CN110225579A - A kind of cooperation interference safe transmission method of physical layer based on wireless energy supply - Google Patents

Abstract

The invention discloses a method for secure transmission of the physical layer of cooperative interference based on wireless power supply. The cooperative interference node uses the expected collection power to send the interference signal, and studies the safe transmission performance of the system in five situations; starting from the known channel CSI, The secrecy performance of different interference cooperation schemes under the ideal situation with known CSI of all channels, the unknown CSI situation of eavesdroppers, the known cooperative channel gain ranking and the limited feedback rate channel CSI situation are respectively studied, and the connection interruption in various situations is derived The closed-form solution of probability and confidentiality interruption probability, and the expression under high SNR approximation; it can be concluded that there is a certain compromise between the connection interruption probability and the confidentiality interruption probability of the system, which can be selected and designed according to different system requirements The system parameters achieve a security-secrecy compromise.

Description

A method for secure transmission of cooperative jamming physical layer based on wireless power supply

technical field

The present invention relates to the technical field of network security transmission, in particular to a wireless power supply-based cooperative interference physical layer security transmission method.

Background technique

The transmission of jamming signals needs to consume energy, although friendly cooperative jamming nodes (Friendly

Jammer, FJ) can assist the secure transmission of legitimate user information, but at the same time consumes its own energy. Due to the selfishness and independence of nodes, interference nodes should at least not harm their own benefits and even receive additional compensation and rewards while helping legal transmission. Therefore, the legal user’s transmitter needs to provide the extra energy required by the FJ node to transmit interference, and the energy required by the cooperative node can be compensated by wireless energy supply, which can encourage the cooperative node to participate in cooperative security transmission, which is more in line with the actual cooperative node participation Collaboration situation.

Literature Jiang and Chen studied the power station to activate legitimate users to transmit information through wireless energy supply, and PB (Power Beacon) acts as the power source of the transmitter. Jiang also studied the physical layer security performance of the system in which the single-antenna PB supplies energy to the multi-antenna transmitter through the wireless function when there is a single-antenna and multi-antenna eavesdropper Eve. It is worth noting that this scheme uses the transmitter to send AN to Very high system security. Huang adopts a model similar to the literature, the difference is that two different situations of collusion and non-collusion of multiple eavesdroppers are considered, assuming that the outdated channel CSI (Outdated CSI, OCSI) is known, and the transmitter adopts the maximum ratio transmission (MRT ) and transmit antenna selection (TAS) two methods of security outage probability (SOP) and average secrecy rate.

Contents of the invention

The purpose of the present invention is to provide a method for secure physical layer transmission of cooperative interference based on wireless energy supply. The cooperative interference node uses the expected collection power to send interference signals, which enhances the security of the system without consuming its own energy.

The technical scheme adopted in the present invention is:

A method for secure transmission of cooperative jamming physical layer based on wireless power supply, comprising the following steps:

A: Establish a limited-rate feedback cooperative interference eavesdropping transmission model considering MISO, including multi-antenna source node S, single-antenna legal user D, wireless-powered cooperative interference node J and single-antenna eavesdropping node E;

Cooperating nodes can be multiple or multi-antenna. Here, for the convenience of calculation and analysis, a single-antenna cooperating node can be regarded as a special case of selecting an optimal coordinating node from multiple coordinating nodes;

Assume that S has the number of antennas N _S , the number of antennas configured by J is N _J , and the rest of the nodes use a single antenna; assuming that the channel is a quasi-static Rayleigh channel, that is, the CSI of the channel in each transmission block remains unchanged, and the channel CSI between different transmission blocks Channel CSI varies independently;

B: Since the cooperative interference node is a selfish and energy-limited node, the energy for sending the interference signal comes from the energy collection of the source node’s transmission power, and the expected value of the collected energy is used instead of the real-time value; using a time-division transmission protocol, the entire transmission process is divided into two stages;

B1: In the first stage, the source node S serves as an energy source to wirelessly supply energy to the cooperative interference node J;

Define α as the time division ratio, assuming that the length of a time slot is T, where the initial αT, 0<α<1 is used for the wireless energy supply in the first stage, and the remaining time slot (1-α)T is used for information transmission;

In the first stage, the energy signal received by the cooperative interference node J from the source node S can be expressed as

Where PS is the transmission power of the source node, _{H SJ} represents the wireless power supply channel, which is a matrix of N _J × N _S and each element is a complex Gaussian random variable with zero mean and variance λ ₁ that obeys independent and identical distribution, x _S represents the energy signal vector of N _S ×1 and satisfies the total power limit condition nS represents a Gaussian additive white noise vector of N _S ×1 and

Therefore, at the end of the first stage, the total energy obtained by the cooperative interference node J is

Wherein η (0<η<1) represents energy conversion efficiency;

Due to the selfishness and friendliness of the cooperative interference node J, that is, to send interference signals with the expectation of collecting energy, the transmission power of node J can be written as

B2: In the second stage, the source node S transmits information to the destination node D, and at the same time, the cooperative interference node sends an interference signal, and the eavesdropping node eavesdrops on useful information;

According to the channel CSI known by the cooperative nodes, the following two cooperative interference schemes are adopted: the linear beamforming BF method and the antenna selection AS method;

B2-1: Linear beamforming BF method

Since multi-antennas are configured at the cooperative nodes, multi-antennas can be used to form beamforming transmission signals to enhance the reliability and security of destination node reception;

Therefore, the received signal of a legitimate user can be expressed as

Among them, h _sd is a vector of N _S ×1, indicating the legal channel coefficient; h _jd is a vector of N _J ×1, indicating the interference channel from cooperative node J to eavesdropping node E; each element in h _sd and h _jd is subject to Independent and identically distributed complex Gaussian random variables with zero mean and variance λ ₂ and λ ₄ ; the beamforming of the source node adopts the maximum ratio transmission, that is, w ₁ = h' _sd /||h _sd || is the beam of N _S ×1 Forming vector, x is the source signal of unit power, w ₂ is the N _J ×1 beamforming vector of the interference signal and ||w ₂ || ² =1, z is the interference signal of unit power, n _d is the legal The AWGN signal with the mean value of the user receiver being 0 and the variance being N ₀ ;

Similarly, the received signal at eavesdropper E can be expressed as

Among them, h _se is a vector of N _S ×1, which represents the eavesdropping channel coefficient; h _je is a vector of N _J ×1, which represents the interference channel from coordinating node J to eavesdropping node E; each element in h _sd and h _jd is subject to The independent and identically distributed complex Gaussian random variables with zero mean and variance of λ ₃ and λ ₅ , _ne is the eavesdropper receiver’s mean value of 0 and variance of N ₀ AWGN signal;

Therefore, combined with formula (3), the end-to-end SINR of legal user D can be expressed as

make Then the above formula can be abbreviated as

It is assumed here that the noise at the eavesdropper can be ignored, because at the eavesdropping user, the interference signal dominates, and this is also a commonly used assumption in the literature, which can be regarded as a worst case;

Therefore, combined with formula (3), the end-to-end signal-to-interference ratio SIR of the eavesdropper E can be expressed as

From equations (7) and (8), it can be seen that the beamforming vector w ₂ is related to the channels h _jd and h _je , and different beamforming vectors w ₂ can be designed according to the known channel CSI information to enhance the transmission of the system and safety performance.

B2-2: Antenna selection method;

According to the principle of AS, the cooperative node selects an optimal antenna from multiple antennas to transmit interference signals based on the set standard, so as to achieve the purpose of enhancing system security;

Assuming that k is the index of the selected antenna, according to equations (7) and (8), the SINR at the legitimate user D and the SIR at the eavesdropper E can be expressed as

in Indicates the channel coefficient between the kth antenna of the cooperative node J and the legitimate user D, Indicates the channel coefficient between the kth antenna of the cooperative node J and the eavesdropper E.

The invention adopts cooperative interference nodes to send interference signals with expected collection power, and studies the security transmission performance of the system in five situations. Starting from the known channel CSI, the secrecy performance of different interference cooperation schemes under the ideal situation of known CSI of all channels, unknown CSI of eavesdroppers, known cooperative channel gain ranking and limited feedback rate channel CSI is studied respectively, and the derivation The closed-form solutions of connection interruption probability and confidentiality interruption probability in various situations are given, and the expressions under high SNR approximation are given; it can be concluded that there is a certain compromise relationship between the connection interruption probability and the confidentiality interruption probability of the system, which can be Select and design system parameters according to different system requirements to achieve a compromise between security and confidentiality.

Further, several schemes are simulated and compared, and the traversal attainable secrecy rate gain obtained under high SNR compared with no feedback is analyzed and studied; it is proved that the key parameters of the system play an important role in the security performance and The advantages of the finite rate feedback method in practice.

Description of drawings

Fig. 1 is the finite rate feedback cooperative interference eavesdropping transmission model of the present invention;

Fig. 2 is the curve that connection interruption probability of the present invention changes with SNR;

Fig. 3 is the curve that connection interruption probability of the present invention changes with SNR;

Fig. 4 is the curve that the security interruption probability of the present invention changes along with SNR;

Fig. 5 is the curve that the connection interruption probability of the present invention changes along with the number of feedback bits B;

Fig. 6 is the curve that connection interruption probability and confidentiality interruption probability vary with slot division factor α of the present invention;

Fig. 7 is a curve showing the variation curve of the required minimum number of feedback bits B with the security interruption probability limit in the present invention.

Detailed ways

As shown in Figure 1, the present invention comprises the following steps:

A: Establish a limited-rate feedback cooperative interference eavesdropping transmission model considering MISO, including multi-antenna source node S, single-antenna legal user D, wireless-powered cooperative interference node J and single-antenna eavesdropping node E;

Cooperating nodes can be multiple or multi-antenna. Here, for the convenience of calculation and analysis, a single-antenna cooperating node can be regarded as a special case of selecting an optimal coordinating node from multiple coordinating nodes;

Assume that S has the number of antennas N _S , the number of antennas configured by J is N _J , and the rest of the nodes use a single antenna; assuming that the channel is a quasi-static Rayleigh channel, that is, the CSI of the channel in each transmission block remains unchanged, and the channel CSI between different transmission blocks Channel CSI varies independently;

B: Since the cooperative interference node is a selfish and energy-limited node, the energy for sending the interference signal comes from the energy collection of the source node’s transmission power, and the expected value of the collected energy is used instead of the real-time value; using a time-division transmission protocol, the entire transmission process is divided into two stages;

B1: In the first stage, the source node S serves as an energy source to wirelessly supply energy to the cooperative interference node J;

Define α as the time division ratio, assuming that the length of a time slot is T, where the initial αT, 0<α<1 is used for the wireless energy supply in the first stage, and the remaining time slot (1-α)T is used for information transmission;

In the first stage, the energy signal received by the cooperative interference node J from the source node S can be expressed as

Where PS is the transmission power of the source node, _{H SJ} represents the wireless power supply channel, which is a matrix of N _J × N _S and each element is a complex Gaussian random variable with zero mean and variance λ ₁ that obeys independent and identical distribution, x _S represents the energy signal vector of N _S ×1 and satisfies the total power limit condition n _S represents the Gaussian additive white noise vector of N _S ×1 and

Therefore, at the end of the first stage, the total energy obtained by the cooperative interference node J is

Wherein η (0<η<1) represents energy conversion efficiency;

Due to the selfishness and friendliness of the cooperative interference node J, that is, to send interference signals with the expectation of collecting energy, the transmission power of node J can be written as

B2: In the second stage, the source node S transmits information to the destination node D, and at the same time, the cooperative interference node sends an interference signal, and the eavesdropping node eavesdrops on useful information;

According to the channel CSI known by the cooperative nodes, the following two cooperative interference schemes are adopted: the linear beamforming BF method and the antenna selection AS method;

B2-1: Linear beamforming BF method

Since multi-antennas are configured at the cooperative nodes, multi-antennas can be used to form beamforming transmission signals to enhance the reliability and security of destination node reception;

Therefore, the received signal of a legitimate user can be expressed as

Among them, h _sd is a vector of N _S ×1, indicating the legal channel coefficient; h _jd is a vector of N _J ×1, indicating the interference channel from cooperative node J to eavesdropping node E; each element in h _sd and h _jd is subject to Independent and identically distributed complex Gaussian random variables with zero mean and variance λ ₂ and λ ₄ ; the beamforming of the source node adopts the maximum ratio transmission, that is, w ₁ = h' _sd /||h _sd || is the beam of N _S ×1 Forming vector, x is the source signal of unit power, w ₂ is the N _J ×1 beamforming vector of the interference signal and ||w ₂ || ² =1, z is the interference signal of unit power, n _d is the legal The AWGN signal with the mean value of the user receiver being 0 and the variance being N ₀ ;

Similarly, the received signal at eavesdropper E can be expressed as

Among them, h _se is a vector of N _S ×1, which represents the eavesdropping channel coefficient; h _je is a vector of N _J ×1, which represents the interference channel from coordinating node J to eavesdropping node E; each element in h _sd and h _jd is subject to The independent and identically distributed complex Gaussian random variables with zero mean and variance of λ ₃ and λ ₅ , _ne is the eavesdropper receiver’s mean value of 0 and variance of N ₀ AWGN signal;

Therefore, combined with formula (3), the end-to-end SINR of legal user D can be expressed as

make Then the above formula can be abbreviated as

It is assumed here that the noise at the eavesdropper can be ignored, because at the eavesdropping user, the interference signal dominates, and this is also a commonly used assumption in the literature, which can be regarded as a worst case;

Therefore, combined with formula (3), the end-to-end signal-to-interference ratio SIR of the eavesdropper E can be expressed as

From equations (7) and (8), it can be seen that the beamforming vector w ₂ is related to the channels h _jd and h _je , and different beamforming vectors w ₂ can be designed according to the known channel CSI information to enhance the transmission of the system and safety performance.

B2-2: Antenna selection method;

According to the principle of AS, the cooperative node selects an optimal antenna from multiple antennas to transmit interference signals based on the set standard, so as to achieve the purpose of enhancing system security;

Assuming that k is the index of the selected antenna, according to equations (7) and (8), the SINR at the legitimate user D and the SIR at the eavesdropper E can be expressed as

in Indicates the channel coefficient between the kth antenna of the cooperative node J and the legitimate user D, Indicates the channel coefficient between the kth antenna of the cooperative node J and the eavesdropper E.

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

Step A: Establish the system model, as shown in Figure 1, consider the finite rate feedback cooperative interference eavesdropping transmission model of MISO, which includes multi-antenna source node S, single-antenna legal user D, wireless power supply cooperative interference node J and single The antenna eavesdrops on node E. What is considered here is a typical eavesdropping model assisted by interference and cooperation. Under the premise that the cooperative nodes do not consume their own resources, they cooperate in the information transmission of legitimate users to enhance the anti-eavesdropping ability of the system. Cooperating nodes can be multiple or multi-antenna. Here, for the convenience of calculation and analysis, a single-antenna cooperating node can be regarded as a special case of selecting an optimal coordinating node from multiple coordinating nodes. Since the energy of the coordinating node comes from the wireless power supply of the source node, the power of its transmitted interference changes with the change of the transmission power of the source node, which is a typical feature different from other systems. It is assumed here that S has the number of antennas N _S , the number of antennas configured by J is N _J , and the rest of the nodes use a single antenna. It is assumed that the channels are all quasi-static Rayleigh channels, that is, the channel CSI in each transmission block is constant, and the channel CSI between different transmission blocks changes independently.

Since the cooperative interference node is a selfish and energy-limited node, the energy of transmitting the interference signal comes from the energy collection of the source node's transmission power. Here, in order to simplify the complexity of the interference node, the expected value of the collected energy is used instead of the real-time value. Therefore, this not only simplifies the processing complexity of the interfering node, but also reduces the overhead of obtaining channel CSI, and on average, the interfering node does not consume its own energy. Time Division (TD) transmission protocol is adopted, and the entire transmission process is divided into two stages. In the first stage, the source node S acts as an energy source to wirelessly supply energy to the cooperative interference node J; in the second stage, the source node S transmits information to the destination node D, while the cooperative interference node sends interference signals, and the eavesdropping nodes eavesdrop on useful information. Define α as the time division ratio, assuming that the length of a time slot is T, where the initial αT, 0<α<1 is used for the wireless energy supply in the first stage, and the remaining time slot (1-α)T is used for information transmission.

In the first stage, the energy signal received by the cooperative interference node J from the source node S can be expressed as

Where PS is the transmission power of the source node, _{H SJ} represents the wireless power supply channel, which is a matrix of N _J × N _S and each element is subject to the zero-mean variance of the IID (Identically and Independently Distributed, iid) is The complex Gaussian random variable of λ ₁ , x _S represents the energy signal vector of N _S ×1 and satisfies the total power constraint condition n _S represents the Gaussian additive white noise vector of N _S ×1 and Therefore, at the end of the first stage, the total energy obtained by the cooperative interference node J is

Wherein η (0<η<1) represents the energy conversion efficiency. Since the noise energy is small compared to the transmitted signal energy, the noise energy is ignored here. Due to the selfishness and friendliness of the cooperative interference node J, that is, to send interference signals with the expectation of collecting energy, the transmission power of node J can be written as

It is assumed that the source node knows the instantaneous CSI of the legitimate channel, but the instantaneous CSI of the eavesdropping channel is unknown. The rationality of this assumption is that the CSI of the legitimate channel can be obtained through various training-based channel estimation methods, such as MMSE, etc., and at the same time, an ideal channel CSI can be obtained by increasing the pilot power; Passive, will not actively report its channel CSI or may report inaccurate channel CSI, so here it is assumed that the source node does not know the status information of the source node and the eavesdropper channel. On the other hand, due to the complexity of the cooperative node and the limitation of system scheduling, the instantaneous CSI of the channel connected to the cooperative interfering node is not easy to obtain. Here, the known ideal CSI, known channel gain ordering and partial CSI based on finite rate feedback are considered Three scenarios, in which the security performance of CSI based on channel gain sorting and finite rate feedback is studied, and compared with ideal CSI. According to the channel CSI known by the cooperative nodes, the following two cooperative interference schemes can be mainly adopted: linear beamforming (BF) and antenna selection (AS). The source node S utilizes all antenna degrees of freedom to adopt the beamforming method to enhance information transmission, and the cooperative node uses the collected expected power to play the role of artificial interference in the information transmission stage. Since the cooperative nodes are also configured with multiple antennas, different cooperative interference schemes can be adopted according to the performance required by the system, implementation complexity and overhead.

Cooperative Interference Transmission Scheme

Since multiple antennas are configured at the coordination node, the multiple antennas can be used to form a beamforming transmission signal to enhance the reliability and security of receiving by the destination node. Therefore, the received signal of a legitimate user can be expressed as

Among them, h _sd is a vector of N _S ×1, indicating the legal channel coefficient; h _jd is a vector of N _J ×1, indicating the interference channel from coordinating node J to eavesdropping node E. The elements in h _sd and h _jd are independent and identically distributed complex Gaussian random variables with zero mean variance λ ₂ and λ ₄ respectively. The beamforming of the source node adopts the maximum ratio transmission, that is, w ₁ =h' _sd /||h _sd || is the beamforming vector of N _S ×1, x is the source signal of unit power, w ₂ is the interference signal N _J ×1 beamforming vector and ||w ₂ || ² =1, z is the interference signal with unit power, n _d is the AWGN signal with zero mean value and N ₀ variance of the legal user receiver. Similarly, the received signal at eavesdropper E can be expressed as

Among them, h _se is a vector of N _S ×1, representing the eavesdropping channel coefficient; h _je is a vector of N _J ×1, representing the interference channel from coordinating node J to eavesdropping node E. The elements in h _sd and h _jd are independent and identically distributed complex Gaussian random variables with zero mean variance λ ₃ and λ ₅ respectively, _ne is the eavesdropper receiver’s mean value is 0 and variance is N ₀ AWGN signal.

Therefore, combined with (3), the end-to-end signal-to-interference-noise ratio (SINR) of legal user D can be expressed as

make Then the above formula can be abbreviated as

It is assumed here that the noise at the eavesdropper can be ignored, because at the eavesdropping user, the interference signal plays a dominant role, and this is also a commonly used assumption in the literature, which can be regarded as a worst case. Therefore, combined with (3-3) the end-to-end signal-to-interference ratio (SIR) of the eavesdropper E can be expressed as

From equations (7) and (8), it can be seen that the beamforming vector w ₂ is related to the channels h _jd and h _je , and different beamforming vectors w ₂ can be designed according to the known channel CSI information to enhance the transmission of the system and safety performance.

Antenna selection method

Antenna selection is a low-complexity and energy-efficient approach to enhance physical layer security ^[52] , which is especially suitable for systems with limited computing power and energy. Considering the selfishness of nodes and the limitation of collected wireless energy, AS is an effective method to save energy and enhance security. According to the principle of AS, the cooperative node selects the best antenna from multiple antennas to transmit interference signals based on certain standards, so as to enhance the security of the system. Assuming that k is the index of the selected antenna, according to equations (7) and (8), the SINR at the legitimate user D and the SIR at the eavesdropper E can be expressed as

in Indicates the channel coefficient between the kth antenna of the cooperative node J and the legitimate user D, Indicates the channel coefficient between the kth antenna of the cooperative node J and the eavesdropper E.

Cooperative interference method based on known channel CSI

The method of transmitting beamforming design and AS is mainly given, which can minimize the impact on legitimate users while damaging the channel quality of eavesdroppers. For the known CSI situation of the channel, different cooperative interference methods and strategies can be designed.

The first case: ideal situation: both the cooperative node-destination node and the cooperative node-eavesdropper CSI are known

When the CSI of h _jd is ideal and known, the legal user receiver can use the zero-forcing method to completely eliminate the interference from the cooperative node J. At the same time, the interference at the eavesdropper can be maximized with known h _je CSI. Therefore, the beamforming vector _w2 is the solution to the following optimization problem:

According to Ding's research results ^[53] , the solution of the above optimization problem can be expressed as

where N _J ×N _J matrix is the orthogonal complement of the h _jd column space, which can be expressed as

Case 2: Cooperating nodes only—the CSI of the destination node is known

When the CSI of h _jd is ideal and known, the legal user receiver can still use the zero-forcing method to completely eliminate the interference from the cooperative node J. Since the CSI of h _je is unknown, it is impossible to use the channel CSI between the cooperating node and the eavesdropper to design the beamforming matrix to increase the interference at the eavesdropper as in the first case. Therefore, the null space of cooperative interference is aligned with legitimate users, so that the interference to legitimate users is 0, and random interference is generated to eavesdroppers. The beamforming vector design method first requires the matrix For singular value decomposition (SVD), we can get

Therefore, the N _J -1 left singular value vectors u _i corresponding to N _J -1 non-zero singular values become The column space of u _i can be expressed as A linear combination of column vectors, so that

Therefore, the beamforming vector can be arbitrarily selected from N _J -1 left singular value vector u _i , which can ensure that the interference at the legal user is 0, and the interference at the eavesdropper is random interference at this time, and 4.3.1 where is the jointly designed beamforming vector, and the interference at the eavesdropper is the maximum interference that can be achieved.

The third case: the channel gain ranking of the cooperative node-destination node is known

When the channel gain ranking of h _jd is known, the AS method can be used to select the antenna index corresponding to the minimum channel gain according to formula (9), which can minimize the influence of the cooperative interference node on legitimate users. Therefore, the criterion chosen by AS is

where h _id is the ith element of h _jd . Note that the antenna index selected by this scheme is a random antenna selection for the eavesdropping channel.

The fourth case: h _jd limited rate feedback: based on the codebook selection method with the least interference of legitimate users

Partial channel CSI between cooperative interfering nodes and legitimate users can be obtained through rate-limited feedback. In the finite rate feedback model, the legal user estimates and quantizes the channel CSI between the cooperative node and the destination node, and selects an optimal codebook to quantize the channel CSI. The codebook set can be obtained by random vector quantization (RVQ), that is, from A codebook is selected from the prequantized codebook sets with a size of 2 as channel CSI and fed back to the coordinating node, where ^B is the number of feedback bits. Quantized codebook set It consists of 2 ^B N _J -dimensional unit norm vectors, and the codebook set is known to both the cooperative node and the legal user. According to a certain selection criterion, the legitimate user transmits the selected codebook index through the feedback channel.

Since the channel CSI corresponding to the codebook codebook is the quantized channel CSI, which has a certain deviation from the actual channel CSI, therefore, there is inevitably some interference to legal users. Taking the interference minimization at the legal user as the optimization goal, the criteria for selecting the beamforming vector are as follows

in is the Channel Direction Information (CDI) vector.

The fifth case: h _jd finite rate feedback: channel orthogonality based codebook selection method

Compared with the fourth case: the limited feedback scheme is similarly adopted and the required feedback is the same, the difference lies in the method of constructing the beamforming vector w ₂ is different. Different from the fourth case, the codebook vector closest to the orthogonality with h _jd is used as the beamforming vector, so as to minimize the interference to legal users. In this subsection, the codebook vector closest to h _jd is first selected as the feedback information, and then the beamforming vector orthogonal to the legal channel is constructed according to the feedback information. Therefore, the codebook selection criteria are as follows

Feedback-based indexing to get the closest codebook vector to h _jd Collaborating nodes are working with The interference signal is sent in the orthogonal direction, which can ensure that the interference to the legal channel is as small as possible. Therefore, using The beamforming vector w ₂ can be constructed using a method similar to 4.3.2. The essence of the method in this subsection lies in the chosen codebook The closer to the real channel h _jd , the closer the constructed beamforming vector w ₂ is to the orthogonal direction of the real channel, so that the interference of legal users can be kept as small as possible.

safety performance

For the proposed cooperative jamming system, the probability of connection interruption, the probability of secrecy interruption and the traversal achievable secrecy rate are considered as safety indicators. For the convenience of description and calculation, define

g _sd ＝||h _sd || ² , g _JS ＝||H _SJ || ² , Thus, formulas (7)-(9) can be written as follows:

According to Simon's confirmation, g _sd respectively obey the Chi distribution with parameter 2N _s , and the corresponding PDF can be written as

Note that when using AS, g _se is the sum of an exponentially distributed Gaussian variable and a unit direction random variable. According to Shah's confirmation, it is known that it obeys an exponential distribution, and the corresponding PDF is

In the following, the system security index under the cooperative jamming method is derived and studied, and the influence of each parameter on the system performance is studied and analyzed.

Probability of connection loss and probability of loss of confidentiality

(1) Ideal situation: both the cooperative node-destination node and the cooperative node-eavesdropper CSI are known

It can be seen from the definition of COP that when the capacity of the legal user channel is not enough to support the transmission rate, that is, C _D < R _t , an interruption event occurs. Therefore, for the ideal CSI situation in 4.3.1, substituting equations (10) and (12) into equation (18), we can get and It can be seen from [55] that g _je obeys Gamma(N _J -1,λ ₅ ), and its PDF is

Substituting equation (18), the connection interruption probability in case 1 can be written as

in According to the definition of the confidentiality outage probability, the analytical formula of the confidentiality outage probability of the system can be obtained by substituting formula (20) and using the calculation:

in

From formula (22), it can be seen that the connection interruption probability of the system is related to the transmission power _{P S} , the time slot division factor α, the number of source node antennas and the transmission rate R _t , and it decreases with the increase of PS, and with the increase of The increase of α and R _t increases. When P _S →∞, the connection outage probability of the system can be written as

Therefore, it can be obtained that the diversity order is d=N _S , that is, the number of transmitting antennas. It can be seen from formula (23) that the secrecy _outage probability has nothing to do with the transmission power PS, only the time slot division factor α, the number of cooperative antennas and the secrecy rate R _e are related. When α is given, the secrecy outage probability approaches a constant, and The secrecy rate decreases with the increase of the number of cooperative node antennas, that is, the anti-eavesdropping ability is enhanced, because more cooperative node antennas provide more degrees of freedom.

(2) Collaborative node - the CSI of the destination node is known

Compared with the case 1 where the CSI of the interference channel is known, the CSI of the interference channel from the coordinating node to the eavesdropper is unknown here. At this time, the SNR of the legitimate user remains unchanged, and the expression of the connection interruption probability is shown in formula (22), which will not be further expressed here. For the SIR of the eavesdropper, it is known from the above that the precoding vector of the coordinating node at this time is w ₂ =u _i , then Obey the exponential distribution, that is, degenerate from the Gamma distribution in case 1 to the exponential distribution Therefore, using a similar method, the analytical formula of the system's confidentiality outage probability can be obtained as

Same as case 1, the secrecy outage probability has nothing to do with the transmission power _PS , but is only related to the slot division factor α and the secrecy rate _Re . When α is given, the secrecy outage probability approaches a constant. The difference is that the secrecy interruption probability has nothing to do with the number of antennas of the cooperative node, that is, the secrecy rate will not decrease with the increase of the number of antennas of the cooperative node, because the CSI of the channel from the cooperative node to the eavesdropper is unknown, and the design of the beamforming vector Unable to utilize channel information, the increase in the number of coordinating node antennas does not provide more degrees of freedom for the eavesdropper. Therefore, the anti-eavesdropping ability of the system cannot be improved by increasing the number of antennas of the cooperative nodes.

(3) Cooperating nodes—the channel gain ranking of the destination nodes is known and the coordinating nodes use AS

Different from the interference channel h _jd from the cooperative node to the legal user in case 2 is completely known, here it is assumed that the order of the channel gains of the antennas corresponding to the interference channel from the cooperative node to the legal user is known, and the cooperative node can adopt the AS scheme. The node only needs to feed back the index of the corresponding optimal antenna, without the need for complete CSI of the channel, which not only reduces the complexity and overhead of the system but also saves the energy of the cooperative nodes. The selected transmit antenna is chosen randomly for eavesdroppers and optimal for legitimate users. Therefore, the security of the system can be increased while reducing the interference of legal receiving nodes. at this time Since the chosen antenna is chosen at random for the eavesdropper, obeys the exponential distribution Exp(λ ₅ ), and the selected antenna is the antenna with the smallest channel gain for legal users, then its CDF can be calculated as

Therefore, its PDF can be expressed as f _gjd (x)=N _J /λ ₄ exp(-xN _J /λ ₄ ), which can be regarded as an exponential distribution whose parameter is λ ₄ /N _J. Based on the definition of the connection interruption probability, the expression of the legal user SINR is substituted, and the expression of the system connection interruption probability in the third case can be written as

Among them, (a) is obtained by substituting the CDF of g _sd into sorting, (b) is obtained by substituting the PDF of g _jd into sorting after using the binomial expansion, and (c) is calculated according to [56] and can be obtained after sorting have to.

Using a method similar to Case 2, the analytical formula of the system’s secrecy outage probability can be obtained as

Due to the complexity of formula (27), in order to better study the influence of system parameters, the analytical formula of outage probability of the system under high transmit power is derived here. When P _S →∞, the connection interruption probability of the system can be calculated as

Different from cases 1 and 2, the connection interruption probability of the system has nothing to do with the transmission power _PS , and tends to a constant with the increase of _PS , so the diversity order is d=0. This is because the transmission power of the cooperative node comes from the wireless energy supply of the source node, the transmission power of the cooperative node is proportional to the transmission power of the source node, and the interference signal and useful signal power transmitted by the cooperative node increase with the increase of the transmission power of the source node. The same scale increases, so that the diversity order is 0. In addition, the connection interruption probability is related to the number of cooperative node antennas, and decreases with the increase of the number of cooperative node antennas. This is because the more the number of cooperative node antennas, the more degrees of freedom the optional antennas provide, and the interference to legitimate users Smaller antennas have a higher probability of occurrence, but an increase in the number of antennas will increase the number of bits required for the feedback index. At the same time, the connection interruption probability is also related to the time slot division factor α, the number of source node antennas and the transmission rate R _t , and the connection interruption probability increases with the increase of α and R _t , and with the increase of the source node antenna number And reduce. The system's secrecy outage probability expression is the same as case 2, it has nothing to do with the sending power _PS , but is only related to the time slot division factor α and the secrecy rate _Re . This is because the antenna selection criterion only considers the channel CSI between the coordinating node and the destination node, which is equivalent to random selection for the eavesdropper.

(4)h _jd limited rate feedback: codebook selection method based on legal user interference minimum

Case 2 is that the CSI of the channel h _jd from the coordinating node to the destination node is completely known. Case 3 is that the channel gain order of h _jd is known. Here, part of the CSI of the channel h _jd is considered to be known, that is, the channel CSI model using finite rate feedback. Since there is a certain error between the feedback quantized channel CSI and the ideal CSI, it can be regarded as part of the channel CSI is known, and the orthogonal space of the channel is not exactly the same as the ideal situation. Unlike situations 1 and 2, g _jd ≠ 0, that is, cooperative The interference signal sent by the interference node will partially leak to the legitimate user end, which will cause interference to the legitimate user. Here, the channel CSI design beamforming vector corresponding to the codebook with the least leakage interference is selected. According to formula (16), the channel gain from the cooperative node to the legal user is The two random variables it consists of are independent of each other, is the inner product of two unit vectors, whose CDF is x∈[0,1]. U is the minimum value of M=2 ^B independent and identically distributed random variables, and its CDF can be expressed as

By the definition of COP, the system connection interruption probability in the fourth case is given by the following theorem.

Theorem 1: The channel CSI from the cooperative node to the destination node adopts finite rate feedback, and the CSI of h _je is unknown, the expression of the connection interruption probability of the cooperative interference system can be written as

where _Φ1 is given by

Proof: According to the definition of connection interruption probability, substituting the legal user SINR expression, using a method similar to formula (27) can be obtained

in Therefore, the CDF of g _jd needs to be calculated as follows

Thus, the PDF of g _jd can be obtained as follows

Substituting equation (36), we can get

Substitute Equation (37) into Equation (32) to get the connection interruption probability of the system.

In the following, the secrecy outage probability is deduced. According to the existing theorem, g _je obeys the exponential distribution Exp(λ ₅ ). Using a method similar to case 2, the analytical formula of SOP can be obtained as

From formula (32), it can be known that the connection interruption probability of the system is related to the transmission power P _S , the number of cooperative node antennas, the number of source node antennas, the time slot division factor α and the secrecy rate _Re , and with the increase of the feedback bit B, the channel The more accurate the CSI, the less interference the cooperating node will have on legitimate users, and the probability of connection interruption decreases with the increase of the number of feedback bits, and increases with the increase of the number of cooperating node antennas, because the more the number of coordinating node antennas More, the rougher the quantization of the channel under the same feedback bit, the more interference leaked by the cooperative interference signal. The probability of connection interruption increases with the increase of α and R _t , and decreases with the increase of the number of source node antennas. The system's secrecy outage probability expression is the same as case 2, it has nothing to do with the transmission power _PS , only related to the time slot division factor α and the secrecy rate _Re . This is because the antenna selection criterion only considers the feedback optimization of the channel CSI between the cooperative node and the destination node, which is random for eavesdroppers.

(5)h _jd finite rate feedback: an orthogonal codebook selection method based on interference channels

Different from the beamforming vectors constructed with the least interference to legal users in case 4, limited rate feedback is also used here, but the codebook index closest to the actual channel is selected and fed back to the coordinating node, and the coordinating node uses the feedback codebook as the real channel The CSI designs the beam vector so that it is orthogonal to the interference channel, so that the interference to legitimate users is minimal. According to the RVQ theory, the actual channel direction and quantized channel direction relationship can be expressed as

in s is with Unit-norm vectors that are orthogonal and independent of v. w ₂ is based on The beam vector constructed by the null space of , in order to avoid interference to legitimate users as much as possible. Therefore there are

According to the quantization unit approximation model ^[46] , w ₂ is a unit norm vector independent of s, then Z=|s ^H w ₂ | ² ～β(1,N _J -2), and ||h _jd || ² obeys Gamma distribution with parameters (N _J ,λ ₄ ), it can be seen from [47] that X=||h _jd || ² sin ² θ obeys Gamma distribution with parameters (N _J -1,λ ₄ δ), where Thus the CDF of g _jd can be expressed as

It can be obtained from the above formula that g _jd obeys the exponential distribution whose parameter is λ ₄ δ. Due to finite rate CSI feedback, This results in the residual interference of the cooperating nodes to legitimate users, and the residual interference depends on the number of bits B of the feedback CSI. When the number of feedback bits B→∞, that is, full feedback, the residual interference approaches 0. g _je obeys the exponential distribution Exp(λ ₅ ), and the analytical formula of the confidentiality interruption probability is the same as formula (38). According to the definition of connection interruption probability, using a method similar to case 3, the analytical formula of connection interruption probability is calculated as follows

At high SNR, Gaussian white noise at the receiver can be ignored, and the connection outage probability of the system can be calculated as

in Given the coding rate R _t , the number of antennas at each node, the time slot division factor α and the probability of successful transmission ξ, the required number of feedback bits B can be calculated by formula (4-43) as follows

From equation (43), it can be seen that the system connection interruption probability has nothing to do with the transmission power _PS , but is related to the number of cooperative node antennas, the number of source node antennas, the time slot division factor α and the secrecy rate _Re , and with the increase of the feedback bit B , the more accurate the δ→0 channel CSI is, the less interference the cooperating node will have on legitimate users, the probability of connection interruption decreases with the increase of the number of feedback bits, and increases with the increase of the number of coordinating node antennas, because The more the number of cooperative node antennas, the rougher the quantization of the channel under the same feedback bit, the more interference leaked by the cooperative interference signal, and the number of feedback bits should increase linearly with the number of cooperative antennas. The probability of connection interruption increases with the increase of α and R _t , and decreases with the increase of the number of source node antennas. The system's secrecy outage probability expression is the same as case 2, it has nothing to do with the transmission power _PS , only related to the time slot division factor α and the secrecy rate _Re .

Discussion: (1) From the above analysis and research, it can be seen that under the same system parameters, the secrecy outage probability of the system with known ideal CSI is constant and minimum. When the ordering CSI of the cooperative interference channel gain is known, use AS, that is, case 3, the cooperative interference node generates interference to the legitimate user and the eavesdropper at the same time, but the selected antenna minimizes the interference to the legitimate user, so that the legitimate user The effect of the eavesdropper is smaller than that of the eavesdropper. In the case of cooperative interference with finite rate feedback, due to a certain error between the feedback channel and the ideal channel CSI, it will cause certain interference to legitimate users, and thus damage the performance of legitimate users. In addition, for non-ideal CSI (AS and rate-limited feedback), at high SNR, legitimate users That is, it has nothing to do with SNR and tends to a constant value, so there is a bit error flat layer effect. As the number of feedback bits B increases, the fed back CSI approaches the real CSI, the residual interference approaches 0, and the system performance approaches the ideal CSI situation. Therefore, the accuracy of CSI estimation and feedback has a significant impact on system performance. In addition, from the perspective of system complexity and overhead, in order to obtain a more ideal channel CSI requires a large system overhead and transmit power, which affects the transmission performance of the system, and the method based on feedback channel sorting index only needs to feed back log ₂ N _J bits, less feedback information is required, but at the same time the performance gain is also limited. The method based on finite rate feedback can obtain relatively accurate channel CSI and the overhead of the system can be controlled according to the performance requirement, which can achieve a better compromise between performance and complexity.

(2) Considering the comparison situation where there is no cooperative node or the cooperative node does not participate in cooperative interference, at this time γ _D = ρ _s ||h _sd || ² , The corresponding connection interruption probability and confidentiality interruption probability are respectively

It can be seen from the above formula that when ρ _s → ∞, p _so → 1, p _co → 0, it can be seen that when there is no cooperative interference node, the eavesdropper can intercept the information of the legitimate user with a probability of 1, so it is necessary to design an anti-eavesdropping interference Algorithms to achieve a compromise between speed and security. The proposed schemes design different interference schemes according to different channel CSIs, which can not only achieve a good compromise between reliability and security, but also have practical application significance.

Simulation and Results Analysis

Carry out Mentor Carlo simulation verification on the above theoretical results and analyze the simulation results. The simulation parameters are set as follows unless otherwise specified: the number of independent simulations is 1×10 ⁶ , the encoding rate R _t sent by the source node = 2bps, the required secrecy rate R _e = 1bps, the conversion efficiency factor of energy harvesting η = 0.8, when The slot division factor α=0.1, the channel parameters are λ ₁ =10, λ ₂ =λ ₃ =λ ₄ =λ ₅ =1, and SNR is defined as the sending signal-to-noise ratio PS / _{N 0} of the source node. For ideal CSI (case 1), known cooperating node-destination node channel CSI (case 2), known coordinating node-destination node channel gain ranking (case 3), finite rate feedback based on legal user interference minimum (case 4) Numerical simulation and analysis are done for the five cases of finite rate feedback based on channel orthogonality (case 5).

Fig. 2 shows the curve of the connection interruption probability of the system changing with the SNR, where the solid line represents the antenna configuration of N _S =2, N _J =2, the dotted line represents the antenna configuration of N _S =4, N _J =2, and the feedback The number of bits B=8. It can be seen from the figure that the derived theoretical results are in good agreement with the actual simulation results, and the asymptotic expression is also consistent with the simulation results in the high SNR region. As analyzed above, the connection interruption probability increases with the increase in the number of source node antennas and decrease, and for case 1 and the diversity gain is the number of antennas _NS of the source node; in other cases, the connection interruption probability of the system decreases with the increase of the number of antennas of the source node, but there is a flat-bottom effect of error codes, that is, as the SNR The increase of tends to a constant value, and the flat bottom decreases with the increase of the number of source node antennas. It can be seen from the figure that the connection interruption probability of case 3 is the worst, followed by cases 4 and 5, and the performance of the connection interruption probability of cases 1 and 2 is the best. The beam vector of . Therefore, obtaining better channel CSI plays an important role in the performance of the system.

Fig. 3 shows the curve of the connection interruption probability of the system changing with SNR, where the solid line represents the antenna configuration of N _S =2, N _J =2, the dotted line represents the antenna configuration of N _S =2, N _J =3, and the feedback The number of bits B = 8, as can be seen from the figure, the derived theoretical results are in good agreement with the actual simulation results. For case 1, the connection interruption probability has nothing to do with the number of cooperative node antennas, while the connection interruption probability in other cases decreases as the number of cooperative node antennas increases, but there is a flat-bottom effect of error codes, that is, as the SNR increases, the probability of connection interruption decreases. At a constant value, the flat bottom decreases as the number of cooperative node antennas increases. Similarly, the connection interruption probability of case 3 is the worst, followed by cases 4 and 5, and the performance of connection interruption probability of cases 1 and 2 is the best, because the good beam vector is designed in case 1 and the CSI that makes full use of the interference channel .

Fig. 4 shows the curve of the security outage probability of the system changing with the SNR, wherein the solid line represents the antenna configuration of N _S =2, N _J =2, the dotted line represents the antenna configuration of N _S =2, N _J =4, and the feedback The number of bits B=8. It can be seen from the figure that the derived theoretical results are in good agreement with the actual simulation results. When the number of cooperative node antennas is 2, the analysis in the ideal situation shows that the security interruption probability of the system is the same, which is consistent with the simulation results. When the number of cooperative node antennas is greater than 2, since the known channel CSI is used to jointly design the beam vector in case 1, the spatial degree of freedom of the channel is fully utilized, and the system's secrecy outage probability is effectively reduced. In other cases, the confidentiality outage probability decreases with the increase of the number of cooperative node antennas, but there is a flat-bottom effect, that is, it tends to a constant value with the increase of SNR.

Fig. 5 shows the change curve of the connection interruption probability of the system with the number of feedback bits B, where SNR=15dB, N _S =4, N _J =4, and the number of feedback bits B=8. Cases 1, 2 and 3 have nothing to do with the number of feedback bits B, so the outage probability does not change with B. The connection interruption probability of cases 4 and 5 is related to the number of feedback bits, and decreases with the increase of the number of bits and gradually approaches case 1. This is because with the increase of the number of bits B, the higher the channel quantization accuracy is, the closer it is to the real channel, so the connection interruption probability of the system increases rapidly. Note that when the number of feedback bits is small, B<2, the connection interruption probability performance of the method based on antenna selection is better than that based on finite rate feedback, because the channel CSI quantization is too rough due to too few feedback bits, which leaks to the destination node. Increased interference leads to a decrease in the SINR of the destination node and a decrease in reliability. With the increase of the feedback bit B, the accuracy of channel quantization is improved, and it is getting closer to the real channel CSI, the interference of the interference signal sent by the interference node to the destination node tends to 0, and the connection interruption probability of the system tends to the ideal situation.

Fig. 6 shows the curves of the system's connection interruption probability and confidentiality interruption probability changing with the time slot division factor α, where N _S =2, N _J =2, and the number of feedback bits B=6. As a comparison item, the figure also shows the COP and SOP without coordination nodes. It can be seen from the figure that with the increase of α, the connection interruption probability of the system in the five situations increases, that is, the reliability decreases. This is because the time slot used for transmitting data is getting shorter and shorter, and the corresponding transmission rate On the other hand, with the increase of α, the security interruption probability of the system in the five situations decreases, that is, the security is enhanced, because the cooperative nodes can collect more energy for transmitting interference signals , the data transmission time slot becomes shorter, the SINR required by the eavesdropper increases, and it is less likely to be eavesdropped. Therefore, there is a compromise relationship between reliability and security. The system without cooperative nodes has a low probability of connection interruption, but a high probability of confidentiality interruption. In data transmission with priority on security, cooperative nodes play an important role.

Fig. 7 shows the change curve of the minimum number of feedback bits B required by systems with different numbers of antennas of cooperative nodes under the constraint condition of confidentiality outage probability ξ, where N _S =4. It can be seen from the figure that in order to achieve the same confidentiality outage probability, as the number of cooperative node antennas increases, the required minimum number of feedback bits B increases; in order to obtain quantization with the same precision, the increase in the number of antennas requires more quantization bits . Note that as the security breaking probability increases, the required minimum number of feedback bits B decreases; only a small bit feedback is needed to achieve the required security breaking probability.

The present invention is based on the secure transmission system performance of the cooperative jamming system with wireless energy supply, and the cooperative jamming nodes send jamming signals with expected collection power, and the secure transmission performance of the system in five situations is studied. Starting from the known channel CSI, the secrecy performance of different interference cooperation schemes under the ideal situation of known CSI of all channels, unknown CSI of eavesdroppers, known cooperative channel gain ranking and limited feedback rate channel CSI is studied respectively, and the derivation The closed-form solutions of connection interruption probability and confidentiality interruption probability in various situations are obtained, and the expressions under high SNR approximation are given; it can be concluded that there is a certain compromise relationship between the connection interruption probability and the confidentiality interruption probability of the system, which can be Select and design system parameters according to different system requirements to achieve a compromise between security and confidentiality. Further, the ergodic reachable secrecy rate of the system is deduced and studied, and several schemes are simulated and compared, and the gain of the ergodic reachable secrecy rate obtained under high SNR compared with no feedback is analyzed and studied; it is proved that the system The important role of the key parameters in the safety performance and the advantages of the finite rate feedback method in practical trials.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it still The technical solutions described in the foregoing embodiments can be modified, or some or all of the technical features can be equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.

Claims (1)

1. A method for secure transmission of cooperative interference physical layer based on wireless energy supply, characterized in that: comprising the following steps: A: Establish a limited-rate feedback cooperative interference eavesdropping transmission model considering MISO, including multi-antenna source node S, single-antenna legal user D, wireless-powered cooperative interference node J and single-antenna eavesdropping node E; Cooperating nodes can be multiple or multi-antenna. Here, for the convenience of calculation and analysis, a single-antenna cooperating node can be regarded as a special case of selecting an optimal coordinating node from multiple coordinating nodes; Assume that S has the number of antennas N _S , the number of antennas configured by J is N _J , and the rest of the nodes use a single antenna; assuming that the channel is a quasi-static Rayleigh channel, that is, the CSI of the channel in each transmission block remains unchanged, and the channel CSI between different transmission blocks Channel CSI varies independently; B: Since the cooperative interference node is a selfish and energy-limited node, the energy for sending the interference signal comes from the energy collection of the source node’s transmission power, and the expected value of the collected energy is used instead of the real-time value; using a time-division transmission protocol, the entire transmission process is divided into two stages; B1: In the first stage, the source node S serves as an energy source to wirelessly supply energy to the cooperative interference node J; Define α as the time division ratio, assuming that the length of a time slot is T, where the initial αT, 0<α<1 is used for the wireless energy supply in the first stage, and the remaining time slot (1-α)T is used for information transmission; In the first stage, the energy signal received by the cooperative interference node J from the source node S can be expressed as Where PS is the transmission power of the source node, _{H SJ} represents the wireless power supply channel, which is a matrix of N _J × N _S and each element is a complex Gaussian random variable with zero mean and variance λ ₁ that obeys independent and identical distribution, x _S represents the energy signal vector of N _S ×1 and satisfies the total power limit condition n _S represents the Gaussian additive white noise vector of N _S ×1 and Therefore, at the end of the first stage, the total energy obtained by the cooperative interference node J is Wherein η (0<η<1) represents energy conversion efficiency; Due to the selfishness and friendliness of the cooperative interference node J, that is, to send interference signals with the expectation of collecting energy, the transmission power of node J can be written as B2: In the second stage, the source node S transmits information to the destination node D, and at the same time, the cooperative interference node sends an interference signal, and the eavesdropping node eavesdrops on useful information; According to the channel CSI known by the cooperative nodes, the following two cooperative interference schemes are adopted: the linear beamforming BF method and the antenna selection AS method; B2-1: Linear beamforming BF method Since multi-antennas are configured at the cooperative nodes, multi-antennas can be used to form beamforming transmission signals to enhance the reliability and security of destination node reception; Therefore, the received signal of a legitimate user can be expressed as Among them, h _sd is a vector of N _S ×1, indicating the legal channel coefficient; h _jd is a vector of N _J ×1, indicating the interference channel from cooperative node J to eavesdropping node E; each element in h _sd and h _jd is subject to Independent and identically distributed complex Gaussian random variables with zero mean and variance λ ₂ and λ ₄ ; the beamforming of the source node adopts the maximum ratio transmission, that is, w ₁ = h' _sd /||h _sd || is the beam of N _S ×1 Forming vector, x is the source signal of unit power, w ₂ is the N _J ×1 beamforming vector of the interference signal and ||w ₂ || ² =1, z is the interference signal of unit power, n _d is the legal The AWGN signal with the mean value of the user receiver being 0 and the variance being N ₀ ; Similarly, the received signal at eavesdropper E can be expressed as Among them, h _se is a vector of N _S ×1, which represents the eavesdropping channel coefficient; h _je is a vector of N _J ×1, which represents the interference channel from coordinating node J to eavesdropping node E; each element in h _sd and h _jd is subject to The independent and identically distributed complex Gaussian random variables with zero mean and variance of λ ₃ and λ ₅ , _ne is the eavesdropper receiver’s mean value of 0 and variance of N ₀ AWGN signal; Therefore, combined with formula (3), the end-to-end SINR of legal user D can be expressed as make Then the above formula can be abbreviated as It is assumed here that the noise at the eavesdropper can be ignored, because at the eavesdropping user, the interference signal dominates, and this is also a commonly used assumption in the literature, which can be regarded as a worst case; Therefore, combined with formula (3), the end-to-end signal-to-interference ratio SIR of the eavesdropper E can be expressed as From equations (7) and (8), it can be seen that the beamforming vector w ₂ is related to the channels h _jd and h _je , and different beamforming vectors w ₂ can be designed according to the known channel CSI information to enhance the transmission of the system and safety performance. B2-2: Antenna selection method; According to the principle of AS, the cooperative node selects an optimal antenna from multiple antennas to transmit interference signals based on the set standard, so as to achieve the purpose of enhancing system security; Assuming that k is the index of the selected antenna, according to equations (7) and (8), the SINR at the legitimate user D and the SIR at the eavesdropper E can be expressed as in Indicates the channel coefficient between the kth antenna of the cooperative node J and the legitimate user D, Indicates the channel coefficient between the kth antenna of the cooperative node J and the eavesdropper E.