CN110300399A - A kind of short distance multi-user concealed communication method and system based on Wi-Fi network interface card - Google Patents

Abstract

The invention discloses a short-distance multi-user concealed communication method and system based on a Wi-Fi network card. The method includes the following sub-steps: Step 1: The Wi-Fi signal source sends out a radio frequency signal, and the receiving end receives the radio frequency signal and calculates according to the radio frequency signal CSI between the Wi-Fi signal source and the receiving end; Step 2: The user end reflects the radio frequency signal sent by the Wi-Fi signal source, and each user uses the network card to simultaneously load its own orthogonally encoded information sequence on the radio frequency signal, after Superimpose to obtain the integrated radio frequency signal; step 3: the receiving end receives the integrated radio frequency signal and calculates the CSI between the Wi-Fi signal source and the receiving end according to the integrated radio frequency signal, and establishes a mapping from the CSI amplitude to the signal level to obtain the integrated signal level, The comprehensive signal level is decoded to obtain the user's information sequence. The invention can transmit the data of multiple users, helps the receiver at the receiving end to separate the important information of multiple users, and ensures the safety of the important information transmitted by the users using the public network connection.

Description

A method and system for short-distance multi-user covert communication based on Wi-Fi network card

technical field

The invention belongs to the technical field of covert communication, and in particular relates to a method and system for short-distance multi-user covert communication based on a Wi-Fi network card.

Background technique

In recent years, with the increasing popularity of wireless networks and mobile smart terminals, short-range wireless data transmission has become an indispensable part of people's daily life. In the short-distance situation, a large amount of user privacy information is transmitted between user devices through Wi-Fi, Bluetooth, NFC and other methods, and there are huge security problems. According to the "Security Report for the First Quarter of 2019" released by Tencent Mobile Security Lab, as of March, the number of public Wi-Fi has reached 984 million, covering public places such as airports, hotels, and large shopping malls, and no risky Wi-Fi has been found. Fi accounts for 58.18%, while risky Wi-Fi accounts for 41.82%, and the number is close to 396 million. Criminals will use risky Wi-Fi to steal users' private information and even threaten the safety of users' property. Bluetooth is not only affected by denial of service attacks, co-channel interference, message modification and resource abuse, but also has the disadvantages of pairing time as long as 6 seconds and only supports point-to-point communication. Using Wi-Fi and Bluetooth to transmit private information in public places, such as identity information, contacts, account passwords, etc., is extremely risky. As the current mainstream technology of near field communication, NFC, due to cost and market constraints, mobile devices with NFC functions only account for 30% of the total devices.

With the enhancement of people's awareness of information protection and the deepening of research, a series of covert communication methods have emerged, such as schemes based on heat reflection and electromagnetic reflection. The advantages of physical connectivity, however, are that these solutions are used to some degree within the context in which they are applicable. Its limitations are obvious, such as the need for power supply, high energy consumption, and the need for external devices to communicate. In the prior art, the impedance when passing through the switch of the network card is different, reflecting the sine wave from the router, and loading its own information on the received signal strength RSSI or transmitting it to the receiving end, although the number of public Wi-Fi of 984 million and the number of network cards Popularity provides the hardware foundation for this, but there are obvious problems correspondingly:

(1) The RSSI signal is greatly affected by the environment, and the RSSI information will change greatly with the change of time and space. The reason is the multi-path transmission of the environment;

(2) The RSSI signal is easy to be captured, and the RSSI information can be captured by any commercial network device in the space, which is not conducive to the concealed transmission of data;

(3) The data transmission rate is limited by the delay of the network card switch, only about 1bps. Therefore, a low-cost, high-precision, and easy-to-use covert communication solution has very important research value.

Contents of the invention

The purpose of the present invention is to provide a short-distance multi-user covert communication method and system based on a Wi-Fi network card to solve the problems existing in the research of covert communication schemes in the prior art.

In order to achieve the above tasks, the present invention adopts the following technical solutions:

A kind of short-distance multi-user covert communication method based on Wi-Fi network card, comprises following sub-steps:

Step 1: The Wi-Fi signal source sends out a radio frequency signal, and the receiving end receives the radio frequency signal and calculates the original channel state information CSI between the Wi-Fi signal source and the receiving end according to the radio frequency signal;

Step 2: the user end reflects the radio frequency signal sent by the Wi-Fi signal source, and the user end includes a plurality of users, and each user uses a network card to simultaneously load respective orthogonally encoded information sequences on the radio frequency signal to obtain a comprehensive radio frequency signal;

Step 3: The receiving end receives the integrated RF signal and calculates the CSI between the Wi-Fi signal source and the receiving end based on the integrated RF signal, establishes the mapping from the CSI amplitude to the signal level to obtain the integrated signal level, and then calculates the integrated signal level Decoding is performed to obtain the information sequence of each user.

Further, step 2 includes the following sub-steps:

Step 2.1: The user end reflects the radio frequency signal sent by the Wi-Fi signal source. The user end includes K users. In order to effectively separate the data from different users at the receiver, we use w _k to encode the transmitted data of user k . The data of each user includes multiple information bits, and the extended information bit d' _k (l) of the kth user's lth bit is expressed as:

d' _k (l)＝μ _k (l)w _k

Among them, k=1,2,3...K and K represents the total number of users at the user end, l=1,2,3...L and L represents the maximum number of information bits, μ _k (l) represents the bipolar sexual position and w _k is expressed as the k-th row vector of the Walsh-Hadmard matrix. The Walsh-Hadmard matrix is an N×N square matrix, and its rows are orthogonal matrices. N is the integral power of 2, and N≥2 and not less than the total number of users;

Step 2.2: Construct according to the number of users and simultaneously load the respective orthogonally coded information sequences on the radio frequency signal to obtain the integrated radio frequency signal as formula Ⅰ:

where β _k represents the CSI power level h ₀ is the original CSI from the Wi-Fi signal source to the receiver at the receiving end, h _k is the CSI from the kth user to the receiver, h _0,k is the CSI from the Wi-Fi signal source to the kth user, p _k is the transmit power of the network card of the kth user, p ₀ is the transmit power at the Wi-Fi signal source, Indicates the attenuation coefficient for adjusting the working state of the network card.

Furthermore, step 3 includes the following sub-steps:

Step 3.1: The receiving end receives the integrated RF signal and calculates the CSI between the Wi-Fi signal source and the receiving end using Formula II according to the integrated RF signal:

in, and z" is the noise term;

Step 3.2: Use formula Ⅲ to establish the mapping from the amplitude of CSI between the Wi-Fi signal source and the receiver to the signal level to obtain the comprehensive signal level d" _k :

d” _k ＝μ _k N Formula Ⅲ

Among them, N=w ^T _N w _N , { } ^T is the transposition of the matrix, μ _k represents the bipolar sequence;

Step 3.3: Decode the integrated signal level d” _k to obtain the bipolar sequence μ _k , and obtain the original information sequence of the kth user according to μ _k , and then obtain the information sequences of all K users.

Further, step 3.3 includes the following sub-steps:

Step 3.3.1: Utilize formula IV to decode the integrated signal level d" _k to obtain the bipolar sequence μ _k :

in, d' _k is the information sequence of the kth user extended by orthogonal coding.

Step 3.3.2: According to the bipolar sequence μ _k , according to the relationship between the bipolar sequence and the information sequence shown in formula V, the information sequence of the kth user is obtained, and then the information sequences of all K users are obtained:

Further, the radio frequency signal is sent to the receiving end by the Wi-Fi signal source based on the data packet, the structure of the data packet is a packet detection sequence, a user identification sequence ID and a payload transmission sequence, and the identification of the kth user The sequence ID _k is expressed as, Wherein, K ₀ ≥ K+1.

A short-distance multi-user covert communication system based on a Wi-Fi network card, including a Wi-Fi signal source, a user terminal, and a receiving terminal;

Including Wi-Fi signal source, client and receiver;

The Wi-Fi signal source is used to send a radio frequency signal to the receiving end;

The user end includes a plurality of users, and each user uses a network card to simultaneously load their own orthogonally coded information sequences on the radio frequency signal, and reflect to the receiving end to superimpose with the original radio frequency signal to obtain a comprehensive radio frequency signal;

The receiving end receives the radio frequency signal sent by the Wi-Fi signal source and the integrated radio frequency signal reflected by the user end, calculates the CSI between the Wi-Fi signal source and the receiving end, and establishes the CSI amplitude to signal level according to the amplitude change of the CSI Flat mapping to obtain the integrated signal level, and then decode the integrated signal level to obtain the information sequence of each user.

Further, the user end reflects the radio frequency signal sent by the Wi-Fi signal source, and the user end includes K users. In order to effectively separate the data from different users at the receiver, we use w _k to encode the sent data of user k . The data of each user includes multiple information bits, and the extended information bit d' _k (l) of the kth user's lth bit is expressed as:

d' _k (l)＝μ _k (l)w _k

Among them, k=1,2,3...K and K represents the total number of users at the user end, l=1,2,3...L and L represents the maximum number of information bits, μ _k (l) represents the bipolar sexual position and w _k is expressed as the k-th row vector of the Walsh-Hadmard matrix. The Walsh-Hadmard matrix is an N×N square matrix, and its rows are orthogonal matrices. N is the integral power of 2, and N≥2 and not less than the total number of users;

Then, according to the number of users, construct and simultaneously load the respective orthogonally coded information sequences on the radio frequency signal, and obtain the integrated radio frequency signal as formula Ⅰ:

where β _k represents the CSI power level h ₀ is the CSI from the Wi-Fi signal source to the receiver at the receiving end, h _k is the CSI from the kth user to the receiver, h _0,k is the CSI from the Wi-Fi signal source to the kth user, and p _k is The transmit power of the network card of the kth user, p ₀ is the transmit power at the Wi-Fi signal source, Indicates the attenuation coefficient for adjusting the working state of the network card.

Further, the receiving end receives the integrated radio frequency signal and uses formula II to calculate the CSI between the Wi-Fi signal source and the receiving end according to the integrated radio frequency signal:

in, and z" is the noise term;

The receiving end uses formula III to establish a mapping from CSI to signal level to obtain the integrated signal level d” _k :

d” _k ＝μ _k N Formula Ⅲ

Among them, N=w ^T _N w _N , { } ^T is the transposition of the matrix, μ _k represents the bipolar sequence;

The receiving end decodes the integrated signal level d" _k to obtain a bipolar sequence μ _k , and obtains the information sequence of the kth user according to μ _k , and then obtains the information sequences of all K users.

Further, the receiving end decodes the integrated signal level d” _k by formula IV to obtain the bipolar sequence μ _k :

And according to the bipolar sequence μ _k , according to the relationship between the bipolar sequence and the information sequence shown in formula V, the information sequence of the kth user is obtained, and then the information sequences of all K users are obtained:

Further, the radio frequency signal is sent to the receiving end by the Wi-Fi signal source based on the data packet, the structure of the data packet is a packet detection sequence, a user identification sequence ID and a payload transmission sequence, and the identification of the kth user The sequence ID _k is expressed as, Wherein, K ₀ is greater than the total number of users K, that is, K ₀ ≥ K+1.

Compared with the prior art, the present invention has the following beneficial effects:

(1) In recent years, the CSI of Wi-Fi devices has been widely used in the field of environmental sensing. It is a more fine-grained channel characteristic than RSSI in Wi-Fi devices. CSI can capture fast fading and slow fading information at the same time, so as to reflect frequency selective channel characteristics and multipath channel effects. Compared with the RSSI-based sensing method, the CSI-based sensing method of the present invention has better sensing performance.

(2) To help extract important information of users from Wi-Fi CSI, a theoretical model is given to show the relationship among Wi-Fi CSI, reflected RF signal and important information.

(3) In order to solve the problem of multi-user concealed communication by Wi-Fi network card, the important information of multiple users is superimposed on the receiving end and cannot be separated. Orthogonal coding and decoding technology helps the receiving end to separate important information of multiple users.

(4) In order to theoretically simulate the relationship between the captured CSI and the important information transmitted by the user, firstly the covert channel communication of the signal transmission process through the wireless network card is analyzed, and then the CSI source (signal helper) and the wireless network at the receiving end are analyzed Finally, a theoretical model is given to analyze the relationship between CSI and transmitted data.

(5) In order to facilitate multi-user data decoding, a scheme that can determine the number and identity of users is proposed.

Description of drawings

Figure 1 is a change diagram of CSI at the receiver when the user configures the Atheros 9580 network card;

Fig. 2 is a structural diagram of user transmission data packets;

Fig. 3 is the received different frequency subcarrier amplitude value figure;

Fig. 4 is the signal-to-noise ratio diagram of different frequency sub-carriers processed through direct current removal;

Fig. 5 is a CSI amplitude diagram for different numbers of users;

Figure 6 is a multi-user scene diagram;

Figure 7 is the experimental topology;

Fig. 8 is a user quantity identification performance diagram;

Fig. 9 is a user ID identification performance diagram;

Fig. 10 is a bit error rate performance diagram with different numbers of users;

Fig. 11 is a relationship diagram of the bit error rate corresponding to the system transmission rate and the switch interval of the network card;

Figure 12 shows the impact of the number of Wi-Fi access points in the environment on system performance.

Detailed ways

Example 1

In this embodiment, a method for concealed communication of short-distance multi-users based on a Wi-Fi network card is disclosed, including the following sub-steps:

Step 1: The Wi-Fi signal source sends out a radio frequency signal, and the receiving end receives the radio frequency signal and calculates the original CSI between the Wi-Fi signal source and the receiving end according to the radio frequency signal;

Step 2: The user end reflects the radio frequency signal sent by the Wi-Fi signal source. The user end includes multiple users. Each user uses a network card to simultaneously load their respective orthogonally encoded information sequences on the radio frequency signal, and obtain a comprehensive radio frequency signal through superposition. Signal;

Step 3: The receiving end receives the integrated RF signal and calculates the CSI between the Wi-Fi signal source and the receiving end based on the integrated RF signal, and establishes a mapping from the CSI amplitude to the signal level according to the amplitude change of the CSI to obtain the integrated signal level , and then decode the integrated signal level to obtain the information sequences of all users.

The invention adjusts the impedance of the mobile user's Wi-Fi network card through the pre-installed software for controlling the impedance of the network card, so that the user's network card can reflect the environmental radio frequency signal (RF) signal in the surrounding environment. The information within the user end can then be carried by the reflected RF signal and transmitted to the receiver at the receiving end, using orthogonal encoding and decoding to help multiple users decode the data at the receiving end.

First, explain how the Wi-Fi network card loads the information sequence:

As shown in Figure 6, the incoming signal from the Wi-Fi source will be transmitted directly to the receiver. In addition, if the impedance of the user's network card changes from one state to another, the auxiliary signal may be reflected by the user's network card and then transmitted to the receiving end. This paper mainly studies the users of static network card configuration. In static cards, when the Wi-Fi card is continuously enabled and disabled, the envelope of the reflected RF signal resembles a square wave of different power levels. As shown in Figure 1, the NIC AR9850 is a static NIC. The signal received by one transmitting antenna of the signal source at user k can be expressed as:

where h _0,k is the CSI from the Wi-Fi signal source to user k, P ₀ and x ₀ are the transmit power and transmit signal at the Wi-Fi signal source, respectively, and Z _k is Gaussian white noise at user k.

It is assumed that the user has pre-installed the software for controlling the impedance of the network card, and the software for controlling the impedance of the network card can control the working state of the user's network card, making it in the on/off state, so that it can reflect the RF signal received by the user. Different working states of the network card will result in different amplitudes of reflected RF signals, as shown in Figure 1. More specifically, the received signal x' _k will be reflected to user k:

Among them, α0 and α1 are the attenuation factors caused by adjusting the working state of the network card.

Using this feature, the user's sensitive information can be modulated into a reflected radio frequency signal, which can then be decoded at the receiving end. More specifically, the software for controlling the impedance of the network card pre-installed in the mobile user controls the working state of the user's network card to be on or off according to the user's sensitive information. Therefore, the reflected signal at user k can be rewritten as:

where d _k is the sensitive information bit in user k. According to (2) and (3), the software controlling the impedance of the network card of user k will force the user's network card if the sensitive information transmitted is '0' and the software controlling the impedance of the network card will force the user's network card if the bit transmitted information is '1'". In this way, the user's sensitive information can be modulated into a radio frequency signal and passed to the receiver at the receiving end.

If K users reflect Wi-Fi signals at the same time, the receiver at the receiving end will receive signals from all K users in addition to receiving signals from the Wi-Fi signal source. Therefore, the signal received by the receiver can be expressed as

where h ₀ is the CSI from the Wi-Fi signal source to the receiver at the receiving end, h _k is the CSI from user k to the receiver, P _k is the transmit power of the network card of user k, and z is the white Gaussian noise at the receiver.

It is mainly aimed at multiple mobile devices configured with static network cards, and a covert channel communication coding scheme based on Wi-Fi network card backscattering. In order to allow multiple users to broadcast data simultaneously, the payload data before transmission is encoded using orthogonal coding techniques. In this way, different users can be distinguished by different orthogonal codes. Therefore, at the receiving end, the corresponding orthogonal decoding techniques can be used to separate the data from different users.

Specifically, step 2 includes the following sub-steps:

Step 2.1: The user end reflects the radio frequency signal sent by the Wi-Fi signal source. The user end includes K users. In order to effectively separate the data from different users at the receiver, we use w _k to encode the transmitted data of user k . The data of each user includes multiple information bits, and the extended information bit d' _k (l) of the kth user's lth bit is expressed as:

d' _k (l)＝μ _k (l)w _k

Among them, k=1,2,3...K and K represents the total number of users at the user end, l=1,2,3...L and L represents the maximum number of information bits, μ _k (l) represents the bipolar sexual position and w _k is expressed as the k-th row vector of the Walsh-Hadmard matrix. The Walsh-Hadmard matrix is an N×N square matrix, and its rows are orthogonal matrices. N is the integral power of 2, and N≥2 and not less than the total number of users (N ≥K);

For ease of explanation, each user's transmitted payload data is encoded using the rows of the Walsh-Hadmard matrix. The Walsh-Hadmard matrix is a square matrix whose rows are orthogonal vectors. Represents a Walsh-Hadmard matrix where W _N is N×N. then it is

and

Where N is the integral power of 2, N>2, and not less than the total number of users (ie N≥K), is the Kronecker (Kronecker) product. Indicates that w _i is the ith row vector of the Walsh-Hadmard matrix H _N . According to the definition of Walsh-Hadmard matrix,

w _i w _j ^T =0 (7)

and

w _i w _i ^T = N (8)

In order to effectively separate the data from different users at the receiving end, w _N is used to encode the transmitted data of user k.

Step 2.2: Construct and load the orthogonally coded information sequences to the radio frequency signal according to the number of users, and obtain the integrated radio frequency signal through superposition as formula Ⅰ:

where β _k represents the CSI power level β _k = α _ηk C _k , h ₀ is the original CSI from the Wi-Fi signal source to the receiver at the receiving end, h _k is the CSI from the kth user to the receiver, h _0,k is the CSI from the Wi-Fi signal source to the kth user, p _k is the transmission power of the network card of the kth user, p ₀ is the transmission power at the Wi-Fi signal source, and α _ηk represents the attenuation coefficient for adjusting the working state of the network card.

In the case of one user, orthogonal coding is not required, so W ₁ =1 is assumed in the case of one user. Note that other orthogonal coding schemes can also be used to support the signaling of multiple users. For ease of interpretation, the transmitted information is encoded using Walsh-Hadmard sequences.

Specifically, step 3 includes the following sub-steps:

Step 3.1: The receiving end receives the integrated RF signal and calculates the CSI between the Wi-Fi signal source and the receiving end using Formula II according to the integrated RF signal:

in, and z" is the noise term;

Step 3.2: The change of the integrated radio frequency signal will cause the amplitude change of CSI, and use formula III to establish the mapping from integrated CSI to signal level to obtain the integrated signal level d" _k :

d” _k ＝μ _k N Formula Ⅲ

Among them, N=w ^T _N w _N , { } ^T is the transposition of the matrix, μ _k represents the bipolar sequence;

Step 3.3: Decode the integrated signal level d” _k to obtain the bipolar sequence μ _k , and obtain the information sequence of the kth user according to μ _k , and then obtain the information sequences of all K users.

Preferably, step 3.3 includes the following sub-steps:

The encoded information bit transmitted in d' at user k has two possible values: '1' and '-1'. Label the received signal level of each coded information symbol "1" and "-1" as 1 and -1, respectively. Suppose there are K users in total, and all users send data at the same time, K (1≤k≤K) users send coded information symbol "1", and the remaining K-k users send coded information symbol "-1". The overall received signal level associated with these transmitted symbols is then defined as k*1+(K-k)*(-1)=K-2k.

Considering the structure and design of the packet, in superimposing the preamble sequence of CSI and transmitter identification code ID, the number of CSI levels is equal to the number of users plus 1, K+1, and the number of signal levels is the same as the CSI level. More specifically, the signal levels involved are: K,K-2,...,K-2(K-1),...,-K where K-2(K-1) is the transmitter The signal level corresponding to the Kth bit in the identification code ID.

The signal level is determined for each symbol of the received load from the signal level obtained in the preamble sequence. More specifically, the mean of the CSI amplitude samples in each symbol period is calculated, then the calculated value is compared with the CSI level associated with the serial number sequence, and finally the calculated value is updated to the most recent CSI level. Then, the calculated CSI levels are mapped to corresponding signal levels. The calculated signal level corresponding to one information bit of a user not performing orthogonal coding is Ss.

Step 3.3.1: Use formula IV to decode the integrated signal level d” _k to obtain the bipolar sequence μ _k :

in Among them, d' _k is the information sequence of the kth user extended by orthogonal coding.

Step 3.3.2: According to the bipolar sequence μ _k , according to the relationship between the bipolar sequence and the information sequence shown in formula V, the information sequence of the kth user is obtained, and then the information sequences of all K users are obtained:

Specifically, in order to maintain the consistency of the description, "1" and "-1" are used to represent "1" and "0" in the preamble sequence respectively. For packet detection, use "-1,1,1,1,-1" in the frame structure as the packet detection code. The data packet is then detected at the receiving end according to the encoding structure of the data packet. Other codes are also available to facilitate packet inspection.

For wireless carrier transmission of CSI, from the beginning of the CSI trace if the time interval between the sudden change of the CSI level is equal to three symbol durations, the sudden change at the beginning causes the CSI level to increase, and the sudden change at the end causes the CSI level to decrease, and then Three consecutive symbols "111" can be detected. The first and last symbol "-1" can also be determined, because the sudden change in the CSI level is caused by the difference of the transmitted symbols. Once the symbol "-1111-1" is detected, it means that a new compression is transmitted.

In order to facilitate data decoding at the receiving end, the user number and identity after data packet detection must be determined first. Therefore, a user identification code is designed in the frame structure to solve this problem. More specifically, the serial ID is designed to identify different users, so as to facilitate data decoding at the receiving end. Specifically, suppose 1 _1×k and -1 _1×k are 1×k vectors with all elements equal to 1 and -1 respectively.

The radio frequency signal is sent to the receiving end by the Wi-Fi signal source relying on the data packet, the structure of the data packet is a packet detection sequence, user identification sequence ID and payload transmission sequence, the identification sequence ID _k of the kth user Expressed as, Wherein, K ₀ is greater than the total number of users K, that is, K ₀ ≥ K+1.

Specifically, the method of determining the number and identity of users

However, to decode the receiver's payload information, the number of users and the identity of the users first need to be known.

Once the symbol "-1111-1" for packet detection is found, the position of the K+1 bit ID related to the user identity can be determined according to the packet structure. These K+1 symbols will be used to determine the number and identity of users.

a) Observation: Since the comprehensive levels of CSI of multiple users will be superimposed linearly, this can also be proved in the measured data after LPF, as shown in Fig. 5. If two users send the reflected signal at the same time, the CSI level obtained by the superimposed identity code will be like a staircase, and if all users send data at the same time, then at the end of each bit duration, the CSI level will occur sudden change.

b) Determination of the number of users: construct a matrix T whose rows are the identity codes of all users,

The k-th row vector is equal to the identification code of the k-th user. To determine the number of users, the number of abrupt changes in the CSI level at the endpoint boundaries of each bit is detected by setting a CSI level change threshold. The number of users sending data at the same time is equal to the number of mutations in the CSI level within the symbol duration, corresponding to the user identity code.

c) User identity determination: According to the experimental observation in Figure 5, the structure of the matrix T is that if the CSI level at the end boundary of the kth symbol of the user identity code position changes suddenly, the user sends data. For example, if only user 1 and user 3 participate in backscatter transmission among all K users, then the corresponding sub-matrix in matrix T can be expressed as:

The CSI level superimposed at the end of the first digit and the third digit of the subscriber identification code will change the CSI level.

Example 2

A short-distance multi-user covert communication system based on a Wi-Fi network card, including a Wi-Fi signal source, a user terminal, and a receiving terminal;

The Wi-Fi signal source is used to send a radio frequency signal to the receiving end;

The user end includes K users, and each user uses a network card to simultaneously load their respective information sequences to the radio frequency signal, and each user uses a network card to simultaneously load their respective orthogonally coded information sequences to the radio frequency signal, and obtain a comprehensive radio frequency signal through superposition. The signal synthesis radio frequency signal is reflected to the receiving end;

The receiving end receives the radio frequency signal sent by the Wi-Fi signal source and the integrated radio frequency signal reflected by the user end, respectively calculates the CSI between the Wi-Fi signal source and the receiving end, and establishes the CSI amplitude to The signal level is mapped to obtain the integrated signal level, and then the integrated signal level is decoded to obtain the information sequences of K users.

Specifically, the user end reflects the radio frequency signal sent by the Wi-Fi signal source, and the user end includes K users. In order to effectively separate the data from different users at the receiver, we use w _k to encode the sent data of user k . The data of each user includes multiple information bits, and the extended information bit d' _k (l) of the kth user's lth bit is expressed as:

d' _k (l)＝μ _k (l)w _k

Among them, k=1,2,3...K and K represents the total number of users at the user end, l=1,2,3...L and L represents the maximum number of information bits, μ _k (l) represents the bipolar sexual position and w _k is expressed as the k-th row vector of the Walsh-Hadmard matrix. The Walsh-Hadmard matrix is an N×N square matrix, and its rows are orthogonal matrices. N is the integral power of 2, and N≥2 and not less than the total number of users (N ≥K); According to the number of users, the RF signal is constructed and loaded with its orthogonally coded information sequence, and the integrated RF signal is obtained through superposition, such as formula Ⅰ:

where β _k represents the CSI power level h ₀ is the original CSI from the Wi-Fi signal source to the receiver at the receiving end, h _k is the CSI from the kth user to the receiver, h _0,k is the CSI from the Wi-Fi signal source to the kth user, p _k is the transmit power of the network card of the kth user, p ₀ is the transmit power at the Wi-Fi signal source, Indicates the attenuation coefficient for adjusting the working state of the network card.

Specifically, the receiving end receives the integrated radio frequency signal and uses Formula II to calculate the CSI between the Wi-Fi signal source and the receiving end according to the integrated radio frequency signal:

in, and z" is the noise term;

The receiving end uses formula III to establish a mapping from the integrated CSI to the signal level to obtain the integrated signal level d” _k :

d” _k ＝μ _k N Formula Ⅲ

Among them, N=w ^T _N w _N , { } ^T is the transposition of the matrix, μ _k represents the bipolar sequence;

The receiving end decodes the integrated signal level d" _k to obtain a bipolar sequence μ _k , and obtains the information sequence of the kth user according to μ _k , and then obtains the information sequences of all K users.

Preferably, the receiving end decodes the integrated signal level d” _k by formula IV to obtain the bipolar sequence μ _k :

in Among them, d' _k is the information sequence of the kth user extended by orthogonal coding.

And according to the bipolar sequence μ _k , according to the relationship between the bipolar sequence and the information sequence shown in formula V, the information sequence of the kth user is obtained, and then the information sequences of all K users are obtained:

Specifically, the radio frequency signal is sent to the receiving end by the Wi-Fi signal source based on the data packet, the structure of the data packet is a packet detection sequence, a user identification sequence ID and a payload transmission sequence, and the identification of the kth user The sequence ID _k is expressed as, Wherein, K ₀ is greater than the total number of users K, that is, K ₀ ≥ K+1.

Example 3

experiment procedure

First, a multi-user covert channel backscatter communication system is constructed in the 5GHz frequency band through commercial Wi-Fi equipment. Then, we collect data through the system to evaluate system performance. The experimental data collection is carried out in an ordinary laboratory, and there are multiple Wi-Fi access points (APs) working in the same frequency band in the surrounding rooms or corridors, which always cause interference to the system.

Evaluation and configuration

The experimental evaluation is designed according to the user scenarios shown in Fig. 6. Specifically, we used a commercial miniature desktop computer equipped with an Intel 5300 Wi-Fi network card and a 5dBi omnidirectional antenna as a signal assistant and a receiver, respectively. And a commercial miniature desktop computer equipped with Atheros AR9580Wi-Fi network card and 5dBi omnidirectional antenna for users. The experimental evaluation is carried out in an office environment, whose topology is shown in Fig. 7. Signal Helper generates a 5GHz helper signal by transmitting User Datagram Protocol (UDP) packets. The packet transmission rate is 1000 packets/second. The receiver is running a CSI tool to capture a CSI trace for the communication process. The user is positioned between the signal assistant and the receiver. Among them, d2 and d1 are the vertical distance from the signal assistant to the user and the vertical distance from the receiver to the user, respectively. The distance between the signal aid and the receiver is 1.2m. The switch interval of the network card is 0.25s by default.

CSI at the receiver and received signal-to-noise ratio (SNR)

Figure 3 shows the CSI trajectories for multiple subcarriers, and the corresponding SNR of the receiver after LPF. The encoded sensitive data leaked from the user is related to the square wave form of the CSI trace in Figure 3. From the figure, it can be seen that not all Wi-Fi subcarriers carry sensitive data from users, which is due to the frequency selectivity of Wi-Fi channels. Therefore, in order to decode sensitive data from users, subcarrier selection is required.

Figure 4 shows the signal-to-noise ratios of the CSI trajectories of each subcarrier at two different moments (corresponding to different packet numbers) after the DC component is removed. It can be seen from the figure that the signal-to-noise ratio is frequency-selective and time-selective. It is not to say that the higher the SNR on a subcarrier, the easier it is to decode sensitive data of users on that subcarrier. Therefore, in order to facilitate the decoding of sensitive data, the preamble characteristic is used to select subcarriers for decoding.

1) User number detection: The performance of user number detection is closely related to the sensitive data decoding of the receiver, especially for multi-user scenarios. Figure 8 shows the performance of user number detection performance when 1, 2, and 3 users participate in the system at the same time, and the d1 distance varies from 10cm to 40cm. It can be seen from the figure that the detection performance decreases with the increase of d1. The reason is that the farther the distance from the user to the receiver is, the weaker the backscatter signal received by the receiver is, thus reducing the performance of user number detection. We can also see from the figure that the lower the detection performance, the more users there are. When d1=10cm, its detection performance can almost reach 100% for different numbers of users.

2) User ID detection performance: Figure 9 illustrates the performance of user identification performance when 1, 2, and 3 users participate in the system at the same time, and the d1 distance is 10-40cm. It can be seen from the figure that as d1 increases, the performance of user ID detection decreases, which is similar to the trend in the figure. We can also see that the detection performance is lower and more users are acquired. When d1=10cm, the detection performance of different number of people to be tested can almost reach 100%, which is similar to that in the figure.

3) System bit error rate (BER): Figure 10 describes the decoding performance of the system when 1, 2, and 3 users participate in the system at the same time, and the d1 distance varies from 10cm to 40cm. It can be seen from the figure that the BER increases with the increase of d1. The reason is that the farther the distance from the user to the receiver is, the weaker the backscattering signal received by the receiver is, thus reducing the decoding performance. It can also be seen from the figure that as the number of users increases, the bit error rate also increases. The bit error rate of a user can almost reach 0 when d1<20cm. For 1, 2 or 3 users involved in the system, the bit error at d1 = 10 cm is less than 1%.

System data rate and NIC switching interval

Figure 11 illustrates the effect of the NIC switch interval on the system data rate and bit error rate when one user is involved. Omni-directional antennas with an antenna gain of 5dBi are respectively arranged at the signal assistant, the user and the receiving end, and d1=10cm. It can be seen from the figure that both the data rate and the bit error rate increase as the network card switching interval decreases. By changing the switching frequency of the network card, the data transmission rate can reach 100 bits, and the bit error rate is very low; therefore, the system can also be applied to some near-field communication scenarios that require lower transmission rates.

Effect of Interference on Receiver Signal-to-Noise Ratio

Figure 12 shows an example, illustrating the impact of interference from the surrounding environment on the average SNR of the receiver in the 1500th data packet when d1 = 20cm, after the DC in the CSI is removed. These data are collected from many experimental measurements. In each experiment, we use Wi-Fi diagnostic software to observe the number of Wi-Fi access points (APs) at 5GHz frequency. It can be seen from the figure that the signal-to-noise ratio decreases as the number of Wi-Fi APs increases. When the number of 5Ghz Wi-Fi APs increases, the interference of these Wi-Fi APs will seriously interfere with the CSI traces, so that the reflected signals carried by the CSI traces at the receiver make it more difficult to decode sensitive data at the receiver. On the other hand, higher data rates are seen in Figure 11 when there is less interference from the surrounding environment.

Claims (10)

1. a kind of short-distance multi-user covert communication method based on Wi-Fi network card, it is characterized in that, comprises following sub-steps: Step 1: The Wi-Fi signal source sends out a radio frequency signal, and the receiving end receives the radio frequency signal and calculates the original channel state information CSI between the Wi-Fi signal source and the receiving end according to the radio frequency signal; Step 2: the user end reflects the radio frequency signal sent by the Wi-Fi signal source, and the user end includes a plurality of users, and each user uses a network card to simultaneously load respective orthogonally encoded information sequences on the radio frequency signal to obtain a comprehensive radio frequency signal; Step 3: The receiving end receives the integrated RF signal and calculates the CSI between the Wi-Fi signal source and the receiving end based on the integrated RF signal, establishes the mapping from the CSI amplitude to the signal level to obtain the integrated signal level, and then calculates the integrated signal level Decoding is performed to obtain the information sequence of each user. 2. the short-distance multi-user covert communication method based on Wi-Fi network card as claimed in claim 1, is characterized in that, step 2 comprises following sub-steps: Step 2.1: The user end reflects the radio frequency signal sent by the Wi-Fi signal source. The user end includes K users. In order to effectively separate the data from different users at the receiver, we use w _k to encode the transmitted data of user k . The data of each user includes multiple information bits, and the extended information bit d' _k (l) of the kth user's lth bit is expressed as: d' _k (l)＝μ _k (l)w _k Among them, k=1,2,3...K and K represents the total number of users at the user end, l=1,2,3...L and L represents the maximum number of information bits, μ _k (l) represents the bipolar sexual position and w _k is expressed as the k-th row vector of the Walsh-Hadmard matrix. The Walsh-Hadmard matrix is an N×N square matrix, and its rows are orthogonal matrices. N is the integral power of 2, and N≥2 and not less than the total number of users; Step 2.2: Construct according to the number of users and simultaneously load the respective orthogonally coded information sequences on the radio frequency signal to obtain the integrated radio frequency signal as formula Ⅰ: where β _k represents the CSI power level h ₀ is the original CSI from the Wi-Fi signal source to the receiver at the receiving end, h _k is the CSI from the kth user to the receiver, h _0,k is the CSI from the Wi-Fi signal source to the kth user, p _k is the transmit power of the network card of the kth user, p ₀ is the transmit power at the Wi-Fi signal source, Indicates the attenuation coefficient for adjusting the working state of the network card. 3. the short-distance multi-user covert communication method based on Wi-Fi network card as claimed in claim 2, is characterized in that, step 3 comprises following sub-steps: Step 3.1: The receiving end receives the integrated RF signal and calculates the CSI between the Wi-Fi signal source and the receiving end using Formula II according to the integrated RF signal: in, and z" is the noise term; Step 3.2: Use formula Ⅲ to establish the mapping from the amplitude of CSI between the Wi-Fi signal source and the receiver to the signal level to obtain the comprehensive signal level d" _k : d” _k ＝μ _k N Formula Ⅲ Among them, N=w ^T _N w _N , { } ^T is the transposition of the matrix, μ _k represents the bipolar sequence; Step 3.3: Decode the integrated signal level d” _k to obtain the bipolar sequence μ _k , and obtain the original information sequence of the kth user according to μ _k , and then obtain the information sequences of all K users. 4. the short-distance multi-user covert communication method based on Wi-Fi network card as claimed in claim 3, is characterized in that, step 3.3 comprises following sub-steps: Step 3.3.1: Use formula IV to decode the integrated signal level d” _k to obtain the bipolar sequence μ _k : in, d' _k is the information sequence of the kth user extended by orthogonal coding. Step 3.3.2: According to the bipolar sequence μ _k , according to the relationship between the bipolar sequence and the information sequence shown in formula V, the information sequence of the kth user is obtained, and then the information sequences of all K users are obtained: 5. the concealed communication method of short distance multi-user based on Wi-Fi network card described in claim 1, it is characterized in that, described radio frequency signal is sent to receiving end by Wi-Fi signal source by relying on data packet, the data packet The structure is packet detection sequence, user identification sequence ID and payload transmission sequence, and the identification sequence ID _k of the kth user is expressed as, Wherein, K ₀ ≥ K+1. 6. A short-distance multi-user covert communication system based on a Wi-Fi network card, characterized in that it includes a Wi-Fi signal source, a user end, and a receiving end; The Wi-Fi signal source is used to send a radio frequency signal to the receiving end; The user end includes a plurality of users, and each user uses a network card to simultaneously load their own orthogonally coded information sequences on the radio frequency signal, and reflect to the receiving end to superimpose with the original radio frequency signal to obtain a comprehensive radio frequency signal; The receiving end receives the radio frequency signal sent by the Wi-Fi signal source and the integrated radio frequency signal reflected by the user end, calculates the CSI between the Wi-Fi signal source and the receiving end, and establishes the CSI amplitude to the signal level according to the amplitude change of the CSI. Flat mapping to obtain the integrated signal level, and then decode the integrated signal level to obtain the information sequence of each user. 7. The short-distance multi-user covert communication system based on Wi-Fi network card as claimed in claim 6, is characterized in that, the user terminal reflects the radio frequency signal that Wi-Fi signal source sends, and described user terminal comprises K users, for To effectively separate the data from different users at the receiver, we utilize w _k to encode the transmitted data of user k. The data of each user includes multiple information bits, and the extended information bit d' _k (l) of the kth user's lth bit is expressed as: d' _k (l)＝μ _k (l)w _k Among them, k=1,2,3...K and K represents the total number of users at the user end, l=1,2,3...L and L represents the maximum number of information bits, μ _k (l) represents the bipolar sexual position and w _k is expressed as the k-th row vector of the Walsh-Hadmard matrix. The Walsh-Hadmard matrix is an N×N square matrix, and its rows are orthogonal matrices. N is the integral power of 2, and N≥2 and not less than the total number of users; Then, according to the number of users, construct and simultaneously load the respective orthogonally coded information sequences on the radio frequency signal, and obtain the integrated radio frequency signal as formula Ⅰ: where β _k represents the CSI power level h ₀ is the CSI from the Wi-Fi signal source to the receiver at the receiving end, h _k is the CSI from the kth user to the receiver, h _0,k is the CSI from the Wi-Fi signal source to the kth user, and p _k is The transmit power of the network card of the kth user, p ₀ is the transmit power at the Wi-Fi signal source, Indicates the attenuation coefficient for adjusting the working state of the network card. 8. the short-distance multi-user covert communication system based on Wi-Fi network card as claimed in claim 7, is characterized in that, The receiving end receives the integrated radio frequency signal and uses formula II to calculate the CSI between the Wi-Fi signal source and the receiving end according to the integrated radio frequency signal: in, and z" is the noise term; The receiving end uses formula III to establish a mapping from CSI to signal level to obtain the integrated signal level d” _k : d” _k ＝μ _k N Formula Ⅲ Among them, N=w ^T _N w _N , { } ^T is the transposition of the matrix, μ _k represents the bipolar sequence; The receiving end decodes the integrated signal level d" _k to obtain a bipolar sequence μ _k , and obtains the information sequence of the kth user according to μ _k , and then obtains the information sequences of all K users. 9. The short-distance multi-user concealed communication method based on Wi-Fi network card as claimed in claim 8, is characterized in that, described receiving end uses formula IV to carry out decoding to integrated signal level d " _k , obtains bipolar Sex sequence μ _k : And according to the bipolar sequence μ _k , according to the relationship between the bipolar sequence and the information sequence shown in formula V, the information sequence of the kth user is obtained, and then the information sequences of all K users are obtained: 10. The method for concealed communication of short-distance multi-user based on Wi-Fi network card according to claim 6, characterized in that, the radio frequency signal is sent to the receiving end by the Wi-Fi signal source on the basis of the data packet, and the data packet The structure is packet detection sequence, user identification sequence ID and payload transmission sequence, and the identification sequence ID _k of the kth user is expressed as, Wherein, K ₀ is greater than the total number of users K, that is, K ₀ ≥ K+1.