CN116546505A - MIMO equipment radio frequency fingerprint identification method and device in multi-user wireless communication scene - Google Patents

Abstract

The application provides a MIMO equipment radio frequency fingerprint identification method under a multi-user wireless communication scene, which comprises the following steps: receiving training signals and any unknown signals transmitted by the other end at a wireless access point/user end, differentiating the training signals to generate a wireless channel matrix, generating signal vectors according to the any unknown signals and the wireless channel matrix, and processing all the signal vectors to obtain all link signals of a transmitting end; training a radio frequency fingerprint feature identification model by using all link signals, and generating a device radio frequency fingerprint library; and acquiring a signal to be identified of the equipment to be authenticated, inputting the signal to be identified into the model for identification, obtaining the radio frequency fingerprint characteristics of the equipment to be authenticated, and comparing the radio frequency fingerprint characteristics with an equipment radio frequency fingerprint library to obtain the radio frequency fingerprint identification result of the equipment to be authenticated. The invention adopting the scheme can realize the separation of a plurality of devices and a plurality of link signals of each device at the same time, and ensure that fingerprint aliasing does not occur, thereby realizing the radio frequency fingerprint identification of the plurality of devices.

Description

Method and device for radio frequency fingerprint identification of MIMO equipment in multi-user wireless communication scenario

technical field

The present application relates to the technical field of wireless network access authentication, and in particular to a method and device for identifying radio frequency fingerprints of MIMO devices in a multi-user wireless communication scenario.

Background technique

In the real environment, wireless networks are extremely vulnerable to various eavesdropping and interference attacks, which bring a series of network security threats such as illegal access and data leakage. Strict access authentication for devices is one of the main means to prevent these threats. Existing access authentication technologies are mainly divided into two categories. The first type is access authentication technology based on cryptographic methods, and the second type is called radio frequency fingerprint identification technology.

The access authentication technology based on cryptographic methods adopts cryptographic methods such as encryption, hash function, and digital signature, and performs access authentication based on the secret value owned by the device to be authenticated. For example, in a WiFi wireless network, a wireless access point (or router) generally uses a WPA or WPA2 protocol to perform access authentication on a mobile device. The WPA or WPA2 protocol uses an encryption algorithm, based on the same pre-shared key (that is, a secret value) used by both parties to the protocol for authentication and key exchange.

The radio frequency fingerprint identification technology is based on the hardware characteristics of the device to be authenticated or the characteristics of the wireless channel to authenticate it, which can be divided into two categories: the first is the radio frequency fingerprint identification technology based on the wireless channel, which identifies the authenticated device and the authenticated device The second type is the radio frequency fingerprint identification technology based on hardware features, which identifies the hardware features of the authenticated device. The hardware characteristics here mainly refer to the circuit tolerances of components and associated circuits that deviate from the nominal values due to factors such as materials, processes, and aging of the RF front-end of the equipment, such as clock jitter, DAC sampling error, mixer frequency offset, Power amplifier nonlinear effects, IQ imbalance, etc.

The access authentication technology based on cryptography increases the message size, transmission overhead and delay, and it is difficult to meet the communication requirements of high bandwidth and low delay in wireless devices with limited computing power. Additionally, as the cryptanalysis industry grows, older cryptography methods continue to be found to have significant vulnerabilities. Quantum computers can also efficiently crack most cryptography methods, posing a great threat to authentication technologies based on cryptography methods. The RF fingerprint identification technology based on hardware features does not have the shortcomings of the above-mentioned cryptography-based methods, but most of the existing research focuses on the RF fingerprint extraction and identification of single-antenna devices, and multi-antenna multi-link devices (ie MIMO devices), or There is no clear solution for radio frequency fingerprinting of all access users in the scenario where multiple users communicate at the same time.

Contents of the invention

The first purpose of this application is to propose a radio frequency fingerprint identification method for MIMO devices in a multi-user wireless communication scenario, which solves the technical problem that there is no system identification of fingerprints for multi-antenna and multi-link devices in existing methods, and realizes simultaneous Separation of multiple devices and multiple link signals of each device, and completely retain the fingerprint information of the corresponding link during separation without fingerprint aliasing, so that multiple link fingerprints of a single MIMO transmitter can be passed through similar A single-antenna single-link transmitter method for fingerprint extraction and identification.

The second purpose of the present application is to propose a radio frequency fingerprint identification device for MIMO equipment in a multi-user wireless communication scenario.

In order to achieve the above purpose, the embodiment of the first aspect of the present application proposes a method for identifying radio frequency fingerprints of MIMO devices in a multi-user wireless communication scenario, including: receiving at the wireless access point/user end the information transmitted by the user end/wireless access point The training signal and any unknown signal are differentiated to generate a wireless channel matrix, and signal vectors are generated according to any unknown signal and wireless channel matrix, and all signal vectors of all user equipment are processed to obtain link signals of all user equipment , where, if the wireless access point is the signal transmitter, the user terminal is the signal receiver, if the user terminal is the signal transmitter, the wireless access point is the receiver, the training signal carries the training sequence, and any unknown signal carries the communication data ; Use all link signals of all devices as training data to train the radio frequency fingerprint feature recognition model, obtain the trained radio frequency fingerprint feature recognition model, and generate the device radio frequency fingerprint library, wherein the device radio frequency fingerprint library is used to store the training results Device information and corresponding RF fingerprint features; obtain the signal to be identified of the device to be authenticated, and input the signal to be identified into the trained RF fingerprint feature recognition model for identification, obtain the RF fingerprint feature of the device to be authenticated, and convert the RF fingerprint of the device to be identified The features are compared with the device radio frequency fingerprint database to obtain the radio frequency fingerprint identification result of the device to be authenticated.

The radio frequency fingerprint identification method for MIMO devices in the multi-user wireless communication scenario of the embodiment of the present application proposes a link separation technology for the phenomenon that the signals of multiple devices in the uplink communication scenario generate time-space aliasing at the wireless access point, so that it can Differentiate the signals of multiple devices, and then perform fingerprint extraction and identification for multiple devices. The link separation technology proposed in this application uses matrix operations, which can simultaneously realize the separation of multiple devices and multiple link signals of each device, and When separating, the fingerprint information of the corresponding link is completely preserved without fingerprint aliasing, so that the fingerprints of multiple links of a single MIMO transmitter can be extracted and identified by a method similar to that of a single-antenna single-link transmitter.

Optionally, in one embodiment of the present application, if the wireless communication scenario is an uplink communication scenario, the method includes:

receiving training signals transmitted by multiple client terminals at the wireless access point, and performing difference on the training signals to obtain a wireless channel matrix;

Receive any unknown signal transmitted by multiple clients through the wireless access point at each moment of the preset time period;

Calculate the signal vector corresponding to this moment according to any unknown signal transmitted by multiple users received at each moment and the wireless channel matrix;

The signal vectors at all moments within the preset time period are spliced, and all link signals of all devices are obtained according to the spliced signal vectors.

Optionally, in one embodiment of the present application, the training signal is differentiated to obtain a wireless channel matrix, including:

Performing a difference on training signals transmitted by multiple user terminals, and splicing the differenced data to obtain first spliced data;

Performing a difference on the training signal received by the wireless access point, and splicing the differenced data to obtain second spliced data;

A wireless channel matrix is obtained by calculating according to the first spliced data and the second spliced data.

Optionally, in one embodiment of the present application, the training signals simultaneously transmitted by multiple client terminals are expressed as:

x ^t0 ,x ^t1 ,…,x ^tK (K>M)

Among them, t0, t1, ..., tK is the time of transmission, M is the total number of antennas at the receiving end, and K is the number of training signals transmitted;

The training signal received by the wireless access point is expressed as:

y ^t1 =HAx ^t1 +Hb+w ^t1

…

y ^tK ＝HAx ^tK +Hb+w ^tK

Among them, t0, t1, ..., tK are the time of transmission, H is the wireless channel matrix between the transmitting and receiving antennas stripped of the RF fingerprint characteristics of the transmitter, A is determined by the RF fingerprint characteristic parameter α _i , b is determined by the radio frequency fingerprint characteristic parameter β _i , b={β ₁ ,β ₂ ,…,β _M } ^T , w is noise;

The data after the difference of the training signals transmitted by multiple clients is expressed as:

The first spliced data is expressed as:

ΔX _t = [Δx _t1 ,Δx _t2 ,…,Δx _tK ]

The data after the difference of the training signal received by the wireless access point is expressed as:

The second spliced data is expressed as:

ΔY _t = [Δy _t1 ,Δy _t2 ,…,Δy _tK ]

The wireless channel matrix is expressed as:

Wherein, ΔY _t represents the second splicing data, and ΔX _t represents the first splicing data;

The signal vector is expressed as:

Wherein, y indicates that the wireless access point receives N signals, y={y ₁ ,y ₂ ,...,y _N } ^T , Represents the wireless channel matrix; the spliced signal vector is expressed as:

Z=[z ₁ ,z ₂ ,…]

All link signals for all devices are represented as:

in, Indicates the set of all link serial numbers contained in the l-th device, and Z ^j indicates the j-th row vector of Z.

Optionally, in one embodiment of the present application, if the wireless communication scenario is a downlink communication scenario, the method includes:

receiving multiple training signals transmitted by the wireless access point within multiple time periods at the user end, wherein the time periods are randomly selected by the wireless access point;

Differentiate multiple training signals to obtain a wireless channel matrix;

Splitting the wireless channel matrix according to multiple time periods to obtain a wireless channel matrix corresponding to each time period;

Receive all unknown signals transmitted by the wireless access point within multiple time periods at the user end;

According to the wireless channel matrix corresponding to each time period and the unknown signal received at the user end in each time period, the signal vector of the time period is calculated and generated, and the signal vectors of all time periods are spliced to obtain the spliced signal vector;

The spliced signal vectors of all client terminals are obtained, and all link signals of all devices are obtained according to the spliced signal vectors of all client terminals.

Optionally, in one embodiment of the present application, a plurality of training signals are differentiated to obtain a wireless channel matrix, including:

Performing a difference on multiple training signals transmitted by the wireless access point to obtain first difference data;

performing a difference on a plurality of training signals received by the user end to obtain second difference data;

A wireless channel matrix is obtained by calculating according to the first difference data and the second difference data.

Optionally, in one embodiment of the present application, the multiple training signals transmitted by the wireless access point are expressed as:

x ^t1 ,x ^t2 ,…,x ^tK

The multiple training signals received by the client are expressed as:

Among them, [Y _j ] _N×K is the signal received by the user terminal in the jth time period, H _j is the channel state corresponding to the jth time period, A is determined by the radio frequency fingerprint characteristic parameter α _i , b is determined by the radio frequency fingerprint characteristic parameter β _i , b={β ₁ ,β ₂ ,…,β _M } ^T , w is noise;

The first difference data is expressed as:

ΔX＝[x ^t2 -x ^t1 ,x ^t3 -x ^t2 ,…,x ^tK -x ^t(K-1) ] _M×(K-1)

Wherein, M is the number of radio frequency links of the wireless access point, and K is the number of training signals;

The second difference data is expressed as:

Among them, N is the number of radio frequency links at the user end, K is the number of training signals, and P is the number of time periods;

The wireless channel matrix is expressed as:

in,

The split wireless channel matrix is expressed as:

H ₁ A,H ₂ A,…,H _P A

Among them, P is the number of time periods;

The signal vector is expressed as:

z _M×K ＝(H _j A) ⁺ y

Among them, z _M×K is the signal vector corresponding to the received signal y _N×K , j is the time period of the signal, H _j is the channel state corresponding to the jth time period, and H _j A represents the jth channel state after splitting Wireless channel matrix;

The spliced signal vector is expressed as:

Z=[z ₁ ,z ₂ ,…]

All link signals for all devices are represented as:

in, Indicates the set of all link serial numbers contained in the l-th device, and Z ^j indicates the j-th row vector of Z.

Optionally, in one embodiment of the present application, the radio frequency fingerprint feature recognition model is a deep neural network, and all link signals of all devices are used as training data to train the radio frequency fingerprint feature recognition model and establish a device radio frequency fingerprint library, including :

Obtain pre-processed training signals corresponding to all link signals of all devices;

With the goal of minimizing the error between the real device label of the input signal and the estimated device label by the neural network, the deep neural network model is trained to obtain a trained radio frequency fingerprint feature recognition model.

Store the obtained device information and corresponding radio frequency fingerprint features into the device radio frequency fingerprint database.

In order to achieve the above purpose, the embodiment of the second aspect of the present invention proposes a radio frequency fingerprint identification device for MIMO equipment in a multi-user wireless communication scenario, including a signal processing module, a model training module, and an identification module, wherein,

The signal processing module is used to receive the training signal and any unknown signal transmitted by the user terminal/wireless access point at the wireless access point/client, and differentiate the training signal to generate a wireless channel matrix. According to any unknown signal and wireless channel The matrix generates signal vectors, processes all signal vectors of all user equipments, and obtains link signals of all user equipments, wherein, if the wireless access point is the signal transmitting end, the user end is the signal receiving end, and if the user end is the signal At the transmitting end, the wireless access point is the receiving end, the training signal carries the training sequence, and any unknown signal carries the communication data;

The model training module is used to use all link signals of all devices as training data to train the radio frequency fingerprint feature recognition model, obtain the trained radio frequency fingerprint feature recognition model, and generate the device radio frequency fingerprint database, wherein the device radio frequency fingerprint database is used It is used to store the device information generated by training and the corresponding RF fingerprint features;

The identification module is used to obtain the signal to be identified of the device to be authenticated, and input the signal to be identified into the trained radio frequency fingerprint feature recognition model for identification, obtain the radio frequency fingerprint feature of the device to be authenticated, and combine the radio frequency fingerprint feature of the device to be identified with the device radio frequency The fingerprint library is compared to obtain the radio frequency fingerprint identification result of the device to be authenticated.

Optionally, in one embodiment of the present application, the radio frequency fingerprint feature recognition model is a deep neural network, and all link signals of all devices are used as training data to train the radio frequency fingerprint feature recognition model and establish a device radio frequency fingerprint library, including :

Obtain pre-processed training signals corresponding to all link signals of all devices;

With the goal of minimizing the error between the real device label of the input signal and the estimated device label by the neural network, the deep neural network model is trained to obtain a trained radio frequency fingerprint feature recognition model.

Store the obtained device information and corresponding radio frequency fingerprint features into the device radio frequency fingerprint library.

Additional aspects and advantages of the application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the application.

Description of drawings

The above and/or additional aspects and advantages of the present application will become apparent and easy to understand from the following description of the embodiments in conjunction with the accompanying drawings, wherein:

FIG. 1 is a schematic diagram of multi-user MIMO uplink and downlink wireless communication according to an embodiment of the present application;

FIG. 2 is a schematic flowchart of a method for identifying radio frequency fingerprints of MIMO devices in a multi-user wireless communication scenario provided by Embodiment 1 of the present application;

FIG. 3 is a schematic structural diagram of a radio frequency fingerprint identification device for MIMO equipment in a multi-user wireless communication scenario provided by an embodiment of the present application.

Detailed ways

Embodiments of the present application are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary, and are intended to explain the present application, and should not be construed as limiting the present application.

In an existing multi-user MIMO wireless communication environment, periodic data communication is performed between multiple terminal users and a wireless access point (Access Point), and uplink communication and downlink communication are performed alternately. FIG. 1 is a schematic diagram of multi-user MIMO uplink and downlink wireless communication according to an embodiment of the present application. FIG. 1(a) is an uplink communication scenario, and FIG. 1(b) is a downlink communication scenario. As shown in FIG. 1(a), during uplink communication, Multiple users send signals to the wireless access point at the same time, as shown in Figure 1(b), during downlink communication, the wireless access point sends signals to multiple users. In an existing communication standard, the total number of antennas of a wireless access point during multi-user MIMO communication is greater than or equal to the total number of antennas of multiple users.

In the uplink communication scenario, the signal transmitting end is multiple users, and the signal receiving end is a wireless access point. Note that the total number of antennas at the transmitting end is M, and the total number of antennas at the receiving end is N, and N≥M; while in the downlink communication scenario, The signal transmitting end is a wireless access point, and the signal receiving end is multiple users. The total number of antennas at the transmitting end is M, and the total number of antennas at the receiving end is N, and N≤M.

At this time, the relationship between the received signal and the sent signal is:

y=HAx+Hb+w

Among them, x is the ideal transmission signal of the transmitter, H is the wireless channel matrix between the transmitting and receiving antennas stripped of the transmitter’s RF fingerprint characteristics (it can be reasonably assumed that in a short period of time, the communication channel status of each user is basically unchanged ), y is the multi-user aliasing signal received by the receiver, w is the noise, A and b are related to the RF fingerprint of the transmitter, and satisfy:

b={β ₁ ,β ₂ ,…,β _M } ^T

In the uplink communication scenario and the downlink communication scenario, the present invention respectively proposes corresponding radio frequency link separation methods for multi-user aliasing signals received by the receiver.

Note that the i-th RF link fingerprint feature is Its richness in frequency dependence and uniqueness in device association determine that it can be used for device radio frequency fingerprinting.

The method and device for identifying radio frequency fingerprints of MIMO devices in a multi-user wireless communication scenario according to the embodiments of the present application will be described below with reference to the accompanying drawings.

FIG. 2 is a schematic flowchart of a method for identifying radio frequency fingerprints of MIMO devices in a multi-user wireless communication scenario provided by Embodiment 1 of the present application.

As shown in Figure 2, the MIMO device radio frequency fingerprint identification method in the multi-user wireless communication scenario includes the following steps:

Step 201: Receive the training signal and any unknown signal transmitted by the user terminal/wireless access point at the wireless access point/client, and differentiate the training signal to generate a wireless channel matrix, and generate a signal according to any unknown signal and the wireless channel matrix vector, all signal vectors of all user equipment are processed to obtain link signals of all user equipment, wherein, if the wireless access point is the signal transmitting end, the user end is the signal receiving end, and if the user end is the signal transmitting end , the wireless access point is the receiving end, the training signal carries the training sequence, and any unknown signal carries the communication data;

Step 202, use all link signals of all devices as training data to train the radio frequency fingerprint feature recognition model, obtain the trained radio frequency fingerprint feature recognition model, and generate a device radio frequency fingerprint library, wherein the device radio frequency fingerprint library is used to store training Generated device information and corresponding RF fingerprint features;

Step 203, obtain the signal to be identified of the device to be authenticated, input the signal to be identified into the trained RF fingerprint feature recognition model for identification, obtain the RF fingerprint feature of the device to be authenticated, and combine the RF fingerprint feature of the device to be identified with the device RF fingerprint library By comparison, the radio frequency fingerprint identification result of the device to be authenticated is obtained.

The radio frequency fingerprint identification method for MIMO devices in the multi-user wireless communication scenario of the embodiment of the present application proposes a link separation technology for the phenomenon that the signals of multiple devices in the uplink communication scenario generate time-space aliasing at the wireless access point, so that it can Differentiate the signals of multiple devices, and then perform fingerprint extraction and identification for multiple devices. The link separation technology proposed in this application uses matrix operations, which can simultaneously realize the separation of multiple devices and multiple link signals of each device, and When separating, the fingerprint information of the corresponding link is completely preserved without fingerprint aliasing, so that the fingerprints of multiple links of a single MIMO transmitter can be extracted and identified by a method similar to that of a single-antenna single-link transmitter.

The MIMO device radio frequency fingerprint identification method in the multi-user wireless communication scenario of the embodiment of the present application can be applied to the uplink communication scenario and the downlink communication scenario of the multi-user MIMO communication environment. There are differences in the process of separating the radio frequency link in the two scenarios. The process of fingerprint extraction and identification is the same.

Optionally, in one embodiment of the present application, if the wireless communication scenario is an uplink communication scenario, the method includes:

Multiple users transmit K training signals to the wireless access point at the same time, and make differences on the training signals, and the wireless access point calculates the wireless channel matrix

Subsequently, at a certain moment, some users transmit any unknown signal x (the signal here carries communication data, but does not carry training sequence);

The wireless access point receives N signals, and can calculate the wireless channel matrix based on the received signals and Calculate the signal vector z _M×1 ;

The wireless access point splices the signal vectors at several times, and directly queries the attribution information of the radiation source in the signal according to the MAC layer protocol in the multi-user MIMO communication, and obtains the set of all link numbers contained in the l-th device Taking the multi-user MIMO uplink communication in the IEEE802.11ax standard as an example, the uplink communication process is controlled by the wireless access point through the trigger frame. The trigger frame clearly defines the USER INFO field for each user, including the 12-bit identity of each user Identification information (Associate Identifier, AID), resource unit (Resource Unit, RU) allocation, coding, modulation information, etc., so that all link signals of all devices can be obtained.

Optionally, in one embodiment of the present application, the training signal is differentiated to obtain a wireless channel matrix, including:

Performing a difference on training signals transmitted by multiple user terminals, and splicing the differenced data to obtain first spliced data;

Performing a difference on the training signal received by the wireless access point, and splicing the differenced data to obtain second spliced data;

The wireless access point calculates a wireless channel matrix according to the first spliced data and the second spliced data.

Optionally, in one embodiment of the present application, the training signals simultaneously transmitted by multiple client terminals are expressed as:

x ^t0 ,x ^t1 ,…,x ^tK (K>M)

Among them, record the transmission time as t0, t1, ..., tK respectively, M is the total number of antennas at the receiving end, and K is the number of training signals transmitted;

The training signal received by the wireless access point is expressed as:

y ^t1 =HAx ^t1 +Hb+w ^t1

…

y ^tK ＝HAx ^tK +Hb+w ^tK

Among them, t0, t1, ..., tK are the time of transmission, H is the wireless channel matrix between the transmitting and receiving antennas stripped of the RF fingerprint characteristics of the transmitter, A and b are related to the RF fingerprint of the transmitter, and w is noise;

Differentiate the transmitted training signals x ^t0 , x ^t1 ,…,x ^tK :

By splicing Δx _t1 , Δx _t2 ,…Δx _tK , we can get:

ΔX _t = [Δx _t1 ,Δx _t2 ,…,Δx _tK ]

Differentiate the signals received at different times:

Among them, ΔW _tj represents the noise term, H is the wireless channel matrix between the transmitting and receiving antennas stripped of the RF fingerprint characteristics of the transmitter, and A is related to the RF fingerprint of the transmitter;

Splicing Δy _t1 , Δy _t2 ,…, we can get:

ΔY _t =[Δy _t1 ,Δy _t2 ,…,Δy _tK ]=HAΔX _t +ΔW _t

Among them, H is the wireless channel matrix between the transmitting and receiving antennas stripped of the RF fingerprint characteristics of the transmitter, A is related to the RF fingerprint of the transmitter, and ΔW _t represents the noise term;

Wireless Access Point Computing

Among them, (*) ⁺ is a generalized inverse of a matrix, ignoring the influence of the noise term ΔW _t , ΔY _t represents the second spliced data, and ΔX _t represents the first spliced data;

The wireless access point receives N signals as y={y ₁ ,y ₂ ,…,y _N } ^T , and the signal vector z _M×1 is expressed as:

Wherein, y indicates that the wireless access point receives N signals, y={y ₁ ,y ₂ ,...,y _N } ^T , w is noise;

The spliced signal vector is expressed as:

Z=[z ₁ ,z ₂ ,…]

All link signals for all devices are represented as:

in, Indicates the set of all link serial numbers contained in the l-th device, and Z ^j indicates the j-th row vector of Z.

Optionally, in an embodiment of the present application, in a downlink communication scenario, a user terminal with a small number of antennas needs to perform radio frequency fingerprinting on a wireless access point with a large number of antennas. Due to the low-rank characteristics of the wireless channel matrix, the radio frequency link separation method in the uplink communication scenario is no longer applicable to the downlink communication. If the wireless communication scenario is a downlink communication scenario, the method includes:

The wireless access point selects P time periods, and the channel states H ₁ , H ₂ , . . . , H _P corresponding to the P time periods are different from each other. Transmitting K training signals to the user end in each time period;

The user terminal differentiates the K training signals to obtain the wireless channel matrix

by splitting The user end can obtain the matrix H ₁ A, H ₂ A,...,H _P A of P time periods respectively;

Subsequently, for all unknown signals x transmitted by the wireless access point in the past P time periods, obtain the corresponding received signal y _N×1 of the user terminal;

The user end calculates and generates signal vectors for this time period according to the wireless channel matrix corresponding to each time period and the unknown signals received in each time period, and splices the signal vectors of all time periods to obtain a spliced signal vector;

The spliced signal vectors of all client terminals are obtained, and all link signals of all devices are obtained according to the spliced signal vectors of all client terminals.

Optionally, in one embodiment of the present application, a plurality of training signals are differentiated to obtain a wireless channel matrix, including:

Performing a difference on the K training signals transmitted by the wireless access point to obtain first difference data;

Performing a difference on the K training signals received by the user end to obtain second difference data;

A wireless channel matrix is obtained by calculating according to the first difference data and the second difference data.

Optionally, in one embodiment of the present application, the multiple training signals transmitted by the wireless access point are expressed as:

x ^t1 ,x ^t2 ,…,x ^tK

The signal received by the user terminal in the jth time period (1≤j≤P) is:

Wherein, H _j is the channel state corresponding to the jth time period;

Compute the difference of K training signals:

ΔX＝[x ^t2 -x ^t1 ,x ^t3 -x ^t2 ,…, ^tK -x ^t(K-1) ] _M×(K-1)

And the difference of the received signal under the jth time period:

It can be seen

ΔY _j ＝H _j AΔX+ΔW _j

definition

then there is

The user terminal estimates the wireless channel matrix by the following formula

The matrix of the split P time periods is expressed as:

H ₁ A,H ₂ A,…,H _P A

For the received signal y _N×1 , calculate the signal vector z _N×1 :

z _M×K ＝(H _j A) ⁺ y

≈x+A ^-1 b

Among them, j is the time period of the signal. Based on the assumption of short-term invariance of the RF fingerprint matrix A and vector b, each row z _i in the vector z is only related to the signal and RF fingerprint transmitted by the i-th RF link at the transmitter. Characteristic parameters α _i , β _i , and receiver noise correlation;

The user side splices several signal vectors:

Z=[z ₁ ,z ₂ ,…]

Different from the uplink communication scenario, all link signals in Z belong to one device, that is, the wireless access point;

All link signals for all devices are represented as:

in, Indicates the set of all link serial numbers contained in the l-th device, and Z ^j indicates the j-th row vector of Z.

Optionally, in one embodiment of the present application, the radio frequency fingerprint feature recognition model is a deep neural network, and all link signals of all devices are used as training data to train the radio frequency fingerprint feature recognition model and establish a device radio frequency fingerprint library, including : Extract the RF fingerprint of the device, perform model training, and store the data generated by the training to obtain the RF fingerprint library of the device. in particular:

All link signals {z ₁ ,z ₂ ,…,z _q } of all devices are used as the training set of the self-supervised denoising autoencoder, with the goal of minimizing the error between the input signal and the denoising reconstructed signal, training The autoencoder learns the potential hidden features of the signal to be recognized {z ₁ ,z ₂ ,…,z _q } and the encoder/decoder neural network parameters.

Store the device information and the corresponding RF fingerprint features generated by training into the RF fingerprint library to generate the device RF fingerprint library.

In order to realize the above embodiments, the present application also proposes a device for identifying radio frequency fingerprints of MIMO devices in a multi-user wireless communication scenario.

FIG. 3 is a schematic structural diagram of a radio frequency fingerprint identification device for MIMO equipment in a multi-user wireless communication scenario provided by an embodiment of the present application.

As shown in FIG. 3 , the radio frequency fingerprint identification device for MIMO equipment in the multi-user wireless communication scenario includes a signal processing module, a model training module, and an identification module, wherein,

The signal processing module is used to receive the training signal and any unknown signal transmitted by the user terminal/wireless access point at the wireless access point/client, and differentiate the training signal to generate a wireless channel matrix. According to any unknown signal and wireless channel The matrix generates signal vectors, processes all signal vectors of all user equipments, and obtains link signals of all user equipments, wherein, if the wireless access point is the signal transmitting end, the user end is the signal receiving end, and if the user end is the signal At the transmitting end, the wireless access point is the receiving end, the training signal carries the training sequence, and any unknown signal carries the communication data;

The model training module is used to use all link signals of all devices as training data to train the radio frequency fingerprint feature recognition model, obtain the trained radio frequency fingerprint feature recognition model, and generate the device radio frequency fingerprint database, wherein the device radio frequency fingerprint database is used It is used to store the device information generated by training and the corresponding RF fingerprint features;

The identification module is used to obtain the signal to be identified of the device to be authenticated, and input the signal to be identified into the trained radio frequency fingerprint feature recognition model for identification, obtain the radio frequency fingerprint feature of the device to be authenticated, and combine the radio frequency fingerprint feature of the device to be identified with the device radio frequency The fingerprint library is compared to obtain the radio frequency fingerprint identification result of the device to be authenticated.

Optionally, in one embodiment of the present application, the radio frequency fingerprint feature recognition model is a deep neural network, and all link signals of all devices are used as training data to train the radio frequency fingerprint feature recognition model and establish a device radio frequency fingerprint library, including :

Obtain pre-processed training signals corresponding to all link signals of all devices;

With the goal of minimizing the error between the real device label of the input signal and the estimated device label by the neural network, the deep neural network model is trained to obtain a trained radio frequency fingerprint feature recognition model.

The obtained device information and corresponding radio frequency fingerprint features are stored in the device radio frequency fingerprint database.

It should be noted that the foregoing explanations for the embodiment of the radio frequency fingerprint identification method for MIMO devices in a multi-user wireless communication scenario are also applicable to the radio frequency fingerprint identification device for MIMO devices in a multi-user wireless communication scenario in this embodiment, and will not be repeated here. repeat.

In the description of this specification, descriptions referring to the terms "one embodiment", "some embodiments", "example", "specific examples" or "some examples" mean specific features described in connection with the embodiment or example, A structure, material or characteristic is included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the described specific features, structures, materials or characteristics may be combined in any suitable manner in any one or more embodiments or examples. In addition, those skilled in the art can combine and combine different embodiments or examples and features of different embodiments or examples described in this specification without conflicting with each other.

In addition, the terms "first" and "second" are used for descriptive purposes only, and cannot be interpreted as indicating or implying relative importance or implicitly specifying the quantity of indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include at least one of these features. In the description of the present application, "plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

Any process or method descriptions in flowcharts or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing custom logical functions or steps of a process , and the scope of preferred embodiments of the present application includes additional implementations in which functions may be performed out of the order shown or discussed, including in substantially simultaneous fashion or in reverse order depending on the functions involved, which shall It should be understood by those skilled in the art to which the embodiments of the present application belong.

The logic and/or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium, For use with instruction execution systems, devices, or devices (such as computer-based systems, systems including processors, or other systems that can fetch instructions from instruction execution systems, devices, or devices and execute instructions), or in conjunction with these instruction execution systems, devices or equipment for use. For the purposes of this specification, a "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use in or in conjunction with an instruction execution system, device or device. More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection with one or more wires (electronic device), portable computer disk case (magnetic device), random access memory (RAM), Read Only Memory (ROM), Erasable and Editable Read Only Memory (EPROM or Flash Memory), Fiber Optic Devices, and Portable Compact Disc Read Only Memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program can be printed, as it may be possible, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or other suitable processing if necessary. The program is processed electronically and stored in computer memory.

It should be understood that each part of the present application may be realized by hardware, software, firmware or a combination thereof. In the embodiments described above, various steps or methods may be implemented by software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented by any one or a combination of the following techniques known in the art: a discrete Logic circuits, ASICs with suitable combinational logic gates, Programmable Gate Arrays (PGA), Field Programmable Gate Arrays (FPGA), etc.

Those of ordinary skill in the art can understand that all or part of the steps carried by the methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. During execution, one or a combination of the steps of the method embodiments is included.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing module, each unit may exist separately physically, or two or more units may be integrated into one module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software function modules. If the integrated modules are implemented in the form of software function modules and sold or used as independent products, they can also be stored in a computer-readable storage medium.

The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, and the like. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limitations on the present application, and those skilled in the art can make the above-mentioned The embodiments are subject to changes, modifications, substitutions and variations.

Claims (10)

1. a MIMO device radio frequency fingerprint identification method under the multi-user wireless communication scene, is characterized in that, comprises the following steps: Receive the training signal and any unknown signal transmitted by the user terminal/wireless access point at the wireless access point/client, and perform a difference on the training signal to generate a wireless channel matrix, according to the arbitrary unknown signal and the wireless channel The matrix generates signal vectors, processes all signal vectors of all user equipments, and obtains link signals of all user equipments, wherein, if the wireless access point is a signal transmitting end, the user end is a signal receiving end, if When the user end is a signal transmitting end, the wireless access point is a receiving end, the training signal carries a training sequence, and the arbitrary unknown signal carries communication data; All link signals of all devices are used as training data to train the radio frequency fingerprint feature recognition model, obtain the trained radio frequency fingerprint feature recognition model, and generate a device radio frequency fingerprint library, wherein the device radio frequency fingerprint library is used to store training generated Device information and corresponding RF fingerprint features; Obtain the unidentified signal of the equipment to be authenticated, and input the unidentified signal into the trained radio frequency fingerprint feature recognition model for identification, obtain the radio frequency fingerprint feature of the equipment to be authenticated, combine the radio frequency fingerprint feature of the equipment to be identified with the The radio frequency fingerprint database of the above-mentioned equipment is compared to obtain the radio frequency fingerprint identification result of the equipment to be authenticated. 2. The method according to claim 1, wherein if the wireless communication scenario is an uplink communication scenario, the method comprises: receiving training signals transmitted by multiple user terminals at the wireless access point, and performing a difference on the training signals to obtain the wireless channel matrix; receiving any unknown signal transmitted by multiple user terminals through the wireless access point at each moment of the preset time period; According to any unknown signal transmitted by multiple users received at each moment and the wireless channel matrix, the signal vector corresponding to the moment is calculated; The signal vectors at all moments within the preset time period are spliced, and all link signals of all the devices are obtained according to the spliced signal vectors. 3. The method according to claim 2, wherein said performing a difference on said training signal to obtain said wireless channel matrix comprises: Performing a difference on the training signals transmitted by the plurality of user terminals, and splicing the differenced data to obtain the first spliced data; Performing a difference on the training signal received by the wireless access point, and splicing the differenced data to obtain second spliced data; The wireless channel matrix is obtained by calculating according to the first spliced data and the second spliced data. 4. The method according to claim 3, characterized in that, the training signals transmitted by the multiple user terminals are expressed as: x ^t0 ,x ^t1 ,…,x ^tK (K>M) Among them, t0, t1, ..., tK is the time of transmission, M is the total number of antennas at the receiving end, and K is the number of training signals transmitted; The training signal received by the wireless access point is expressed as: y ^t1 =HAx ^t1 +Hb+w ^t1 … y ^tK ＝HAx ^tK +Hb+w ^tK Among them, t0, t1,..., tK are the time of transmission, H is the wireless channel matrix between the transmitting and receiving antennas stripped of the radio frequency fingerprint characteristics of the transmitter, A is determined by the radio frequency fingerprint characteristic parameter α _β , b is determined by the radio frequency fingerprint characteristic parameter β _β , b=β ₁ ,β ₂ ,…,β _M } ^T , w is noise; The training signals transmitted by the plurality of user terminals are differentially expressed as: The first splicing data is expressed as: ΔX _t = [Δx _t1 ,Δx _t2 ,…,Δx _tK ] The training signal received by the wireless access point is differentially expressed as: The second splicing data is expressed as: ΔY _t = [Δy _t1 ,Δy _t2 ,…,Δy _tK ] The wireless channel matrix is expressed as: Wherein, ΔY _t represents the second mosaic data, and ΔX _t represents the first mosaic data; The signal vector is expressed as: Wherein, y indicates that the wireless access point receives N signals, y={y ₁ ,y ₂ ,...,y _N } ^T , Indicates the wireless channel matrix; The signal vector after the splicing is expressed as: Z=[z ₁ ,z ₂ ,…] All link signals for all devices described are: in, Indicates the set of all link serial numbers contained in the l-th device, and Z ^j indicates the j-th row vector of Z. 5. The method according to claim 1, wherein if the wireless communication scenario is a downlink communication scenario, the method comprises: receiving multiple training signals transmitted by the wireless access point within multiple time periods at the user end, wherein the time periods are randomly selected by the wireless access point; Performing a difference on the plurality of training signals to obtain the wireless channel matrix; Splitting the wireless channel matrix according to the multiple time periods to obtain a wireless channel matrix corresponding to each time period; receiving all unknown signals transmitted by the wireless access point within the plurality of time periods at the user end; According to the wireless channel matrix corresponding to each time period and the unknown signal received at the user end in each time period, the signal vector of the time period is calculated and generated, and the signal vectors of all time periods are spliced to obtain the spliced signal vector; The spliced signal vectors of all user terminals are acquired, and all link signals of all devices are obtained according to the spliced signal vectors of all user terminals. 6. The method according to claim 5, wherein said performing a difference on said plurality of training signals to obtain said wireless channel matrix comprises: Performing a difference on a plurality of training signals transmitted by the wireless access point to obtain first difference data; performing a difference on a plurality of training signals received by the user end to obtain second difference data; The wireless channel matrix is obtained by calculating according to the first differential data and the second differential data. 7. The method according to claim 6, wherein the multiple training signals transmitted by the wireless access point are expressed as: x ^t1 ,x ^t2 ,…,x ^tK The multiple training signals received by the client are expressed as: Among them, [y _j ] _N×K is the signal received by the user terminal in the jth time period, H _j is the channel state corresponding to the jth time period, A is determined by the radio frequency fingerprint characteristic parameter α _i , b is determined by the radio frequency fingerprint characteristic parameter β _i , b={β ₁ ,β ₂ ,…,β _M } ^T , w is noise; The first differential data is expressed as: ΔX＝[x ^t2 -x ^t1 ,x ^t3 -x ^t2 ,…,x ^tK -x ^t(K-1) ] _M×(K-1) Wherein, M is the number of radio frequency links of the wireless access point, and K is the number of training signals; The second differential data is expressed as: Among them, N is the number of radio frequency links at the user end, K is the number of training signals, and P is the number of time periods; The wireless channel matrix is expressed as: in, The wireless channel matrix after the split is expressed as: H ₁ A,H ₂ A,…,H _P A Among them, P is the number of time periods; The signal vector is expressed as: z _M×K ＝(H _j A) ⁺ y Among them, z _M×K is the signal vector corresponding to the received signal y _N×K , j is the time period of the signal, H _j is the channel state corresponding to the jth time period, and H _j A represents the jth channel state after splitting Wireless channel matrix; The signal vector after the splicing is expressed as: Z=[z ₁ ,z ₂ ,…] All link signals for all devices described are: in, Indicates the set of all link serial numbers contained in the l-th device, and Z ^j indicates the j-th row vector of Z. 8. The method according to claim 1, wherein the radio frequency fingerprint feature recognition model is a deep neural network, and all link signals of all devices are used as training data to train the radio frequency fingerprint feature recognition model and establish a device radio frequency Fingerprint library, including: Obtain pre-processed training signals corresponding to all link signals of all devices; With the goal of minimizing the error between the real device label of the input signal and the estimated device label by the neural network, the deep neural network model is trained to obtain a trained radio frequency fingerprint feature recognition model. The obtained device information and corresponding radio frequency fingerprint features are stored in the device radio frequency fingerprint database. 9. A radio frequency fingerprint identification device for MIMO equipment in a multi-user wireless communication scenario, characterized in that it includes a signal processing module, a model training module, and an identification module, wherein, The signal processing module is configured to receive a training signal and any unknown signal transmitted by the user terminal/wireless access point at the wireless access point/client, and perform a difference on the training signal to generate a wireless channel matrix, according to the Generate signal vectors from any unknown signal and the wireless channel matrix, process all signal vectors of all user equipments, and obtain link signals of all user equipments, wherein, if the wireless access point is a signal transmitting end, the The user end is a signal receiving end, if the user end is a signal transmitting end, the wireless access point is a receiving end, the training signal carries a training sequence, and the arbitrary unknown signal carries communication data; The model training module is used to use all link signals of all devices as training data to train the radio frequency fingerprint feature recognition model, obtain a trained radio frequency fingerprint feature recognition model, and generate a device radio frequency fingerprint database, wherein the device The radio frequency fingerprint library is used to store the device information generated by training and the corresponding radio frequency fingerprint features; The identification module is used to obtain the signal to be identified of the device to be authenticated, and input the signal to be identified into the trained radio frequency fingerprint feature recognition model for identification, obtain the radio frequency fingerprint feature of the device to be authenticated, and convert the signal to be identified to The radio frequency fingerprint feature of the identification device is compared with the radio frequency fingerprint library of the device to obtain the radio frequency fingerprint identification result of the device to be authenticated. 10. The device according to claim 9, wherein the radio frequency fingerprint feature recognition model is a deep neural network, and all link signals of all devices are used as training data to train the radio frequency fingerprint feature recognition model and establish a device radio frequency Fingerprint library, including: Obtain pre-processed training signals corresponding to all link signals of all devices; With the goal of minimizing the error between the real device label of the input signal and the estimated device label by the neural network, the deep neural network model is trained to obtain a trained radio frequency fingerprint feature recognition model. The obtained device information and corresponding radio frequency fingerprint features are stored in the device radio frequency fingerprint database.