MIMO full-duplex secure communication parameter optimization method, optimization device and communication method

Abstract

The present invention discloses a MIMO full-duplex secure communication parameter optimization method, an optimization device, and a communication method based on the optimized parameter setting. The parameter optimization method constructs transmission signals of Alice and Bob, adds artificial noise thereto, and performs power allocation and dimension allocation on the effective signal and the artificial noise according to decision variables. At the same time, based on the singular value decomposition of the channel, a specific expression form of the effective signal and the artificial noise is constructed. MIMO full-duplex communication is performed with the constructed transmission signal, and the received signal of each end is obtained. The optimal power allocation coefficient, power allocation matrix, and signal dimension are obtained with the goal of maximizing the system average confidentiality rate. Communication is performed based on the optimized parameters, and the confidentiality performance of the MIMO full-duplex communication system can be improved from the physical layer.

Description

Technical Field

[0001] The present invention belongs to the technical field of physical layer security of wireless communication systems, and more specifically, relates to a MIMO full-duplex secure communication parameter optimization method, an optimization device, and a communication method. Background Art

[0002] In recent years, with the development of mobile internet, wireless networks have been increasingly used in education, scientific research, finance, transportation, and other fields, gradually becoming an integral part of daily work, study, and life. The widespread adoption of 5G technology has further expanded the application scope of wireless networks. However, due to the broadcast nature of wireless communication systems, they face more complex security issues than traditional wired communication systems.

[0003] Physical layer security is a key technology for ensuring secure information transmission in wireless communication systems. It explores the physical layer, the lowest level of communication model, and offers a different approach to confidentiality than upper-layer key encryption. Based on information theory, it leverages the time-varying, random, and heterogeneous nature of wireless channels, exploiting the differences between legitimate and eavesdropped channels to ensure secure information transmission.

[0004] In a MIMO full-duplex communication system, each antenna on both ends can simultaneously transmit and receive signals on the same frequency, significantly increasing the system's communication capacity. Regarding secure communication strategies, some literature has explored full-duplex receiver scenarios, where the receiver simultaneously transmits artificial noise while receiving the signal, but these studies do not consider bidirectional communication scenarios. Some public literature also considers bidirectional communication scenarios, but these full-duplex scenarios use separate transmit and receive antennas, rather than the currently state-of-the-art MIMO full-duplex mode. Therefore, physical layer security technologies for secure communication in MIMO full-duplex communication systems are currently lacking. Summary of the Invention

[0005] In view of the above defects or improvement needs of the prior art, the present invention provides a MIMO full-duplex secure communication parameter optimization method and a communication method, the purpose of which is to improve the confidentiality performance of MIMO full-duplex communication from the physical layer.

[0006] To achieve the above object, according to one aspect of the present invention, a method for optimizing parameters of MIMO full-duplex secure communication is provided, comprising:

[0007] Determine the number of antennas N on Alice, Bob, and Eve respectively A 、N B 、N E , and the transmission power P of Alice and Bob A 、P B, where Alice and Bob are legitimate communication terminals, and Eve is an eavesdropping terminal;

[0008] Construct the transmission signals of Alice and Bob respectively, where the transmission signal of Alice includes the effective signal s A and artificial noise z A , effective signal s A The total power is φ A P A , dimension is r b , the power allocation matrix is ​​Σ A ; Artificial noise z A The total power is (1-φ A )P A , dimension is N A -r b And the power in each dimension is the same; the transmitted signal at Bob's end includes the effective signal s B and artificial noise z B , where the effective signal s B The total power is φ B P B , dimension is r a , the power allocation matrix is ​​Σ B ; Artificial noise z B The total power is (1-φ B )P B , dimension is N B -r a And the power in each dimension is the same, φ A and φ B are the power allocation coefficients at Alice and Bob respectively;

[0009] The channel H from Alice to Bob is AB and the channel H from Bob to Alice BA Do singular value decomposition, in, is an orthogonal unitary matrix, with V A The first r b Column as valid signal s A The precoding matrix W A 、The rest N A -r b Column as artificial noise z A The precoding matrix Z A , in, is an orthogonal unitary matrix, with V B The first r a Column as valid signal s B The precoding matrix W B、The rest N B -r a Column as artificial noise z B The precoding matrix Z B ;

[0010] Construct effective signals s respectively A 、s B and artificial noise z A 、z B , where s A =W A u A , z A =Z A m A , s B =W B u B , z B =Z B m B ,u A 、m A is the valid information symbol vector and artificial noise symbol vector at Alice’s end, u B 、m B are the valid information symbol vector and artificial noise symbol vector at Bob's end;

[0011] Alice and Bob transmit signals simultaneously, and obtain the signals received by Alice, Bob, and Eve respectively;

[0012] Construct the system average confidentiality rate, and calculate the optimal power allocation coefficient φ with the goal of maximizing the system average confidentiality rate A and φ B , power allocation matrix Σ A and Σ B and dimension r b and r a .

[0013] In one embodiment, a mask beamforming method is used to process the effective signal and artificial noise of each transmission signal so that the receiving end can receive the original transmission information.

[0014] In one embodiment, each element in the channel satisfies a cyclic complex Gaussian distribution with a mean of 0 and a variance of 1.

[0015] In one embodiment, after receiving the signal, Alice and Bob process the original received signal to remove artificial noise interference in the signal.

[0016] In one embodiment, the signals received by Alice, Bob, and Eve are y A 、yB and y E ,in,

[0017] y A =H BA W B u B +H BA Z B m B +η A +n A ;

[0018] y B =H AB W A u A +H AB Z A m A +η B +n B ;

[0019] y E =G AE W A u A +G AE Z A m A +G BE W B u B +G BE Z B m B +n E ;

[0020] Among them, G AE , G BE are the channels from Alice to Eve and from Bob to Eve respectively; η A and η B They represent the interference terms caused by self-interference at Alice and Bob respectively, A 、n B They are transmission channels H BA 、H AB Additive Gaussian white noise, n E Transmission channel G AE , G BE Total additive white Gaussian noise;

[0021] Processing the original received signal, including:

[0022] The signal y received by Alice A Left multiplication Take the first r a Column as processing result, Indicates U B The conjugate transpose of

[0023] The signal y received by Bob B Left multiplication Take the first r b Column as processing result, Indicates U A The conjugate transpose of .

[0024] In one embodiment, the system average confidentiality rate Among them, R A 、R B are the instantaneous confidentiality rates of Alice and Bob respectively, is the average confidentiality rate of Eve, [x] + Indicates that when the value of x is less than 0, the result is set to 0.

[0025] In one embodiment, the average eavesdropping rate of Eve is converted into an upper bound optimization of the average eavesdropping rate through Jensen inequality, so that the problem of maximizing the system average confidentiality rate is converted into the problem of maximizing the lower bound of the system average confidentiality rate.

[0026] In one embodiment, the process of lower-bounding the maximum system average secrecy rate includes:

[0027] Step S61: Setting the initial distribution coefficient φ A and φ B , power allocation matrix Σ A and Σ B ;

[0028] Step S62: Based on the convex optimization concept, the CVX solver is used to obtain the optimal power allocation matrix under the current effective signal dimension and power allocation coefficient; the particle swarm algorithm is used to obtain the optimal power allocation coefficient under the current effective signal dimension and power allocation matrix;

[0029] Step S63: Update the optimal power allocation coefficient and power allocation matrix under the current effective signal dimension, repeat step S62 until the lower bound of the system average confidentiality rate reaches the maximum, and then end the iteration;

[0030] Step S64: Traverse each effective signal dimension scenario, repeat steps S62 to S63, and finally compare the performance to output the optimal power allocation coefficient φ A and φ B , power allocation matrix Σ A and Σ B and dimension r b and r a .

[0031] According to another aspect of the present invention, a MIMO full-duplex secure communication parameter optimization system is provided, comprising:

[0032] Scenario parameter input unit, used to input the number of antennas N on Alice, Bob, and Eve A 、N B 、N E , and the transmission power P of Alice and Bob A 、P B , where Alice and Bob are legitimate communication terminals, and Eve is an eavesdropping terminal;

[0033] The transmission signal construction unit is used to construct the transmission signals of Alice and Bob respectively, wherein the transmission signal of Alice includes the valid signal s A and artificial noise z A , effective signal s A The total power is φ A P A , dimension is r b , the power allocation matrix is ​​Σ A ; Artificial noise z A The total power is (1-φ A )P A , dimension is N A -r b And the power in each dimension is the same; the transmitted signal at Bob's end includes the effective signal s B and artificial noise z B , where the effective signal s B The total power is φ B P B , dimension is r a , the power allocation matrix is ​​Σ B ; Artificial noise z B The total power is (1-φ B )P B , dimension is N B -r a And the power in each dimension is the same, φ A and φ B are the power allocation coefficients at Alice and Bob respectively;

[0034] The channel decomposition unit is used to decompose the channel H from Alice to Bob. AB and the channel H from Bob to Alice BA Do singular value decomposition, in, is an orthogonal unitary matrix, with V A The first r b Column as valid signal sA The precoding matrix W A 、The rest N A -r b Column as artificial noise z A The precoding matrix Z A , in, is an orthogonal unitary matrix, with V B The first r a Column as valid signal s B The precoding matrix W B 、The rest N B -r a Column as artificial noise z B The precoding matrix Z B ;

[0035] Valid / noise signal construction unit, used to construct valid signal s respectively A 、s B and artificial noise z A 、z B , where s A =W A u A , z A =Z A m A , s B =W B u B , z B =Z B m B ,u A 、m A is the valid information symbol vector and artificial noise symbol vector at Alice’s end, u B 、m B are the valid information symbol vector and artificial noise symbol vector at Bob's end;

[0036] A signal receiving unit, configured to obtain the signals received by Alice, Bob, and Eve respectively after Alice and Bob transmit signals simultaneously;

[0037] Solving unit, used to construct the system average confidentiality rate, with the goal of maximizing the system average confidentiality rate, and calculate the optimal power allocation coefficient φ A and φ B , power allocation matrix Σ A and Σ B and dimension r b and r a .

[0038] According to another aspect of the present invention, there is provided a MIMO full-duplex secure communication method, comprising:

[0039] According to the above MIMO full-duplex secure communication parameter optimization method, the optimal power allocation coefficient φ in the current communication scenario is obtained. A and φ B , power allocation matrix Σ A and Σ B and dimension r b and r a , and construct the transmission signals on Alice and Bob’s ends;

[0040] Communication is performed based on the constructed transmission signals of Alice and Bob.

[0041] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0042] The present invention targets MIMO full-duplex communication scenarios. Each antenna on Alice's end and each antenna on Bob's end can simultaneously transmit and receive signals at the same frequency. Therefore, a transmission signal with artificial noise added is constructed for both Alice and Bob. The transmission signals are constructed in the same manner by Alice and Bob, and both contain artificial noise and valid signals. The artificial noise is used to improve the communication security performance of the system. Regarding the artificial noise and valid signal, the transmission channel is subjected to singular value decomposition. After obtaining the precoding matrices of the valid signal and artificial noise, the valid signal and noise signal of the corresponding communication end are constructed. The transmission signals for Alice and Bob are thus constructed. The power allocation coefficients of the artificial noise and valid signal, the signal dimension, and the power allocation matrix of the valid signal are used as decision variables. The constructed transmission signals at both ends are then transmitted simultaneously, and the received signals at both ends are obtained. The system average security rate is calculated based on the received signals. The decision variables in the transmission signal are optimized with the goal of maximizing the system average security rate, thereby optimizing the MIMO full-duplex secure communication parameters. In actual communication, constructing the transmission signal with optimized communication parameters can maximize the average confidentiality rate in the MIMO full-duplex communication system in full full-duplex mode, thereby improving confidentiality. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A flowchart of a method for optimizing parameters of MIMO full-duplex secure communication according to an embodiment of the present invention;

[0044] Figure 2 is a schematic diagram of a MIMO channel model in full full-duplex mode according to an embodiment;

[0045] Figure 3 is a variation of the average confidentiality rate of an embodiment with the effective signal dimension;

[0046] Figure 4 1 is a curve showing the average confidentiality rate and the power allocation coefficient changing with the transmit power under a fixed effective signal dimension in one embodiment, wherein (a) is the curve showing the average confidentiality rate changing with the transmit power, and (b) is the curve showing the power allocation coefficient changing with the transmit power;

[0047] Figure 5 1 is a performance comparison diagram of a full full-duplex mode and a transmit / receive antenna separation mode according to an embodiment. DETAILED DESCRIPTION

[0048] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0049] like Figure 1 FIG. 1 is a flowchart of a method for optimizing parameters of MIMO full-duplex secure communication in an embodiment, which mainly includes the following steps:

[0050] Step S100: Determine the number of antennas N at Alice, Bob, and Eve respectively. A 、N B 、N E , and the transmission power P of Alice and Bob A 、P B , where Alice and Bob are legitimate communication ends, and Eve is an eavesdropping end.

[0051] First, determine the current communication scenario, such as Figure 2 As shown in Figure 2, in this communication scenario, there are two legitimate full-duplex communication terminals, Alice and Bob, and one eavesdropping terminal, Eve. The number of antennas and total transmit power of each communication terminal are known.

[0052] Step S200: construct transmission signals at Alice and Bob respectively, wherein the transmission signal at Alice includes the valid signal s A and artificial noise z A , the transmitted signal at Bob's end includes the effective signal s B and artificial noise z B .

[0053] Construct the transmission signal x of Alice's end respectively A and the transmitted signal x on Bob's end B , specifically:

[0054] xA =s A +z A ;

[0055] x B =s B +z B ;

[0056] Among them, s A and z A are the effective signal and artificial noise at Alice’s end respectively; s B 、z B are the effective signal and artificial noise at Bob's end, respectively. That is, artificial noise is added to the transmitted signals at both Alice and Bob's ends, and the allocation of the effective signal and artificial noise at each communication end is the decision variable that needs to be optimized. This allocation of the effective signal and artificial noise at each communication end includes the allocation of power and signal dimensions.

[0057] For Alice, assume that the valid signal s A The power distribution coefficient is φ A , then the effective signal s A The transmission power is φ A P A , artificial noise z A The transmission power is (1-φ A )P A Each antenna on Alice's side corresponds to one dimension. Assuming that the effective signal s A The dimension is r b , then the artificial noise z A The dimension is N A -r b . And for r b There are effective signals, assuming that the power distribution matrix Σ A P A P A Perform power allocation, that is, the transmission power of each effective signal may be different, and the power allocation matrix Σ A is a diagonal matrix and the elements in the diagonal matrix represent the power allocation ratio of the effective signal in each dimension. At this time, the energy of the effective signal transmitted by Alice is And for N A -r b artificial noise, whose power is evenly distributed, Alice's artificial noise power (1-φ A )P A Evenly distributed among these N A -r b At this time, the energy of the artificial noise transmitted by Alice is is the unit diagonal matrix.

[0058] Similarly, for Bob, assuming the valid signal s B The power distribution coefficient is φ B , then the effective signal s B The transmission power is φ B P B , artificial noise z B The transmission power is (1-φ B )P B Each antenna on Bob's side corresponds to one dimension. Assuming that the effective signal s B The dimension is r a , then the artificial noise z B The dimension is N B -r a . And for r a There are effective signals, assuming that the power distribution matrix Σ B P B P B Perform power allocation, that is, the transmission power of each effective signal may be different, and the power allocation matrix Σ B is a diagonal matrix and the elements in the diagonal matrix represent the power allocation ratio of the effective signal in each dimension. At this time, the energy of the effective signal transmitted by Bob is For N B -r a artificial noise, whose power is evenly distributed, Alice's artificial noise power (1-φ B )P B Evenly distributed in N B -r a At this time, the energy of the artificial noise transmitted by Bob is is the unit diagonal matrix.

[0059] Step S300: Connect the channel H from Alice to Bob AB Perform singular value decomposition to obtain the valid signal s on Alice's side A The precoding matrix W A and artificial noise z A The precoding matrix Z A , the channel H from Bob to Alice BA Perform singular value decomposition to obtain the effective signal s on Bob's side B The precoding matrix W B and artificial noise z B The precoding matrix Z B .

[0060] After determining the distribution of the effective signal and artificial noise in the transmitted signal in step S200, it is necessary to construct the representation of the effective signal and artificial noise. This step is based on decomposing the transmitted signal and obtaining the corresponding precoding matrix for constructing a specific signal representation.

[0061] For Alice, the total signal space allocated to valid information is r. b , the spatial dimension allocated to artificial noise is N A -r b . Set channel H AB Perform singular value decomposition, let in, is an orthogonal unitary matrix, take out V A The first r b The columns serve as the basis of the effective signal space, that is, the precoding matrix W of the effective signal A =V A (:,1:r b ). V A The remaining Nr b The columns serve as the basis of the artificial noise signal space, that is, the precoding matrix Z of the artificial noise signal A =V A (:,r b +1:N). In this case, the precoding matrix is ​​related to the decision variable of the signal dimension.

[0062] Similarly, for Bob, the total signal space allocated to the effective information is r a , the spatial dimension allocated to artificial noise is N B -r a . Set channel H BA Perform singular value decomposition, let Remove V B The first r a The columns serve as the basis of the effective signal space, that is, the precoding matrix W of the effective signal B =V B (:,1:r a ). V B The remaining Nr a The columns serve as the basis of the artificial noise signal space, that is, the precoding matrix Z of the artificial noise signal B =V B (:,r a +1:N).

[0063] Step S400: Constructing a valid signal s based on Alice's precoding matrix A and artificial noise z A , construct the effective signal s based on the precoding matrix at Bob's endB and artificial noise z B .

[0064] Among them, for Alice:

[0065] s A =W A u A ;

[0066] z A =Z A m A ;

[0067] Then, the transmitted signal at Alice's end can be expressed as:

[0068] x A =s A +z A =W A u A +Z A m A ;

[0069] in, is the vector of valid information symbols sent by the legitimate communication terminal Alice, Represents the vector composed of artificial noise symbols sent by the legitimate communication terminal Alice.

[0070] Similarly, for Bob, there are:

[0071] s B =W B u B ;

[0072] z B =Z B m B ;

[0073] Then, the transmitted signal at Bob's end can be expressed as:

[0074] x B =s B +z B =W B u B +Z B m B ;

[0075] in, is the vector composed of valid information symbols of the legal communication terminal Bob, m B Represents the vector composed of artificial noise symbols sent by the legitimate communication end Bob.

[0076] Step S500: Alice and Bob transmit signals simultaneously, and obtain signals received by Alice, Bob, and Eve respectively.

[0077] Communicate with the transmission signal constructed in step S400 and obtain the received signals of three segments respectively. Define the legitimate communication terminals Alice and Bob, and the signals received by the eavesdropping terminal Eve as y A 、y B and y E , then according to the conventional method, the expressions of each receiving end can be analyzed as follows:

[0078] y A =H BA x B +η A +n A

[0079] =H BA (W B u B +Z B m B )+η A +n A

[0080] =H BA W B u B +H BA Z B m B +η A +n A

[0081] y B =H AB x A +η B +n B

[0082] =H AB (W A u A +Z A m A )+η B +n B

[0083] =H AB W A u A +H AB Z A m A +η B +n B

[0084] y E =GAE x A +G BE x B +n E

[0085] =G AE (W A u A +Z A m A )+G BE (W B u B +Z B m B )+n E

[0086] =G AE W A u A +G AE Z A m A +G BE W B u B +G BE Z B m B +n E

[0087] Among them, η A and η B They represent the interference terms caused by self-interference at Alice and Bob respectively, A 、n B They are transmission channels H BA 、H AB Additive Gaussian white noise, n E Transmission channel G AE , G BE Total additive white Gaussian noise.

[0088] For the signal y received by Alice A , due to r a <N A , the received signal contains the artificial noise term emitted by Bob: H BA Z B m B . This term will greatly affect the received signal-to-noise ratio at Alice's end. Therefore, in one embodiment, the received signal is preprocessed and the received signal vector is multiplied left by Then remove the front r a Valid information symbols, Indicates U BThrough this process, Alice's received signal can eliminate the artificial noise information transmitted by Bob. The received signal after processing is:

[0089]

[0090] in Indicates taking out the first r of the matrix a The column vectors of the columns form a new matrix, and since is an orthogonal unitary matrix, so Multiply left by n A does not change its statistical properties, so In this way, the artificial noise signal transmitted by Bob mixed in the signal received by Alice is filtered out.

[0091] Similarly, the signal received by Bob is also mixed with the artificial noise emitted by Alice. Therefore, in one embodiment, the received signal is also pre-processed on Bob's side, and the received signal vector is multiplied left by And remove the front r b Valid information symbols, Indicates U A The conjugate transpose of , the received signal after processing is:

[0092]

[0093] Step S600: Construct the system average confidentiality rate, and calculate the optimal power allocation coefficient φ with the goal of maximizing the system average confidentiality rate. A and φ B , power allocation matrix Σ A and Σ B and dimension r b and r a .

[0094] After receiving the signal at each end, the average confidentiality rate of the system can be constructed as:

[0095]

[0096] Among them, R A 、R B are the instantaneous confidentiality rates of Alice and Bob respectively, is the average confidentiality rate of Eve, [x] + This means that when the value of x is less than 0, the result is set to 0, which ensures that the average confidentiality rate of the system is not less than 0. The confidentiality rates of each end can be constructed in a conventional manner.

[0097] Specifically,

[0098]

[0099]

[0100]

[0101] Among them, σ A , σ B and σ E The ambient noise power at Alice, Bob, and Eve respectively, N A dimensional unit diagonal matrix, N B dimensional unit diagonal matrix and N E dimensional unit diagonal matrix; ρ is the signal-to-noise ratio loss constant, which is the residual self-interference η of Alice and Bob A and η B Appears as a constant in the transmission rate expressions of Alice and Bob; E x {*} means finding the average value of * with respect to x.

[0102] Furthermore, to facilitate the solution, the average eavesdropping rate at Eve is converted into the upper bound of the average eavesdropping rate through Jensen’s inequality. The upper bound expression after conversion is:

[0103]

[0104] in, is the upper bound of the average eavesdropping rate, G AE Z A =G AEZ , G BE Z B =G BEZ .

[0105] Furthermore, we can write the theorem based on the average rate of the MIMO system: The closed-form expression is:

[0106]

[0107] Where T3 is the number of non-zero diagonal elements in Φ3, Φ 31 is the number of non-zero diagonal elements in Φ3, T2 is the number of non-zero diagonal elements in Φ2, Φ 21 is the non-zero diagonal element in Φ2. C(n,p,Φ) is the ergodic formula for the channel capacity when the number of transmitting antennas n and the number of receiving antennas p and the covariance matrix of the transmitted signal is Φ. Φ3 and Φ2 are defined as follows:

[0108]

[0109]

[0110] Substituting the upper bound of the average eavesdropping rate into the formula of the system's average confidentiality rate, the problem of maximizing the system's average confidentiality rate is ultimately transformed into the problem of maximizing the lower bound of the system's average confidentiality rate.

[0111] After constructing the expression of the system's average confidentiality rate, the decision variables can be determined with the goal of maximizing the system's average confidentiality rate.

[0112] In one embodiment, the solution can be performed according to the following steps:

[0113] Step S610: Initialize decision variables;

[0114] Step S620: Based on the convex optimization concept, the CVX solver is used to obtain the optimal power allocation matrix under the current effective signal dimension and power allocation coefficient; the particle swarm algorithm is used to obtain the optimal power allocation coefficient under the current effective signal dimension and power allocation matrix;

[0115] Step S630: updating the optimal power allocation coefficient and power allocation matrix under the current effective signal dimension, and repeating step S620 until the lower bound of the system average confidentiality rate reaches the maximum, and then ending the iteration;

[0116] Step S640: Traverse each effective signal dimension scenario, repeat steps S620 to S630, and finally compare the performance to output the optimal power allocation coefficient φ A and φ B , power allocation matrix Σ A and Σ B and dimension r b and r a .

[0117] Furthermore, after determining the optimal communication system parameters, the optimal average confidentiality rate of the system can be obtained based on the optimal communication system parameters, thereby evaluating the confidentiality of the entire communication system.

[0118] It should be noted that the order of the steps in the above embodiments is only for illustrative purposes, but is not limited thereto. The order of execution of the steps can be flexibly adjusted without affecting the implementation of the solution.

[0119] Accordingly, the present application also protects a MIMO full-duplex secure communication method, in which the optimal power allocation coefficient φ under the current communication scenario is obtained according to the MIMO full-duplex secure communication parameter optimization method described above. A and φ B , power allocation matrix ΣA and Σ B and dimension r b and r a , and construct transmission signals for Alice and Bob. The constructed transmission signals have the same form as in the parameter optimization method. Communication is then performed based on the constructed transmission signals for Alice and Bob. Furthermore, during communication, after receiving the signal, the original received signal is processed using the method described above to remove artificial noise interference.

[0120] The following is a specific example. Alice and Bob have 10 antennas, each with a transmit power of 10W. Eve has 6 antennas, and the channel noise power is are all 1, and the signal-to-noise ratio loss constant ρ=0.5. Figure 3 The figure shows the change of the system average security rate with the effective signal dimension when the communication parameter optimization method of the present invention is optimized. When the determined signal dimension is different, the system average security rate will also be different. The dimension of the effective signal at each end corresponding to the maximum system average security rate can be obtained through the solution tool. Figure 4 The curves of average confidentiality rate and power allocation coefficient changing with transmission power under fixed effective signal dimension are shown. The selected signal dimension combination is Figure 3 The optimal effective signal dimension in , where (a) the average confidentiality rate of the system is affected by the signal transmission power, and (b) illustrates how the dimension of the effective signal at each end is affected by the signal transmission power. Figure 5 The figure shows a performance comparison between the full-duplex mode and the transmit-receive antenna separation mode under the optimal parameter settings for both Alice and Bob. It can be seen that the full-duplex mode has a 2.3 times confidentiality rate gain compared to the transmit-receive antenna separation mode, which illustrates the excellent performance of the present invention.

[0121] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A MIMO full-duplex secure communication parameter optimization method, characterized in that: include: Determine the number of antennas N on Alice, Bob, and Eve respectively A 、N B 、N E , and the transmission power P of Alice and Bob A 、P B , where Alice and Bob are legitimate communication terminals, and Eve is an eavesdropping terminal; Construct the transmission signals of Alice and Bob respectively, where the transmission signal of Alice includes the effective signal s A and artificial noise z A , effective signal s A The total power is φ A P A , dimension is r b , the power allocation matrix is ​​Σ A ; Artificial noise z A The total power is (1-φ A )P A , dimension is N A -r b And the power in each dimension is the same; the transmitted signal at Bob's end includes the effective signal s B and artificial noise z B , where the effective signal s B The total power is φ B P B , dimension is r a , the power allocation matrix is ​​Σ B ; Artificial noise z B The total power is (1-φ B )P B , dimension is N B -r a And the power in each dimension is the same, φ A and φ B are the power allocation coefficients at Alice and Bob respectively; The channel H from Alice to Bob is AB and the channel H from Bob to Alice BA Do singular value decomposition, in, is an orthogonal unitary matrix, with V A The first r b Column as valid signal s A The precoding matrix W A 、The rest N A -r b Column as artificial noise z A The precoding matrix Z A , in, is an orthogonal unitary matrix, with V B The first r a Column as valid signal s B The precoding matrix W B 、The rest N B -r a Column as artificial noise z B The precoding matrix Z B ; Construct effective signals s respectively A 、s B and artificial noise z A 、z B , where s A =W A u A , z A =Z A m A , s B =W B u B , z B =Z B m B ,u A 、m A is the valid information symbol vector and artificial noise symbol vector at Alice’s end, u B 、m B are the valid information symbol vector and artificial noise symbol vector at Bob's end; Alice and Bob transmit signals simultaneously, and obtain the signals received by Alice, Bob, and Eve respectively; Construct the system average confidentiality rate, and calculate the optimal power allocation coefficient φ with the goal of maximizing the system average confidentiality rate A and φ B , power allocation matrix Σ A and Σ B and dimension r b and r a .

2. The MIMO full-duplex secure communication parameter optimization method according to claim 1, wherein: The effective signal and artificial noise of each transmitted signal are processed by the mask beamforming method so that the receiving end can receive the original transmitted information.

3. The MIMO full-duplex secure communication parameter optimization method according to claim 1, wherein: Each element in the channel satisfies a cyclic complex Gaussian distribution with a mean of 0 and a variance of 1.

4. The MIMO full-duplex secure communication parameter optimization method according to claim 1, wherein: After receiving the signal, Alice and Bob process the original received signal to remove artificial noise interference in the signal.

5. The MIMO full-duplex secure communication parameter optimization method according to claim 4, wherein: The signals received by Alice, Bob, and Eve are y A 、y B and y E ,in, y A =H BA W B u B +H BA Z B m B +η A +n A ; y B =H AB W A u A +H AB Z A m A +η B +n B ; y E =G AE W A u A +G AE Z A m A +G BE W B u B +G BE Z B m B +n E ; Among them, G AE , G BE are the channels from Alice to Eve and from Bob to Eve respectively; η A and η B They represent the interference terms caused by self-interference at Alice and Bob respectively, A 、n B They are transmission channels H BA 、H AB Additive Gaussian white noise, n E Transmission channel G AE , G BE Total additive white Gaussian noise; Processing the original received signal, including: The signal y received by Alice A Left multiplication Take the first r a Column as processing result, Indicates U B The conjugate transpose of The signal y received by Bob B Left multiplication Take the first r b Column as processing result, Indicates U A The conjugate transpose of .

6. The MIMO full-duplex secure communication parameter optimization method according to claim 1, wherein: System average confidentiality rate Among them, R A 、R B are the instantaneous confidentiality rates of Alice and Bob respectively, is the average confidentiality rate of Eve, [x] + Indicates that when the value of x is less than 0, the result is set to 0.

7. The MIMO full-duplex secure communication parameter optimization method according to claim 6, wherein: Through Jensen's inequality, the average eavesdropping rate of Eve is converted into the upper bound optimization of the average eavesdropping rate, so that the problem of maximizing the average confidentiality rate of the system is transformed into the problem of maximizing the lower bound of the average confidentiality rate of the system.

8. The MIMO full-duplex secure communication parameter optimization method according to claim 7, wherein: The process of lower bounding the maximum system average confidentiality rate includes: Step S61: Setting the initial distribution coefficient φ A and φ B , power allocation matrix Σ A and Σ B ; Step S62: Based on the convex optimization concept, the CVX solver is used to obtain the optimal power allocation matrix under the current effective signal dimension and power allocation coefficient; the particle swarm algorithm is used to obtain the optimal power allocation coefficient under the current effective signal dimension and power allocation matrix; Step S63: Update the optimal power allocation coefficient and power allocation matrix under the current effective signal dimension, repeat step S62 until the lower bound of the system average confidentiality rate reaches the maximum, and then end the iteration; Step S64: Traverse each effective signal dimension scenario, repeat steps S62 to S63, and finally compare the performance to output the optimal power allocation coefficient φ A and φ B , power allocation matrix Σ A and Σ B and dimension r b and r a .

9. A MIMO full-duplex secure communication parameter optimization system, characterized in that: include: Scenario parameter input unit, used to input the number of antennas N on Alice, Bob, and Eve A 、N B 、N E , and the transmission power P of Alice and Bob A 、P B , where Alice and Bob are legitimate communication terminals, and Eve is an eavesdropping terminal; The transmission signal construction unit is used to construct the transmission signals of Alice and Bob respectively, wherein the transmission signal of Alice includes the valid signal s A and artificial noise z A , effective signal s A The total power is φ A P A , dimension is r b , the power allocation matrix is ​​Σ A ; Artificial noise z A The total power is (1-φ A )P A , dimension is N A -r b And the power in each dimension is the same; the transmitted signal at Bob's end includes the effective signal s B and artificial noise z B , where the effective signal s B The total power is φ B P B , dimension is r a , the power allocation matrix is ​​Σ B ; Artificial noise z B The total power is (1-φ B )P B , dimension is N B -r a And the power in each dimension is the same, φ A and φ B are the power allocation coefficients at Alice and Bob respectively; The channel decomposition unit is used to decompose the channel H from Alice to Bob. AB and the channel H from Bob to Alice BA Do singular value decomposition, in, is an orthogonal unitary matrix, with V A The first r b Column as valid signal s A The precoding matrix W A 、The rest N A -r b Column as artificial noise z A The precoding matrix Z A , in, is an orthogonal unitary matrix, with V B The first r a Column as valid signal s B The precoding matrix W B 、The rest N B -r a Column as artificial noise z B The precoding matrix Z B ; Valid / noise signal construction unit, used to construct valid signal s respectively A 、s B and artificial noise z A 、z B , where s A =W A u A , z A =Z A m A , s B =W B u B , z B =Z B m B ,u A 、m A is the valid information symbol vector and artificial noise symbol vector at Alice’s end, u B 、m B are the valid information symbol vector and artificial noise symbol vector at Bob's end; A signal receiving unit, configured to obtain the signals received by Alice, Bob, and Eve respectively after Alice and Bob transmit signals simultaneously; Solving unit, used to construct the system average confidentiality rate, with the goal of maximizing the system average confidentiality rate, and calculate the optimal power allocation coefficient φ A and φ B , power allocation matrix Σ A and Σ B and dimension r b and r a .

10. A MIMO full-duplex secure communication method, characterized in that: include: The MIMO full-duplex secure communication parameter optimization method according to any one of claims 1 to 8 obtains the optimal power allocation coefficient φ in the current communication scenario A and φ B , power allocation matrix Σ A and Σ B and dimension r b and r a , and construct the transmission signals on Alice and Bob’s ends; Communication is performed based on the constructed transmission signals of Alice and Bob.

Images