阅读说明：本技术 一种无线设备识别方法、系统及数据处理终端 (Wireless device identification method and system and data processing terminal ) 是由 张志为 李旭飞 沈玉龙 何怡 祝幸辉 王建东 曾水光 刘洋洋 于 2021-07-03 设计创作，主要内容包括：本发明属于物理层设备识别技术领域,公开了一种无线设备识别方法、系统及数据处理终端,所述无线设备识别方法包括：构建基于DSSS帧前导分形维数的无线设备识别模型；采样信号并依据设备辐射缺陷进行信号特征提取；利用DSSS帧前导计算信号分形维数；利用DSSS帧前导分形维数进行设备识别。本发明使用DSSS帧前导分形维数作为无线设备识别中的一种可用新辐射特征,该特征有效提高了物理层设备识别准确度。实验结果表明,本发明提供的基于DSSS帧前导分形维数的无线设备识别方法能够保证与已知特征相较性能增益。在不同噪声环境下,本发明的识别方法仍能保证可观的性能增益,尤其在高噪声环境下识别准确度提升更为显著。(The invention belongs to the technical field of physical layer equipment identification, and discloses a wireless equipment identification method, a system and a data processing terminal, wherein the wireless equipment identification method comprises the following steps: constructing a wireless equipment identification model based on DSSS frame leading fractal dimension; sampling signals and extracting signal characteristics according to the radiation defects of the equipment; calculating a signal fractal dimension by utilizing a DSSS frame preamble; and identifying the equipment by using the DSSS frame preamble fractal dimension. The invention uses DSSS frame leading fractal dimension as an available new radiation characteristic in wireless equipment identification, and the characteristic effectively improves the identification accuracy of physical layer equipment. Experimental results show that the DSSS frame preamble fractal dimension-based wireless equipment identification method provided by the invention can ensure performance gain compared with known characteristics. Under different noise environments, the identification method can still ensure considerable performance gain, and particularly, the identification accuracy is improved more obviously under the high-noise environment.)

1. A wireless device identification method, the wireless device identification method comprising:

constructing a wireless equipment identification model based on DSSS frame leading fractal dimension;

sampling signals and extracting signal characteristics according to the radiation defects of the equipment;

calculating a signal fractal dimension by utilizing a DSSS frame preamble;

and identifying the equipment by using the DSSS frame preamble fractal dimension.

2. The wireless device identification method of claim 1, wherein the constructing the wireless device identification model based on the DSSS frame preamble fractal dimension comprises:

each frame signal is subjected to the influence of hardware defects of radio frequency links of a transmitter and a receiver, wherein the influence of the hardware defects comprises a digital-to-analog converter conversion error, a frequency error of a frequency converter and a power amplifier amplification deviation; any hardware defect in the radio frequency link can cause a wireless signal to present a specific radiation characteristic, and the radiation characteristic is estimated by using a physical layer radio frequency signal at a receiver end;

the radiation characteristic is referred to as the characteristic for short, because the frame preamble field is defined in advance in the wireless communication protocol, the characteristic estimation is carried out by using the preamble field of the wireless signal frame, and the receiver can use the preamble field as a reference signal of the characteristic estimation; the payload part refers to actual communication transmission data, and the content of the actual communication transmission data cannot be known by a receiving end in advance, so the part is not used for characteristic estimation for description;

the leader field is a fixed sequence consisting of 0-1 symbols and is defined as follows:

X＝[x₀，x₁，…，x_i，…x_N]；

wherein x is_1≤i≤NI/Q modulation symbols.

3. The wireless device identification method of claim 1, wherein sampling the signal and extracting signal features based on device radiation defects comprises sampling using a frame preamble field defined in the 802.11b specification PLCP and extracting signal features based on each frame, comprising:

(1) sampling and frame detection: sampling signals at a sampling frequency of 22MHz, and roughly judging the starting position and the ending position of each frame in the sampled signals by using a power matching algorithm;

(2) frame positioning: after frame detection is roughly carried out, a sliding window is adopted to match a received signal sample with a 128-bit lead code specified by a protocol, so that accurate frame positioning is realized;

(3) and (3) feature calculation: after accurate frame positioning, using frame preamble for feature calculation; and calculating signal characteristics based on the original sample, and calculating fractal dimension characteristics by using the despread frame preamble.

4. The wireless device identification method of claim 3, wherein in step (3), the signal characteristics include carrier frequency offset, I/Q imbalance, amplitude error, phase error, and synchronization error.

5. The wireless device identification method of claim 1, wherein the calculating the signal fractal dimension using the DSSS frame preamble comprises:

the actual received signal is represented as:

X′＝[x′₀，x′₁，...，x′_i，…x′_N]；

the fractal dimension is then determined by unifying the received symbols into the same reference symbol by phase rotationThe calculation method comprises the following steps:

wherein L is_τ，nAnd L_1，nRespectively as follows:

L_τ，n＝|x′_n·τ-x′_(n-1)·τ|；

L_1，n＝|x′_n-x′_n-1|；

where X' denotes a receiver-side despread frame preamble field, where each symbol differs from the symbol in X due to a physical defect in the radio frequency chain.

6. The wireless device identification method of claim 1, wherein utilizing DSSS frame preamble fractal dimension for device identification comprises:

(1) acquiring 1000 frames of signals for each network card, respectively calculating six equipment characteristics including fractal dimension, carrier frequency offset, amplitude error, I/Q imbalance, phase error and synchronous error of each frame sample, forming six-dimensional signal characteristic vectors, and normalizing the calculated characteristic vectors to serve as original training data;

(2) training input data by adopting a kNN algorithm and an SVM algorithm, and carrying out equipment identification on a signal frame of unknown equipment according to a trained model;

(3) and extracting a signal feature vector from the signal frame of the unknown equipment according to a feature extraction method, inputting the signal feature vector into the trained kNN and SVM models, and obtaining a model output result which is the identified equipment identity.

7. A wireless device identification system for implementing the wireless device identification method according to any one of claims 1 to 6, the wireless device identification system comprising:

the model building module is used for building a wireless equipment identification model based on DSSS frame preamble fractal dimension;

the characteristic extraction module is used for sampling signals and extracting the characteristics of the signals according to the radiation defects of the equipment;

a fractal dimension calculating module for calculating a signal fractal dimension by using the DSSS frame preamble;

and the equipment identification module is used for identifying the equipment by utilizing the DSSS frame preamble fractal dimension.

8. A computer device, characterized in that the computer device comprises a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to carry out the steps of: constructing a wireless equipment identification model based on DSSS frame leading fractal dimension; sampling signals and extracting signal characteristics according to the radiation defects of the equipment; calculating a signal fractal dimension by utilizing a DSSS frame preamble; and identifying the equipment by using the DSSS frame preamble fractal dimension.

9. A computer-readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of: constructing a wireless equipment identification model based on DSSS frame leading fractal dimension; sampling signals and extracting signal characteristics according to the radiation defects of the equipment; calculating a signal fractal dimension by utilizing a DSSS frame preamble; and identifying the equipment by using the DSSS frame preamble fractal dimension.

10. An information data processing terminal characterized by being configured to implement the wireless device identification system according to claim 7.

Technical Field

The invention belongs to the technical field of physical layer equipment identification, and particularly relates to a wireless equipment identification method, a wireless equipment identification system and a data processing terminal.

Background

Currently, the number of wireless devices is expected to increase in the foreseeable future with advances in technologies such as the internet of things, 5G, and LPWAN. With the rapid increase of the number of wireless access devices, wireless device identification is gradually faced with new technical challenges, such as efficient access identification. Furthermore, device identification based techniques also play an increasingly important role in many emerging applications, such as device localization, tracking, and the like. The traditional identification method is to store identity information in devices, such as media access control address (MAC), International Mobile Equipment Identity (IMEI) and radio frequency identification (rfid), however, the increase of the application scene of wireless devices brings new challenges, the distribution and management of identity information are increasingly complex, and they cannot cope well with physical attacks, and an attacker can easily access or steal identity-related information by damaging the hardware of the wireless device.

The latest research is the use of radio signal radiation signatures for device identification. Here, the radiation characteristics refer to non-reproducible and non-forgeable statistical characteristics that can be measured from the wireless signal. These characteristics are mainly caused by physical imperfections of the transceiver in the wireless device, which are randomly generated during the manufacturing process; they are therefore difficult to control, predict and clone, and can serve as endogenous, non-reproducible and non-counterfeitable identities of device identification. Wireless device identification based on signal radiation characteristics is a powerful addition to traditional information-based storage methods because 1) it is endogenous, i.e., does not require human additional identity; 2) the method has universal applicability and can be applied to all wireless devices; 3) the method has robustness to hardware attack, and increases the cost of physical attack; 4) it exhibits high compatibility, i.e. it is compatible with existing information storage methods; 5) it improves communication efficiency (no explicit communication overhead); 6) it provides privacy protection, i.e. does not require the collection of private data such as MAC, IMEI or biometric features.

However, wireless device identification schemes based on signal radiation characteristics are still limited in their limited identification accuracy and are currently not in practical widespread use. This can be attributed to two main factors: first, the recognizable scale of a single feature, i.e., the maximum number of devices that can be discerned without a significant error rate, is limited; secondly, a multi-feature joint identification scheme is an effective solution, but the deep research on a hardware defect mechanism is lacked at present, so that the development of more signal feature extraction is hindered. Therefore, it is important to research new radiation features that can effectively improve the device identification accuracy to promote the development of the device identification method based on the radiation features.

Direct Sequence Spread spectrum (dsss) (direct Sequence Spread spectrum) is a wireless Spread spectrum communication mode with high security and high interference immunity. DSSS spreads the spectrum of a signal at the transmitting end by using a high-rate spreading sequence, and despreads the signal at the receiving end with the same spreading code sequence, restoring the spread signal to the original signal. The direct sequence spread spectrum technology is widely applied to military communication and confidential industries, is even popularized to some civil high-end products such as signal base stations, wireless televisions, cellular mobile phones, baby monitors and the like, and is a reliable and safe industrial application scheme.

Through the above analysis, the problems and defects of the prior art are as follows:

(1) the traditional identification method is to store identity information in devices, such as media access control addresses, international mobile equipment identities and radio frequency identification, however, the increase of the application scenes and the scale of wireless devices brings new challenges, the distribution and the management of the identity information are increasingly complex, and the identity information cannot be well responded to physical attacks, and an attacker can easily access or steal identity-related information by damaging the hardware of the wireless devices.

(2) The existing wireless device identification scheme based on signal radiation characteristics is still limited by limited identification accuracy, and cannot be widely applied in practice at present.

(3) The recognizable scale of a single feature, i.e. the maximum number of devices that can be discerned without a significant error rate, is limited; the multi-feature combined recognition scheme is an effective solution, but the deep research on the mechanism of hardware defects is lacked at present, so that the development of more signal feature extraction is hindered.

The difficulty in solving the above problems and defects is:

(1) the wireless device identity information is stored in the device, and as the application scene types and scales of the wireless devices increase, the complexity of the distribution and management of the device identity information also increases day by day. This problem cannot be ameliorated by improving existing methods, requiring conceptual updates in the device identification technology.

(2) Radiation signatures are non-reproducible and non-counterfeitable statistical signatures measured from wireless signals, which are mainly caused by physical imperfections in the transceiver of the wireless device that occur randomly during the manufacturing process, and are difficult to control, predict and clone.

The significance of solving the problems and the defects is as follows:

(1) the conceptual updating is carried out on the aspect of equipment identification technology, and the problems of equipment identity information distribution and management in a complex wireless equipment environment are solved.

(2) Performing signal feature modeling according to the radiation features of the equipment, calculating the fractal dimension of the signal, and improving the feature extraction precision; and the device identification accuracy is improved by jointly using a plurality of device characteristics.

Disclosure of Invention

The present invention provides a wireless device identification method, a system and a data processing terminal, and particularly relates to a wireless device identification method, a system and a data processing terminal based on a DSSS frame preamble fractal dimension.

The present invention is achieved as such, and a wireless device identification method includes the steps of:

step one, constructing a wireless equipment identification model based on DSSS frame leading fractal dimension, and ensuring the accuracy of extracted equipment characteristics;

sampling signals and extracting signal characteristics according to the radiation defects of the equipment;

step three, calculating a signal fractal dimension by utilizing the DSSS frame preamble for the equipment identification stage;

and step four, identifying the equipment by using the DSSS frame preamble fractal dimension.

Further, in the step one, the constructing a wireless device identification model based on the DSSS frame preamble fractal dimension includes:

each frame signal is subjected to the influence of hardware defects of radio frequency links of a transmitter and a receiver, wherein the influence of the hardware defects comprises a digital-to-analog converter conversion error, a frequency error of a frequency converter and a power amplifier amplification deviation; any hardware defect in the radio frequency link can cause a wireless signal to present a specific radiation characteristic, and the radiation characteristic is estimated by using a physical layer radio frequency signal at a receiver end;

the radiation characteristic is referred to as the characteristic for short, because the frame preamble field is defined in advance in the wireless communication protocol, the characteristic estimation is carried out by using the preamble field of the wireless signal frame, and the receiver can use the preamble field as a reference signal of the characteristic estimation; the payload part refers to actual communication transmission data, and the content of the actual communication transmission data cannot be known by a receiving end in advance, so the part is not used for characteristic estimation for description;

the leader field is a fixed sequence consisting of 0-1 symbols and is defined as follows:

X＝[x₀，x₁，...，x_i，...x_N]；

wherein x is_1≤i≤UI/Q modulation symbols.

Further, in step two, the sampling the signal and extracting the signal characteristics according to the device radiation defect, including sampling by using a frame preamble field defined in the 802.11b specification PLCP, and extracting the signal characteristics according to each frame, includes:

(1) sampling and frame detection: sampling signals at a sampling frequency of 22MHz, and roughly judging the starting position and the ending position of each frame in the sampled signals by using a power matching algorithm;

(2) frame positioning: after frame detection is roughly carried out, a sliding window is adopted to match a received signal sample with a 128-bit lead code specified by a protocol, so that accurate frame positioning is realized;

(3) and (3) feature calculation: after accurate frame positioning, using frame preamble for feature calculation; and calculating signal characteristics based on the original sample, and calculating fractal dimension characteristics by using the despread frame preamble.

Further, in step (3), the signal characteristics include carrier frequency offset, I/Q imbalance, amplitude error, phase error, and synchronization error.

Further, in step three, the calculating the fractal dimension of the signal by using the DSSS frame preamble includes:

the actual received signal is represented as:

X′＝[x′₀，x′₁，...，x′_i，...x′_N]；

the fractal dimension is then determined by unifying the received symbols into the same reference symbol by phase rotationThe calculation method comprises the following steps:

wherein L is_τ，nAnd L_1，nRespectively as follows:

L_τ，n＝|x′_n·τ-x_(n-1)·τ|；

L_1，n＝|x′_n-x′_n-1|；

where X' denotes a receiver-side despread frame preamble field, where each symbol differs from the symbol in X due to a physical defect in the radio frequency chain.

Further, in the fourth step, the device identification is performed by using the preamble fractal dimension of the DSSS frame, including:

(1) acquiring 1000 frames of signals for each network card, respectively calculating six equipment characteristics including fractal dimension, carrier frequency offset, amplitude error, I/Q imbalance, phase error and synchronous error of each frame sample, forming six-dimensional signal characteristic vectors, and normalizing the calculated characteristic vectors to serve as original training data;

(2) training input data by adopting a kNN algorithm and an SVM algorithm, and carrying out equipment identification on a signal frame of unknown equipment according to a trained model;

(3) and extracting a signal feature vector from the signal frame of the unknown equipment according to a feature extraction method, inputting the signal feature vector into the trained kNN and SVM models, and obtaining a model output result which is the identified equipment identity.

Another object of the present invention is to provide a wireless device identification system applying the wireless device identification method, the wireless device identification system comprising:

the model building module is used for building a wireless equipment identification model based on DSSS frame preamble fractal dimension;

the characteristic extraction module is used for sampling signals and extracting the characteristics of the signals according to the radiation defects of the equipment;

a fractal dimension calculating module for calculating a signal fractal dimension by using the DSSS frame preamble;

and the equipment identification module is used for identifying the equipment by utilizing the DSSS frame preamble fractal dimension.

It is a further object of the invention to provide a computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of:

constructing a wireless equipment identification model based on DSSS frame leading fractal dimension; sampling signals and extracting signal characteristics according to the radiation defects of the equipment; calculating a signal fractal dimension by utilizing a DSSS frame preamble; and identifying the equipment by using the DSSS frame preamble fractal dimension.

It is another object of the present invention to provide a computer-readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of:

constructing a wireless equipment identification model based on DSSS frame leading fractal dimension; sampling signals and extracting signal characteristics according to the radiation defects of the equipment; calculating a signal fractal dimension by utilizing a DSSS frame preamble; and identifying the equipment by using the DSSS frame preamble fractal dimension.

Another object of the present invention is to provide an information data processing terminal for implementing the wireless device identification system.

By combining all the technical schemes, the invention has the advantages and positive effects that: the wireless equipment identification method provided by the invention uses DSSS frame leading fractal dimension as an available new radiation characteristic in wireless equipment identification, and the characteristic effectively improves the identification accuracy of physical layer equipment. The invention provides a new radiation characteristic, which is called a fractal dimension of a DSSS frame preamble, and can improve the accuracy of wireless equipment identification in a DSSS system. The wireless equipment identification method provided by the invention selectively combines the DSSS frame preamble fractal dimension, the carrier frequency offset, the amplitude error, the I/Q imbalance, the phase error and the synchronization error, and effectively improves the identification accuracy of physical layer equipment by combining multiple characteristics.

Experimental results show that the DSSS frame preamble fractal dimension-based wireless equipment identification method provided by the invention can ensure performance gain compared with known characteristics. For the wired case, the SVM-based recognition accuracy is improved by 4.3%, and the kNN-based recognition accuracy is improved by 5.3%. Under different noise environments, considerable performance gain can still be guaranteed, and especially, the identification accuracy is improved more remarkably under the high-noise environment.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments of the present invention will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a flowchart of a wireless device identification method according to an embodiment of the present invention.

Fig. 2 is a schematic diagram of a wireless device identification method according to an embodiment of the present invention.

Fig. 3 is an architecture diagram of a wireless device identification method according to an embodiment of the present invention.

FIG. 4 is a block diagram of a wireless device identification system according to an embodiment of the present invention;

in the figure: 1. a model building module; 2. a feature extraction module; 3. a fractal dimension calculation module; 4. and a device identification module.

Fig. 5 is a schematic diagram of IEEE802.11b frame timing signals according to an embodiment of the present invention.

Fig. 6 is a diagram of a transmitter-receiver radio frequency link provided by an embodiment of the present invention.

FIG. 7 is a line graph of experimental results provided by an embodiment of the present invention;

in the figure: 1 is DSSS frame leading fractal dimension characteristics, 2, 3, 4, 5 and 6 respectively represent carrier frequency offset, amplitude error, I/Q imbalance, phase error and synchronization error; the red line is the accuracy of device identification without using fractal dimension features (i.e., using 2, 3, 4, 5, 6 features), and the black line is the accuracy of device identification with adding fractal dimension features (i.e., using 1, 2, 3, 4, 5, 6 features).

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

In view of the problems in the prior art, the present invention provides a method, a system and a data processing terminal for identifying a wireless device, which are described in detail below with reference to the accompanying drawings.

As shown in fig. 1, a method for identifying a wireless device according to an embodiment of the present invention includes the following steps:

s101, constructing a wireless equipment identification model based on DSSS frame leading fractal dimension, and ensuring the accuracy of extracted equipment characteristics;

s102, sampling signals and extracting signal characteristics according to the radiation defects of the equipment;

s103, calculating a signal fractal dimension by utilizing the DSSS frame preamble for the equipment identification stage;

and S104, identifying the equipment by using the DSSS frame preamble fractal dimension.

Fig. 2 is a schematic diagram of a wireless device identification method according to an embodiment of the present invention, and fig. 3 is an architecture diagram of the wireless device identification method according to the embodiment of the present invention.

As shown in fig. 4, a wireless device identification system provided in an embodiment of the present invention includes:

the model building module 1 is used for building a wireless equipment identification model based on DSSS frame preamble fractal dimension;

the characteristic extraction module 2 is used for sampling signals and extracting signal characteristics according to the radiation defects of the equipment;

a fractal dimension calculating module 3, configured to calculate a signal fractal dimension by using a DSSS frame preamble;

and the device identification module 4 is used for identifying the device by using the DSSS frame preamble fractal dimension.

The technical solution of the present invention will be further described with reference to the following examples.

As shown in fig. 3 and fig. 6, a DSSS frame preamble fractal dimension-based wireless device identification system provided in an embodiment of the present invention includes the following steps:

1. wireless equipment identification model based on DSSS frame preamble fractal dimension

As shown in fig. 6, each frame signal experiences the effects of hardware imperfections in the transmitter and receiver rf chains, including but not limited to digital-to-analog converter conversion errors, frequency converter frequency errors, and power amplifier amplification deviations. Any hardware defect in the rf link may cause the wireless signal to exhibit a specific radiation characteristic, which may be estimated using the receiver-side physical layer rf signal. For convenience, the radiation features will be referred to hereinafter simply as features. Feature estimation is typically performed using the preamble field of a wireless signal frame, mainly because the frame preamble field is defined in advance in the wireless communication protocol; therefore, the receiver can use it as a reference signal for feature estimation. Fig. 5 shows an exemplary diagram of an IEEE802.11b frame obtained in an experiment, in which a payload portion refers to actual communication transmission data, the content of which cannot be known by a receiving end in advance, and thus this portion is not used for the characteristic estimation described in the present invention. The preamble field is typically a fixed sequence of 0-1 symbols, which can be defined as follows:

X＝[x₀，x₁，...，x_i，...x_N]

wherein X is a transmitting end preamble field vector, X_1≤i≤NThe symbols are modulated for I/Q quadrature co-directional signals.

2. Sampling signals and feature extraction using device radiation defects

The frame preamble field defined in the 802.11b specification (physical layer convergence protocol, PLCP) is used for sampling and extracting signal features from each frame.

(1) Sampling and frame detection: the method comprises the steps of firstly sampling signals at a sampling frequency of 22MHz to obtain original signals, and then roughly judging the starting position and the ending position of each frame in the sampled signals by using a power matching algorithm.

(2) Frame positioning: after frame detection is roughly performed, the invention matches a received signal sample with a 128-bit lead code specified by a protocol by adopting a sliding window, thereby realizing accurate frame positioning.

(3) And (3) feature calculation: once the frame is accurately positioned, the present invention uses the frame preamble for rf signature computation. The invention calculates the signal characteristics of carrier frequency offset, I/Q imbalance, amplitude error, phase error, synchronous error and the like based on the original sample, and further calculates the fractal dimension characteristic by utilizing the despread frame preamble.

3. Calculating frame leading fractal dimension by using characteristics

The actually received wireless signal is represented as:

X′＝[x′₀，x′₁，...，x′_i，...x′_N]

for the purpose of calculation, the fractal dimension is determined by unifying the received symbols into the same reference symbols by phase rotationThe calculation method comprises the following steps:

where τ is the scale of units of measurement, L_τ，nAnd L_1，nRespectively as follows:

L_τ，n＝|x′_n·τ-x′_(n-1)·τ|

L_1，n＝|x′_n-x′_n-1|

here, X' denotes a receiver-side despread frame preamble field, where each symbol is different from the symbol in X due to a physical defect in the radio frequency chain.

4. Device identification using frame preamble fractal dimension

In the experiment, 25 network cards are selected to run an IEEE802.11B protocol, and the USRP B210 is used for deploying an IEEE802.11B receiving protocol to communicate with the network cards so as to collect physical layer signals. Collecting 1000 frame signals for each network card, respectively calculating six equipment characteristics of fractal dimension, carrier frequency offset, amplitude error, I/Q unbalance, phase error and synchronous error of each frame sample after frame positioning and frame leading interception are carried out, forming six-dimensional signal characteristic vectors, and normalizing the calculated characteristic vectors to serve as original training data; and training input data by adopting a kNN algorithm and an SVM algorithm, and identifying the equipment according to the trained model aiming at the signal frame of the unknown equipment. And extracting a signal feature vector from the signal frame of the unknown equipment according to the feature extraction method, inputting the signal feature vector into the trained kNN and SVM models, and obtaining a model output result, namely the identified equipment identity.

Fig. 7 shows experimental data of the present invention, and results show that the method for identifying a wireless device based on a DSSS frame preamble fractal dimension of the present embodiment can ensure performance gain compared with known features. For the wired case, the SVM-based recognition accuracy is improved by 4.3%, and the kNN-based recognition accuracy is improved by 5.3%. Under different noise environments, considerable performance gain can still be guaranteed, and especially, the identification accuracy is improved more remarkably under the high-noise environment.

The test in the above embodiment is the ieee802.11b communication protocol, which adopts the direct sequence spread spectrum mode; in addition, any communication protocol using direct sequence spread spectrum mode can use the proposed method for device feature extraction and device identification. Identification schemes based on this physical layer characteristic have many application values in future wireless communication networks. For example: the method comprises the steps of tracking equipment based on physical layer equipment identification, realizing differentiated management and control based on the equipment physical layer identification, fusing heterogeneous network safety based on the physical layer equipment identification, enhancing the concealment of user identity information and the like.

In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When used in whole or in part, can be implemented in a computer program product that includes one or more computer instructions. When loaded or executed on a computer, cause the flow or functions according to embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, the computer instructions may be transmitted from one website site, computer, server, or data center to another website site, computer, server, or data center via wire (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL), or wireless (e.g., infrared, wireless, microwave, etc.)). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that includes one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.

The above description is only for the purpose of illustrating the present invention and the appended claims are not to be construed as limiting the scope of the invention, which is intended to cover all modifications, equivalents and improvements that are within the spirit and scope of the invention as defined by the appended claims.