefficient method, and the K-means clustering method is used to group similar signal features into the same class;

[0015] The spectrum crowding coefficient is calculated based on the distance between cluster centers, the number of clusters, and the number of clusters in the cluster;

[0016] The current humidity data around the device for recording characteristic information is detected by a humidity detection instrument;

[0017] The current temperature data of the device for inputting characteristic information is detected by a temperature detection instrument, and the temperature influence coefficient is calculated based on the current temperature data of the device;

[0018] The environmental suitability index is calculated by weighted summation based on the spectrum crowding coefficient, current humidity data and temperature influence coefficient.

[0019] Preferably, the distance between the cluster centers is obtained by obtaining the frequency points in each cluster and the cluster center of each cluster, and calculating the average value of the distance from each frequency point in the cluster to the cluster center through Euclidean distance.

[0020] Preferably, the step of determining the reason why the device fails to perform radio frequency fingerprint identification according to the interference degree of the device is to determine the interference degree of the device according to the calculated device interference index, and determine, according to the interference degree of the device, whether the reason why the radio frequency fingerprint identification fails is because the device is interfered with or because the device is an unauthorized device.

[0021] Preferably, the step of calculating the device interference index is:

[0022] Collect the usage information of the equipment in the database used by the information entry module, and calculate the equipment aging coefficient through the usage information of the equipment;

[0023] The equipment damage coefficient is obtained by setting sensors on the equipment and collecting data transmitted by the sensors and information extracted from the database used by the information entry module;

[0024] Use a magnetic field detection device to detect the magnetic field strength data around the device within the detection time period, and count the magnetic field strength and the detection time point, perform a differential operation on the detected magnetic field strength data, and calculate the peak difference in the differential data as the magnetic field variation coefficient;

[0025] The equipment interference index is obtained by weighted summing the equipment aging coefficient, equipment damage coefficient and magnetic field variation coefficient.

[0026] Preferably, the equipment aging coefficient calculation step is:

[0027] Extract the time when the device information was entered from the database according to the device ID, and calculate the total usage time of the device according to the time when the device information was entered and the current time;

[0028] Extract the equipment maintenance times from the database according to the equipment ID, and calculate the equipment maintenance frequency by the equipment maintenance times and the total equipment usage time;

[0029] The equipment aging coefficient is calculated based on the total equipment usage time and the equipment maintenance frequency.

[0030] Preferably, the equipment damage coefficient calculation step is:

[0031] The accelerometer on the RFID fingerprint recognition device detects vibration or falling of the device, and counts the number of times the device has been moved during the total use time.

[0032] The temperature sensor on the RFID device detects the temperature of the device and records the number of abnormal temperature events during the total use time.

[0033] Extract the device log information from the database based on the device ID and count the number of error reports in the log;

[0034] The equipment damage coefficient is calculated through the number of equipment movements, the number of times the suitable temperature is exceeded, and the number of error reports.

[0035] Technical effects and advantages of the present invention:

[0036] Determine the environment in which the device is located, collect the device's RF fingerprint feature information, and enter the device's RF fingerprint feature information into the database. Collect the RF fingerprint feature information of unknown devices and match it with the authorized device information in the database. For devices that fail verification, determine the reason why the device fails RF fingerprint recognition based on the interference situation of the device. This effectively prevents the failure of device identity authentication due to external factors affecting the device and improves the accuracy of device identity recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is the overall flow chart of the present invention;

[0038] Figure 2 A flowchart of steps for obtaining the affected conditions of the device of the present invention. DETAILED DESCRIPTION

[0039] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. In addition, the forms of the various structures recorded in the following embodiments are only examples. The radio frequency fingerprint recognition system based on wireless communication devices involved in the present invention is not limited to the various structures recorded in the following embodiments. All other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0040] The present invention provides a radio frequency fingerprint recognition system based on a wireless communication device, including an information input module, an information recognition module, and a problem feedback module. The modules are connected by signals, and the steps of data processing between the modules are as follows:

[0041] An information entry module is used to determine the environment in which the device is located, collect the radio frequency fingerprint feature information of the device, and enter the radio frequency fingerprint feature information of the device into the database;

[0042] When performing RF fingerprint feature extraction, environmental factors will affect the accuracy of the extracted features. Therefore, when performing feature extraction, it is necessary to consider whether the current environment is suitable for performing RF fingerprint feature extraction. This embodiment determines whether to perform feature extraction through the environmental suitability index.

[0043] In this embodiment, it should be specifically explained that Figure 1 As shown, the steps for obtaining the environmental suitability index are:

[0044] Setting a detection time, collecting spectrum data around the device through a radio frequency sensor within the detection time period, the detection time can be one minute or two minutes, the radio frequency sensor is a device used to detect and measure radio frequency signals, radio frequency signals refer to radio frequency signals within the radio frequency range, usually including frequency bands from hundreds of kilohertz to several gigahertz, the radio frequency sensor can capture and analyze these radio frequency signals, and provide information about the radio spectrum;

[0045] Extract frequency information from the collected spectrum data through Fourier transform, which is a mathematical tool that converts a signal from a time domain representation to a frequency domain representation, and can be used to analyze the frequency components of a signal and help understand the characteristics of the signal in the frequency domain;

[0046] The acquired frequency information is clustered, the number of clusters of the frequency information is determined by the silhouette coefficient method, and the similar signal features are grouped into the same class using the K-means clustering method. The number of clusters can reflect the number of different signal sources in the spectrum, and the cluster center of each cluster is counted;

[0047] According to the distance of cluster centers, the number of clusters, and the number of clusters, the spectrum crowding coefficient is calculated, and its calculation formula is: Where YJ represents the spectrum crowding coefficient, K is the number of clusters, and N i Indicates that there are N in each cluster i i frequency points, D i It is expressed as the average distance from all frequency points in the cluster to the cluster center. A higher spectrum crowding coefficient may indicate that there are more signal sources in the spectrum. When the spectrum is very crowded, it means that there are many wireless devices running on the same frequency band, which may cause signal interference and reduce the accuracy of feature extraction. A lower crowding coefficient may indicate a relatively idle spectrum, which has less impact on the accuracy of feature extraction.

[0048] The calculation of the average distance from all frequency points in the cluster to the cluster center takes into account both the compactness within the cluster and the number of frequency points in the cluster. Assume a cluster i with three frequency points in the cluster: x i1 , x i2 , x i3 , change and the cluster center is C i , calculate the distance from the three frequency points in the cluster to the cluster center through the Euclidean distance , calculate the average distance from the three frequency points to the cluster center ;

[0049] The current humidity data HD around the device for recording characteristic information is detected by a humidity detection instrument;

[0050] The current temperature data of the device with recorded characteristic information is detected by the temperature detection instrument, and the temperature influence coefficient is calculated based on the current temperature data of the device. The calculation formula is: , where TE is the temperature influence coefficient, T is the current temperature data of the device, and T 标准 Indicates the optimal operating temperature of the equipment. The optimal operating data of the equipment is provided by the equipment manufacturer in the product specification;

[0051] The environmental suitability index is calculated based on the spectrum crowding coefficient, current humidity data, and temperature influence coefficient. A high spectrum crowding coefficient may indicate that there are multiple signal sources on the spectrum or that the signal distribution is relatively dispersed, which may increase the complexity of the device when recording features in the spectrum. High humidity may cause signal propagation attenuation, affecting the device's perception and recording of RF signals. Temperature affects the performance of electronic components and the overall working state of the device. Too high or too low a temperature may cause the device to degrade or be damaged. The calculation formula is: , where SE represents the environmental suitability index, YJ represents the spectrum crowding coefficient, HD represents the current humidity data, TE represents the temperature influence coefficient, a1, a2, and a3 represent the weight coefficients of the spectrum crowding coefficient YJ, the current humidity data HD, and the temperature influence coefficient TE. This embodiment does not perform specific calculations on their specific values.

[0052] The environmental suitability index is compared with the preset threshold. If the environmental suitability index is greater than the preset threshold, the radio frequency fingerprint feature of the device is extracted. If the environmental suitability index is less than the preset threshold, an alarm is issued to the relevant information entry personnel, indicating that the current environment is not suitable for radio frequency fingerprint feature entry. The environmental suitability index is verified again after the device environment is processed accordingly.

[0053] For devices whose environmental suitability index is greater than a preset threshold, the wavelet transform method is used to extract the radio frequency fingerprint features. The wavelet transform is a signal processing technology used to decompose signals into components of different scales. It provides a method for analyzing signals in the time domain and frequency domain, and can capture the instantaneous characteristics of the signal. The wavelet basis function is a set of functions, which are usually obtained by scaling and translating a mother wavelet function. The core idea of ​​the wavelet transform is to use different scale and translation parameters to analyze the signal. By changing the scale and translation, the signal can be observed at different frequency and time scales.

[0054] In this embodiment, it should be specifically explained that the steps of extracting the radio frequency fingerprint features using wavelet transform are as follows:

[0055] The radio frequency signal returned by the receiving device during recording is subjected to denoising and normalization processing to ensure the quality and consistency of the signal. The denoising and normalization processing of the radio frequency signal is performed by applying a wavelet denoising method, selecting a bandpass filter for filtering, and using a maximum-minimum normalization processing method to preprocess the signal;

[0056] Select appropriate wavelet basis functions and scales to perform wavelet transform on the preprocessed signal. For example, the RF fingerprint signal of device A obtained by preprocessing is:

[0057] , the length of the RF fingerprint signal of device A is N, k is the time coordinate of the RF fingerprint signal, k ranges from 0, 1, 2, ... N-1, and Haar wavelet is selected as the basis function for wavelet transform. The RF fingerprint signal of device A is transformed into , where W A (a, b) represents the wavelet coefficients, a is the scale parameter, b is the translation parameter, is the Haar wavelet basis function;

[0058] The Haar wavelet basis function is , where t represents the parameters of scale a and translation b, usually expressed as , where k represents the time coordinate of the RF fingerprint signal, and t determines the translation position of the Haar wavelet basis function on the signal. By adjusting a and b, the signal changes can be observed on different time and frequency scales;

[0059] For example, if the length of the RF fingerprint signal of device A is 8, the Haar wavelet basis function is used for transformation. Assuming that the scale parameter is 2 and the translation parameter is 0, the calculation formula is: ,in is the Haar wavelet basis function;

[0060] By adjusting the scale and translation parameters, wavelet coefficients at different scales and frequencies can be obtained. The obtained wavelet coefficients are used to represent the characteristics of the RF fingerprint. The characteristics representing the RF fingerprint are stored in the database, and the storage time and the ID information of the input device are recorded, so as to facilitate the subsequent search and extraction of the device database information through the ID information, and perform system detection on the input device at regular intervals, and record the detection results in the log information. In the RF fingerprint recognition system, log information refers to the records of various events, operations and status information generated and recorded by the system, such as errors, exceptions and fault information occurring in the system.

[0061] Information identification module, used to collect radio frequency fingerprint feature information of unknown devices and match it with authorized device information in the database;

[0062] The wavelet coefficients in the database are used as training data sets, and the random forest algorithm is used to train the model of the training data sets. The trained random forest model is used to identify unknown devices and determine whether the unknown devices have passed the verification.

[0063] The random forest is an ensemble learning method based on decision trees. The random forest is an ensemble learning method that constructs multiple decision trees to perform classification or regression tasks and improves the overall classification accuracy by using multiple decision trees.

[0064] In this embodiment, it should be specifically explained that the steps of identifying an unknown device are:

[0065] Mark the wavelet coefficients of known devices with corresponding device labels to construct a training data set;

[0066] Use the random forest algorithm to train the model of the training data set, perform Bootstrap sampling on the training data set, and generate multiple random training sets. Each training set is extracted from the original data set with replacement;

[0067] For each training set and selected wavelet coefficient feature, a decision tree is constructed. The decision tree can be established using the CART decision tree algorithm in the random forest.

[0068] Collect the radio frequency signal of unknown devices and extract their feature vectors;

[0069] Use the trained random forest model to identify unknown devices, and use the decision tree set in the training data to determine which known device the unknown device belongs to.

[0070] If the random forest model recognizes the corresponding wavelet coefficient features in the database, it is determined that the unknown device is authorized and passes the verification. If the random forest model does not recognize the corresponding wavelet coefficient features in the database, it is determined that the unknown device fails the verification.

[0071] The problem feedback module is used to determine the reason why the device fails the RF fingerprint recognition according to the interference degree of the device.

[0072] The reason why the device's RF fingerprint fails is not only because the device is not authorized, but may also be due to the aging of the device, damage to the device, and changes in the magnetic field where the device is located.

[0073] As the device ages, the performance of the RF sensor may decline, including accuracy, sensitivity and response speed, which may lead to inaccuracy in biometric collection and affect the quality of RF fingerprint. Aging may also cause problems in the signal processing unit inside the device, such as decreased processing speed, increased noise, etc., reducing the accuracy of recognition. It may also cause instability in the power supply, resulting in unreliable operation of the device during the RF fingerprint recognition process.

[0074] Damage may cause problems with the signal processing unit inside the device, affecting the processing and analysis of the RF signal, which may cause errors, noise or other problems, thereby affecting the recognition results of the RF fingerprint. A damaged device may have unstable power supply, causing the device to work unreliably during the RF fingerprint recognition process.

[0075] The instability and changes in the magnetic field may cause the quality of the RF signal to deteriorate, which will affect the device's accurate collection of biometric features, especially when the magnetic field strength fluctuates greatly. Changes in the magnetic field may cause the accuracy of the RF fingerprint recognition algorithm to decrease. The model trained by the device under different magnetic field conditions may not be able to effectively cope with the feature changes caused by magnetic field changes, resulting in reduced recognition performance. Changes in the magnetic field may require the device to be calibrated more frequently. If the device does not perform magnetic field calibration in a timely manner, the recognition performance will be affected.

[0076] In this embodiment, it should be specifically explained that Figure 2 As shown, the steps for obtaining the affected status of the device are:

[0077] The time of entering the verified device information is extracted from the database by verifying the device ID, and the total device usage time is calculated using the following formula: , the T 总Expressed as the total equipment usage time, T 现 Represents the current time, T 录 Indicates the time when the device information is entered;

[0078] Extract the maintenance times of the verified equipment from the database by the equipment ID and calculate the maintenance frequency of the equipment. The calculation formula is: , where FM represents the maintenance frequency of the equipment, NM represents the number of maintenance times of the equipment, and T 总 It is expressed as the total usage time of the equipment;

[0079] The equipment aging coefficient is calculated by the total equipment usage time and maintenance frequency. The calculation formula is: , where AF is the equipment aging factor, FM is the equipment maintenance frequency, the higher the equipment maintenance frequency, the higher the equipment aging degree, T 总 It is expressed as the total use time of the equipment. The longer the equipment is used, the higher the degree of equipment aging. k is a set constant used to adjust the relationship between the equipment use time and the maintenance frequency. The range of the equipment aging coefficient is between 0 and 1. 0 means that the equipment is completely new, and 1 means that the equipment is completely aged.

[0080] The accelerometer on the RFID fingerprint recognition device is used to detect whether the device has been vibrated or dropped, and the number of times the device has moved during the total use time is counted. The accelerometer is a sensor used to measure the acceleration of an object. It is usually based on micro-electromechanical system technology and uses a tiny mechanical structure to sense changes in acceleration. The accelerometer can measure the acceleration of an object on three axes, usually the X, Y and Z axes. In the RFID fingerprint recognition device, the accelerometer can be used to detect whether the device has been vibrated, dropped or other sudden movements;

[0081] The temperature sensor on the RFID fingerprint recognition device is used to detect whether the device is at an appropriate temperature and record the number of abnormal device temperatures during the total use time. The temperature sensor is a sensor used to measure ambient temperature and can sense and convert temperature changes into electrical signals. In the RFID fingerprint recognition device, the temperature sensor is used to monitor the ambient temperature around the device. The appropriate temperature is provided by the device manufacturer in the product specification.

[0082] Extract the log information of the verified device by device ID in the database, and count the total number of logs and the number of error reports in the log;

[0083] The equipment damage coefficient is calculated by the number of equipment movements, the number of times the temperature exceeds the appropriate temperature, and the number of error reports. The calculation formula is: , where DF represents the equipment damage coefficient, NY represents the number of times the equipment is moved during use. The fewer times the equipment is moved during the total use time of the equipment, the less damage the equipment will suffer. NA represents the number of abnormal equipment temperature. Abnormal equipment temperature is usually caused by equipment damage. The more abnormal equipment temperature occurs, the greater the equipment damage will be. NC represents the number of error reports. LG represents the total number of logs. The higher the proportion of error reports, the more abnormal equipment times there are, and the higher the equipment damage coefficient will be.

[0084] Use a magnetic field detection device to detect the magnetic field strength data around the device within a detection time period, and count the magnetic field strength and the detection time point to ensure that the data is a time series. The magnetic field detection device is a device used to measure and monitor the surrounding magnetic field strength. These devices usually include magnetic field sensors. These sensors can sense the strength of the magnetic field and convert it into electrical signals. They are widely used in different applications, including scientific research, industrial applications, environmental monitoring, and some security and technical fields. The detection time period can be one minute or two minutes;

[0085] Perform a differential operation on the magnetic field strength data. The differential operation refers to calculating the data difference between adjacent time points. The calculation formula is: For the first time point, since there is no data from the previous moment, it is ignored. For example, the following time series of magnetic field strength data is detected: , perform differential operation on the magnetic field intensity data to obtain ;

[0086] Calculate the peak value in the differential data, that is, the positive and negative values ​​with the largest absolute values, and record the peak value in the differential data as the magnetic field variation coefficient. The calculation formula is: , where PE is the magnetic field variation coefficient, F max It is represented by the positive value with the largest absolute value in the differential data, F min It is represented by the negative value with the largest absolute value in the differential data;

[0087] The equipment interference index is obtained by weighted summation of the equipment aging coefficient, equipment damage coefficient and magnetic field change coefficient. The calculation formula is: , where DT is the device interference index, AF is the device aging coefficient, the higher the device aging coefficient, the lower the performance and reliability of the device, which has a certain impact on the device's RF fingerprint recognition. Device aging leads to hardware component loss, failure, sensor performance degradation, etc., thereby affecting the device's function and performance, DF is the device damage coefficient, the higher the device damage coefficient, the greater the impact of the device's RF fingerprint recognition accuracy and reliability, device damage may cause hardware component failure or performance degradation, thereby affecting RF fingerprint recognition, PE is the magnetic field variation coefficient, the greater the change in the magnetic field where the device is located, the greater the impact on the device's RF fingerprint recognition, the change in the magnetic field may affect the propagation and collection of RF signals, and thus affect the accuracy and stability of the RF fingerprint recognition system, b1, b2, b3 are weight coefficients of the device aging coefficient AF, the device damage coefficient DF, and the magnetic field variation coefficient PE, and this embodiment does not perform specific calculations on specific values;

[0088] The device interference index is compared with the preset threshold. If the device interference index is greater than the preset threshold, it is determined that the reason for the failure of the device verification is that the device is interfered with. An alarm is issued to the relevant staff to remind them that the device is interfered with. The staff will perform equipment maintenance on the device. If the device interference index is less than the preset threshold, the device is determined to be an unauthorized device and an early warning is issued to remind that the device is an unauthorized device.

[0089] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0090] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A radio frequency fingerprint recognition system based on wireless communication equipment, characterized in that: It includes information entry module, information identification module and problem feedback module. The modules are connected by signals. The data processing steps between the modules are as follows: The information entry module determines the environment in which the device is located according to the calculated environmental suitability index, collects the radio frequency fingerprint feature information of the device, and enters the radio frequency fingerprint feature information of the device into the database; The information recognition module collects the radio frequency fingerprint feature information of unknown devices and matches it with the authorized device information in the database; Through the problem feedback module, the reasons why the device failed to pass the RF fingerprint recognition are determined according to the interference degree of the device; The step of determining the reason why the device fails to perform radio frequency fingerprint identification according to the interference degree of the device is to determine the interference degree of the device according to the interference index of the calculated device, and to determine whether the reason why the radio frequency fingerprint identification fails is because the device is interfered or because the device is an unauthorized device according to the interference degree of the device; The steps for calculating the interference index of the device are as follows: Collect the usage information of the equipment in the database used by the information entry module, and calculate the equipment aging coefficient through the usage information of the equipment; The equipment damage coefficient is obtained by setting sensors on the equipment and collecting data transmitted by the sensors and information extracted from the database used by the information entry module; Use a magnetic field detection device to detect the magnetic field strength data around the device within the detection time period, and count the magnetic field strength and the detection time point, perform a differential operation on the detected magnetic field strength data, and calculate the peak difference in the differential data as the magnetic field variation coefficient; The equipment interference index is obtained by weighted summing the equipment aging coefficient, equipment damage coefficient and magnetic field variation coefficient; The steps for calculating the equipment aging coefficient are as follows: Extract the time when the device information was entered from the database according to the device ID, and calculate the total usage time of the device according to the time when the device information was entered and the current time; Extract the equipment maintenance times from the database according to the equipment ID, and calculate the equipment maintenance frequency by the equipment maintenance times and the total equipment usage time; The equipment aging coefficient is calculated based on the total use time of the equipment and the maintenance frequency of the equipment; The equipment damage coefficient calculation steps are as follows: The accelerometer on the RFID fingerprint recognition device detects vibration or falling of the device, and counts the number of times the device has been moved during the total use time. The temperature sensor on the RFID device detects the temperature of the device and records the number of abnormal temperature events during the total use time. Extract the device log information from the database based on the device ID and count the number of error reports in the log; The equipment damage coefficient is calculated through the number of equipment movements, the number of times the suitable temperature is exceeded, and the number of error reports.

2. The radio frequency fingerprint recognition system based on wireless communication equipment according to claim 1, characterized in that: The steps for calculating the environmental suitability index are as follows: Collect spectrum data through radio frequency sensors around the device; Extract frequency information from the collected signal data by Fourier transform; The acquired frequency information is clustered, the number of clusters of the frequency information is determined by the silhouette coefficient method, and the K-means clustering method is used to group similar signal features into the same class; The spectrum crowding coefficient is calculated based on the distance between cluster centers, the number of clusters, and the number of clusters in the cluster; The current humidity data around the device for recording characteristic information is detected by a humidity detection instrument; The current temperature data of the device for inputting characteristic information is detected by a temperature detection instrument, and the temperature influence coefficient is calculated based on the current temperature data of the device; The environmental suitability index is calculated by weighted summation based on the spectrum crowding coefficient, current humidity data and temperature influence coefficient.

3. The radio frequency fingerprint recognition system based on wireless communication equipment according to claim 2, characterized in that: The distance between the cluster centers is obtained by obtaining the frequency points in each cluster and the cluster center of each cluster, and calculating the average value of the distance from each frequency point in the cluster to the cluster center through the Euclidean distance.

Images