Intelligent reflecting surface assisted secure coexistence radio system design method

Abstract

The application discloses a smart reflecting surface assisted safe symbiotic radio system design method, which comprises the following steps: establishing a system and signal model, designing a user terminal receiving scheme; jointly designing a PT precoding vector and an RIS reflection coefficient, and proposing an optimization problem of minimizing PT transmission power under PU SNR and SU BER constraints; and converting the optimization problem and solving the optimization problem by adopting an AO-based optimization algorithm and an SDR technology. The method improves an existing RIS assisted SR network, jointly designs a PT precoding vector and an RIS phase shift coefficient, minimizes transmission power under PU and SU performance and RIS constant modulus constraints, and can realize the reduction of overall power consumption while meeting the performance of a wireless communication system and the guarantee of wireless communication system communication, thereby guaranteeing higher safety than a prior art scheme.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a design method for a smart reflector-assisted secure symbiotic radio system. Background Technology

[0002] Energy efficiency (EE) and spectrum efficiency (SE) are two key performance indicators for sixth-generation (6G) wireless communication networks due to the massive connectivity demands of future Internet of Things (IoT) scenarios such as smart cities and factories. Symbiotic radio (SR) based on reconfigurable intelligent surfaces (RIS) is a promising technology that achieves both high EE and SE. SR utilizes cognitive backscatter communication to achieve reciprocal spectrum sharing and highly reliable backscatter communication. SR consists of two subsystems: a primary system and a secondary system. In the primary system, the primary transmitter (PT) uses an active radio to send information to the primary user (PU), while in the secondary system, the RIS, acting as a secondary transmitter (ST), uses backscatter radio to send radio frequency (RF) signals to the secondary user (SU) by periodically switching load impedance.

[0003] Furthermore, RIS, as an innovative technology, can intelligently improve the communication environment and enhance signal transmission. In recent years, research on RIS-assisted SR networks has attracted increasing attention, as it not only improves the transmission performance of the PU but also enables SU communication. A common feature of existing technologies is that both the PU and SU employ Maximum Likelihood (ML) detection to receive information from the base station (BS) and RIS, thus placing certain demands on the receiver's computing power. The ML scheme cannot be used when the SU lacks sufficient computing power. Secondly, the ML scheme also requires the SU to have sufficiently high received signal power to correctly demodulate the main information; when the main information is not what the SU needs, transmission energy is wasted. Moreover, SU demodulation of the main information may also lead to potential information leakage from the PU, jeopardizing its communication security. Summary of the Invention

[0004] Based on the aforementioned background technology, this invention proposes a design method for a smart reflector-assisted secure coexistence radio system. It improves upon existing RIS-assisted SR networks by employing a hypothesis-testing-based receiving method at the SU end to receive only information from the RIS. This system not only supports RIS-enabled SU communication but also improves the security and energy efficiency of communication from the PT to the PU. The precoding vector of the PT and the passive beamforming of the RIS are jointly designed to minimize the PT transmit power under the QoS constraints of the PU and SU. The optimization problem is non-convex, and it is solved using an iterative algorithm based on alternating optimization (AO) and semidefinite relaxation (SDR). Simulation results verify the security performance of the proposed system and its superiority over traditional communication schemes.

[0005] A design method for a smart reflector-assisted safe symbiotic radio system includes the following steps:

[0006] Step 1: Establish system and signal models, and design user-end receiving schemes;

[0007] The system includes a main transmitter (PT), a reconfigurable smart surface (RIS), a single-antenna primary user (PU), and a secondary user (SU). The PT is equipped with one antenna, while the RIS is equipped with an independent passive reflection unit. The PT plans to transmit one data symbol to the PU, while the RIS, while assisting the PT in its transmission, transmits one data symbol to the SU using reflected signals. The sum of these two data symbols is the symbol set.

[0008] For the user-end receiving scheme, the PU end uses a maximum likelihood ML detector to decode the primary and secondary information. In step 1, for the user-end receiving scheme, the SU end uses a hypothesis testing-based receiving scheme to decode the secondary information.

[0009] Step 2: Jointly design the PT precoding vector and RIS reflection coefficient, and propose an optimization problem to minimize the PT transmit power under the constraints of PU signal-to-noise ratio (SNR) and SU bit error rate (BER).

[0010] Step 3: Transform the proposed optimization problem and solve it using an AO-based optimization algorithm and SDR technique;

[0011] The signal-to-noise ratio (SNR) constraint of PU and the bit error rate (BER) constraint of SU in the conversion optimization problem are transformed into two sub-problems, and the approximate optimal solution is obtained by solving the optimization problem.

[0012] The beneficial effects achieved by this invention are as follows:

[0013] (1) An improvement was made to the existing RIS-assisted SR network. By jointly designing the PT precoding vector and the RIS phase shift coefficient, the transmit power was minimized under the constraints of PU and SU performance and RIS constant mode. This can reduce the overall power consumption and ensure the communication of the wireless communication system while meeting the performance requirements of the wireless communication system.

[0014] (2) This system not only supports RIS-enabled SU communication, but also improves the energy efficiency of communication from PT to PU.

[0015] (3) It can ensure that the bit error rate of the SU (eavesdropper) is higher (i.e., the probability of decoding the received signal into the correct signal is lower), and it can ensure that the SNR of the eavesdropper is lower than the SNR required for demodulation (i.e., the received signal is very small, increasing its decoding error), thus ensuring higher security than existing technical solutions.

[0016] (4) As the number of passive reflection units increases, the required transmission power decreases, and the proposed algorithm outperforms existing solutions. Therefore, RIS can significantly improve system performance when equipped with a large number of components. Attached Figure Description

[0017] Figure 1 This is a basic flowchart illustrating a design method for a smart reflector-assisted secure symbiotic radio system provided in an embodiment of the present invention.

[0018] Figure 2 This is a system model diagram of a secure wireless communication system based on a smart reflector-assisted secure symbiotic radio system design method provided in an embodiment of the present invention.

[0019] Figure 3 This is a transmission frame structure diagram of a secure wireless communication system based on a smart reflector-assisted secure symbiotic radio system design method provided in an embodiment of the present invention.

[0020] Figure 4 This is a simulation coordinate diagram of the known statistical channel state information of the PT in a smart reflector-assisted safe symbiotic radio system design method provided in an embodiment of the present invention.

[0021] Figure 5 The diagram shows the known channel state information of the PT in a smart reflector-assisted safe symbiotic radio system design method provided in this embodiment of the invention, and the performance curves of transmit power, PU, ​​SU SNR and BER under different receive signal-to-noise ratios.

[0022] Figure 6The present invention provides a design method for a smart reflector-assisted safe symbiotic radio system, in which the PT has known channel state information, and the transmit power, SU SNR and BER performance curves at different distances between RIS and PT.

[0023] Figure 7 The present invention provides a design method for a smart reflector-assisted safe symbiotic radio system, in which the PT has known statistical channel state information, and transmit power, SU SNR and BER performance curves under different numbers of RIS components. Detailed Implementation

[0024] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings.

[0025] Reference Figures 1-3 As an embodiment of the present invention, a design method for a smart reflector-assisted safe symbiotic radio system is provided, the specific steps of which are as follows:

[0026] S1: Establish system and signal models, and design user-end receiving schemes;

[0027] SNR stands for Signal to Noise Ratio, which is the ratio of the strength of the received useful signal to the strength of the received interference signal (noise).

[0028] BER stands for Bit Error Rate.

[0029] CG stands for Complex Gaussian.

[0030] PT stands for Primary Transmitter.

[0031] ST stands for Secondary Transmitter.

[0032] RIS stands for Reconfigurable Intelligent Surface.

[0033] PU stands for Primary User.

[0034] SU stands for Secondary User.

[0035] The RIS-assisted security SR system includes a PT, RIS, a single-antenna PU, and a SU wireless network, where the PT is equipped with N... t The antenna is equipped with N. rThis invention utilizes an independent passive reflection unit (PT). By jointly designing the precoding vector of the PT and the passive beamforming of the RIS, it minimizes the PT transmit power under the different performance characteristics of the PU and SU, and the constant mode constraint of the RIS. Since the SU can employ the ML scheme to eavesdrop on PU information, making it impossible to guarantee the security of PU information, this invention, based on a hypothesis testing-based receiving scheme at the SU end, combines the RIS with a symbiotic wireless communication system security approach. For multiple non-convex constraints in the optimization problem, this invention uses an iterative optimization method to obtain a heuristic solution. The AO algorithm is first used to solve for the precoding vector under the constraints of PU SNR and SU BER. The optimal solution to the original problem can be constructed from the relaxation problem solution. Then, the phase shift coefficient is optimized by maximizing the SU BER. The phase shift coefficient is approximated by using a positive semi-definite relaxation and Gaussian randomization method to solve for the reflection coefficient of the RIS. Finally, the optimal solution is obtained through repeated iterative optimization.

[0036] Establish system and signal models, and design reception schemes for different user receivers: This system includes a PT, RIS, a single-antenna PU, and a SU, where the PT is equipped with N t The antenna is equipped with N. r An independent passive reflection unit, such as Figure 2 As shown. PT plans to transmit LT data symbols to PU. While assisting PT transmission, RIS transmits L data symbols to SU using reflected signals. Therefore, this network is often referred to as a symbiotic radio system, where s l (t) and c l Time relationship Figure 3 As shown, one sub-symbol period consists of T main symbol periods. and It is a set of symbols.

[0037] like Figure 2 As shown, the channel between PT and RIS is The channel between PT and PU is The channel between PT and SU is The channel between RIS and PU is The channel between RIS and SU is The channel follows quasi-static block fading during the LT coherence time and is known at the PT end; the RIS reflection coefficient is expressed as:

[0038]

[0039] in Represents the field of complex numbers, θ n ∈[0, 2π] represents the reflection angle of the nth reflecting unit (n∈1, ..., N).r If the received signals at PU and SU can be represented as follows:

[0040]

[0041]

[0042] in,{·} H denoted as Hermitian transpose; W is the PT transmit precoding vector; Φ = diag{φ} is the RIS reflection coefficient correlation matrix. It is the transmitted signal at PT, and RIS transmits it through c. l Controlling the reflection channel parameters enables symbol transmission; n p (t) and n s (t) is a value with zero mean and power δ. 2 Additive white Gaussian noise.

[0043] The master information is decoded using a maximum likelihood (ML) detector in the PU. and secondary information c l ,Right now:

[0044]

[0045] in Represents vector [s] l , ..., s l (T), c l ] T The estimated value. For ease of representation, assume... Thus, the PU and SU decoders under the ML scheme are obtained. l The average signal-to-noise ratio (SNR) of (t) is

[0046]

[0047] A hypothesis-based reception scheme is adopted at the SU end: To introduce a hypothesis-based SU reception scheme, it is assumed that the RIS uses binary phase-shift keying modulation (BPSK), thus... Specifically, if the RIS transmits a symbol "0", it will adjust its impedance so that very little of the incident signal is reflected; while if it wants to transmit a "1", some of the incident signal will be reflected. According to c l The value is either 0 or 1, and the receiver z l (t) can be represented as and Two hypotheses

[0048]

[0049] For unknown PU signalsl (t), which is approximated as a Gaussian distribution with zero mean and 1 variance at the SU end, thus obtaining z l (t) follows a Gaussian distribution in:

[0050]

[0051] Will with c l The relevant received signal is represented as vector z l =[z l (1), ..., z l (T)] T .when When they are independent, z l The elements are also independent of each other, at this time

[0052] In ML, c l The detection can be achieved through the following likelihood ratio test:

[0053]

[0054] Where Z = ||z l || 2 The energy for receiving signals, Indicates the received signal z l In the assumption The probability density function (PDF) is given below. Based on the above tests, when T is relatively large, the closed-form solution of the bit error rate (BER) can be approximated as:

[0055]

[0056] in min(·) represents taking the minimum value and sum. `max(·)` represents taking the maximum value. Based on the above detection process, SU does not need to demodulate s. l (t).

[0057] S2: Jointly design the PT precoding vector w and RIS reflection coefficient φ, and propose an optimization problem to minimize the PT transmit power under PU SNR and SUBER constraints. The problem is formulated as follows:

[0058]

[0059] stγ p ≥r p ,

[0060] P b ≤Λs ,

[0061]

[0062] The first constraint is the SNR constraint of PU, Γ p The second constraint is the BER constraint of SU, set as the PU SNR threshold. s To limit the bit error rate threshold, the third constraint is the unit constant modulus constraint of RIS. "min" indicates the minimization operation; "st" indicates the constraint condition, φ n Denotes the nth reflection coefficient, ||·|| 2 Represents the square norm.

[0063] S3: Transform the proposed optimization problem and solve it using AO-based optimization algorithms and SDR techniques; [The remaining text appears to be incomplete and requires further context.] l Substituting ∈{0,1} into the PU SNR constraint, the constraint can be written as:

[0064] γ p =w H R h w / (2δ 2 )≥Γ p ,

[0065] in

[0066] Further transform the Sub-BER constraints:

[0067] when At that time, the SU bit error rate constraint is:

[0068]

[0069] in And at this time

[0070] when Similarly, we can obtain

[0071]

[0072] in And at this time

[0073] Therefore, according to and The relationship between the magnitudes of these factors allows us to transform the problem of minimizing the PT transmit power into the following two problems. Let {w} i , φ i Let}, i = 1, 2, represent two sets of solutions to the following two optimization problems, respectively:

[0074]

[0075] stw H R h w≥2δ 2 Γ p ,

[0076] w H R g w≥(λ s -1)δ 2 ,

[0077]

[0078]

[0079] stw H R h w≥2δ 2 Γ p ,

[0080]

[0081]

[0082] Where "arg min" represents the minimization operation. Then, when the objective function value ||w1|| 2 <||w2|| 2 When the condition is met, the final solution to the problem is {w1, φ1}; otherwise, it is {w2, φ2}. The solution to the two problems described above will be discussed below. Since they have the same form, only one problem will be discussed.

[0083] The optimization problem is solved using an AO-based optimization algorithm. The specific steps are as follows:

[0084] Update w: Update w by solving the following optimization problem:

[0085]

[0086] st tr(R h W)≥2δ 2 r p ,

[0087] tr(R g W)≥(λ s -1)δ 2 ,

[0088] rank(W) = 1.

[0089] Here, "tr" represents the trace of the matrix, and "rank" represents the rank of the matrix. Relaxing the rank-1 constraint to W≥0 (this constraint relaxation is a positive semidefinite relaxation technique, i.e., SDR, which is also used in the following problem), the problem is transformed into a convex semidefinite programming (SDP) problem, which can be solved efficiently using interior-point methods such as CVX. Since the original problem contains only two inequality constraints, the optimal solution w of the original problem can be constructed from the solution of the relaxed problem through rank-1 decomposition.

[0090] Update φ: Update φ by solving the following optimization problem.

[0091] When w is given, the objective function value is determined, so only one feasible solution for φ needs to be found; however, this may lead to slow iterative convergence. Therefore, the SU BER constraint is transformed into an objective function, and a SU receiver performance optimization problem under PU performance constraints is established, i.e.

[0092]

[0093]

[0094]

[0095] in Let G r1 =diag(g r1 ), and use the equation Φg r1 =G r1 φ, the objective function can be rewritten as:

[0096]

[0097] in Similarly, define H r1 =diag(h r1 The constraint can be rewritten as:

[0098]

[0099] in

[0100] Therefore, the subproblem of optimizing the RIS reflection coefficient can be expressed as:

[0101]

[0102]

[0103]

[0104]

[0105] Where "max" represents the optimization operation, and "diag" represents the diagonal of the matrix. Relaxing the rank-1 constraint to... The relaxed problem becomes a convex SDP problem, which can be solved using the interior-point method. Based on the optimal solution of the relaxed problem, an approximate optimal solution to the original problem can be obtained using the Gaussian randomization method. (The obtained approximate optimal solution represents the final radio system parameters.)

[0106] Reference Figures 4-7 This paper presents a verification test of a design method for a safe symbiotic radio system assisted by an intelligent reflector, which verifies and explains the technical effects of the method and demonstrates its actual effectiveness.

[0107] In this method, since the optimization problem of minimizing PT transmit power is non-convex, an alternating optimization (AO) method is heuristically employed. This involves sequentially fixing one of the precoding vector w and the RIS reflection coefficient φ while updating the other. The AO algorithm is summarized in the table below.

[0108]

[0109] To verify the performance of the above secure reception scheme, simulation experiments were conducted using MATLAB for the corresponding scenarios, and the CVX software package was used to solve the optimization problem.

[0110] The simulation was programmed using MATLAB software, and the specific parameters used in the simulation are given in the detailed implementation. Transmit power, signal-to-noise ratio (SNR), and bit error rate (BER) were used as performance metrics. For each simulation scenario, 100 channels were generated to obtain the average performance results.

[0111] The simulation design is as follows: Number of PT antennas N t =4; Number of RIS reflection units N r =80; the user channel conforms to the Ricean channel model. Noise power δ 2 = -70dBm; PU signal-to-noise ratio requirement set to Γ p =10dB; SU bit error rate requirement set to P b =10 -2 ;Transmission block length T = 50;Symbol s l (t) represents a Complex Gaussian (CG) signal. PT is located at (0,0), and the transmit antenna angle is π / 4. RIS is located at (30,0), and the transmit antenna angle is 3π / 4. PU is located at (25,15), and SU is located at (30,10), as shown below. Figure 4 As shown. The path loss model for all channels follows the following model:

[0112] η(d)=C0(d / d0) -α

[0113] Where C0 is the path loss when the reference distance D0 = 1; d represents the link distance; and α represents the path loss exponent. The remaining channels are generated in the same manner. α will be used. PR α Rpu α Rsu α Ppu and α Psu Let represent the path loss exponents of the PT to RIS, RIS to PU, RIS to SU, PT to PU, and PT to SU channels, respectively. In the simulation below, C0 = -20dB and α are set to... PR =2.2, α Rpu =α Rsu =α Ppu =α Psu =2.8. The channel G from PT to RIS is given by the following formula:

[0114]

[0115] Where κ is the Rice factor, κ=1 represents a pure line-of-sight channel, κ=0 represents a Rayleigh fading channel, and G LOS and G NLOS These represent the line-of-sight channel and the Rayleigh fading channel, respectively. κ PR κ Rpu κ Rsu κ Ppu 、 and κ Psu These represent the Rice factors of the PT to RIS, RIS to PU, RIS to SU, PT to PU, and PT to SU channels, respectively.

[0116] In the simulation below, κ is set PR =κ Rpu =k Rsu =0.7, κ Ppu =κ Rsu =0.7. The convergence threshold for the AO algorithm is set to ∈≤10. -3 .

[0117] To better evaluate the performance of the proposed algorithm, consider the contrast scheme without subsystems: In this scheme, there is no subsystem transmission, and the optimal precoding vector at the PTx end can be obtained by solving the following optimization problem:

[0118]

[0119] stγ p ≥Γ p ,

[0120] Additionally, "CG-T" is the theoretical value calculated by the formula. The rest represent the actual bit error rate detected. Transmit power, signal-to-noise ratio (SNR), and bit error rate (BER) are used as performance indicators. For each simulation scenario, 100 channels are generated to obtain the average performance result.

[0121] Let P b =10 -2 ,like Figure 5 As shown in the top left figure, the transmit power varies with the required PU SNR by Γ p The increase in SNR with increasing PU SNR, and the superior performance of the proposed algorithm compared to the comparative scheme, demonstrates that introducing RIS is beneficial for improving the system's energy efficiency. The lower left figure shows the actual signal-to-noise ratio (SNR) of the user as a function of PU SNR requirements. p The relationship between the required signal-to-noise ratio (SNR) and the actual SNR of the PU (Power Injection Unit) is as follows: the actual SNR of the SU (Supply Injection Unit) meets the required SNR requirement, while the actual SNR of the SU (Supply Injection Unit) increases with the required SNR requirement, but its SNR remains lower than that of the PU's demodulated main signal, thus verifying the security of the proposed system. The right figure shows that the SU's received BER changes with Γ. p The relationship between actual and theoretical BER changes with Γ p The actual BER is basically consistent with the theoretical BER as the SU SNR increases. In addition, the actual BER was compared when the main signal was QPSK, 16QAM and 64QAM. Their actual BER performance was better than that of the complex Gaussian signal (CG). The performance improvement was more obvious as the SU SNR increased. Among them, the QPSK signal had the lowest randomness, so its BER performance was the best.

[0122] Let Γ p =10dB,P b =10 -2 ,like Figure 6 The top left figure shows the relationship between transmit power and the distance between RIS and PT, as follows: Figure 5 The distance between the RIS and PT was increased from 30m to 50m. As shown in the figure, the required transmit power increases with the increase in RIS distance. Within a certain distance, the comparative scheme consumes more power than the proposed scheme. However, at shorter RIS distances, power consumption increases significantly due to the need to simultaneously serve the SU. The lower left figure shows the relationship between the actual SU SNR and distance. The actual SU SNR decreases continuously with increasing distance and fails to meet the demodulated main signal SNR requirement. The right figure shows the relationship between the SU received BER and distance. Both the actual and theoretical BER show an increasing trend with increasing distance, with the actual BER slightly better than the theoretical BER. Furthermore, the actual BER was compared when the main signal was QPSK, 16QAM, and 64QAM. Their actual BER performance was better than that of the complex Gaussian signal (CG), with QPSK having the lowest randomness, thus exhibiting the best BER performance.

[0123] Let Γ p=10dB,P b =10 -2 ,like Figure 7 The top left figure shows the relationship between transmit power, RIS, and the number of components N. r The relationship between them, N r Increase from 40 to 160. As seen in the left graph, with N... r With the increase of N, the required transmit power decreases, and the proposed algorithm outperforms the comparative scheme. Therefore, RIS significantly improves system performance even with a large number of components. The lower left figure shows the actual SNR of SU as a function of N. r The relationship between the changes in SU's actual SNR and N r It increases with increasing N, but when N increases r When sufficiently large, its signal-to-noise ratio (SNR) tends to level off and remains consistently below the actual SNR requirement of the PU, ensuring the security of the proposed system. The right figure shows that the SU's received BER varies with N. r The relationship between N and its changes, both in practice and in theory. r The actual BER decreases as the main signal increases, and it basically matches the theoretical BER. In addition, the actual BER was compared when the main signal was QPSK, 16QAM, and 64QAM. Their actual BER performance was better than CG. Among them, QPSK signal has the lowest randomness, so its BER performance is the best.

[0124] In RIS-assisted SR, a common feature of existing technologies is that both the PU and SU ends employ maximum likelihood detection (ML) to receive information from the base station (BS) and RIS, thus placing certain demands on the receiver's computing power. When the SU lacks sufficient computing power, the ML scheme cannot be used. Secondly, the ML scheme also requires the SU to have sufficiently high received signal power to correctly demodulate the main information; when the main information is not what the SU needs, transmission energy is wasted. Furthermore, SU demodulation of the main information may lead to potential information leakage from the PU, jeopardizing its communication security. Based on existing RIS-assisted SR networks, this system uses ML at the PU end to receive information from the base station (PT) and RIS, while at the SU end, a hypothesis-based reception method is used to receive only information from the RIS. This system not only supports RIS-enabled SU communication but also improves the security and energy efficiency of communication from the PT to the PU.

[0125] This paper proposes a method for jointly designing the precoding vector of the phototransistor (PT) and the passive beamforming of the reflector beamforming (RIS), minimizing the PT transmit power under QoS constraints of the phototransistor (PU) and reflector beamforming (SU). The proposed problem is non-convex. To address this, the objective problem is first transformed into solving the precoding vector under PU SNR and SU BER constraints using the AO algorithm. The optimal solution to the original problem can be constructed from the relaxation problem solution. Then, the phase shift coefficient is optimized by maximizing the SU BER. The phase shift coefficient is approximated by solving the RIS reflection coefficient using a positive semi-definite relaxation and Gaussian randomization method. Simulation results verify the effectiveness of the proposed method.

[0126] This solution can guarantee higher security than existing technologies. For example, it can ensure that the bit error rate of the SU (eavesdropper) is higher (i.e., the probability of decoding the received signal into the correct signal is lower), and it can ensure that the SNR of the eavesdropper is lower than the SNR required for demodulation (i.e., the received signal is very small, increasing its decoding error), thereby ensuring security.

[0127] The above description is only a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. Any equivalent modifications or changes made by those skilled in the art based on the content disclosed in the present invention should be included within the scope of protection set forth in the claims.

Claims

1. A design method for a smart reflector-assisted secure symbiotic radio system, characterized in that: The method includes the following steps: Step 1: Establish system and signal models, and design user-end receiving schemes; The system includes a main transmitter (PT), a reconfigurable smart surface (RIS), a single-antenna main user unit (PU), and... Secondary user SU, of which PT is equipped with The RIS is equipped with an antenna. A separate passive reflection unit; PT plans to transmit to PU Data symbols While assisting PT transmission, RIS transmits signals to SU via reflected signals. Data symbols , and A set of symbols; For the user-side receiving scheme, the PU end uses a maximum likelihood ML detector to decode the master information. and secondary information In step 1, for the user-end receiving scheme, the SU end uses a hypothesis testing-based receiving scheme to decode the secondary information. : Step 2, Jointly design PT precoding vectors and RIS reflection coefficient An optimization problem is proposed to minimize the transmit power of the PT under the constraints of PU signal-to-noise ratio (SNR) and SU bit error rate (BER). In step 2, the PT precoding vector is jointly designed. and RIS reflection coefficient An optimization problem is proposed to minimize the PT transmit power under the constraints of PU SNR and SU BER, which is expressed as: The first constraint is the SNR constraint of PU. The first constraint is the set PU SNR threshold, and the second constraint is the SU BER constraint. To limit the bit error rate threshold, the third constraint is the constant modulus constraint of RIS; min indicates the minimization operation; st represents the constraint condition. Indicates the first Reflectance coefficient Represents the square of the norm; Step 3: Transform the proposed optimization problem and solve it using an AO-based optimization algorithm and SDR technique; The signal-to-noise ratio (SNR) constraint and the bit error rate (BER) constraint in the conversion optimization problem are transformed into two sub-problems, and the approximate optimal solution is obtained by solving the optimization problem. In step 3, Substituting the PU SNR constraint, the PU SNR constraint is transformed into: in , For the channel between PT and RIS, This is the channel between PT and PU. For the channel between RIS and PU, The power of additive white Gaussian noise; In step 3, the Sub-BER constraints are transformed into: when At that time, the SU bit error rate constraint is: in And at this time , and For signal Gaussian distribution The parameters in For the channel between PT and SU, This is the channel between RIS and SU; when At that time, we obtained: in And at this time ; according to and Based on the relationship between the magnitudes, the problem of minimizing the PT transmit power is transformed into the following two problems; let... These can be represented as two sets of solutions to the following two optimization problems, namely: Where arg min represents the minimization operation. The number of independent passive reflection units equipped in the RIS; then when the objective function value At that time, the final solution to the problem is Otherwise, for ; Given that the transmit power (PT) is known in the CSI, the problem of finding a heuristic solution, i.e., minimizing the transmit power, is formulated as follows: The solution is obtained using an AO-based optimization algorithm and SDR technique. The specific steps are as follows: renew Solving the following optimization problem for Update: Where tr represents the trace of the matrix and rank represents the rank of the matrix; the rank-1 constraint is relaxed to The problem is then transformed into a convex semidefinite programming problem (SDP), which can be solved using the interior-point method. Since the original problem contains only two inequality constraints, the optimal solution to the original problem can be constructed from the solution of the relaxation problem through rank-one decomposition. ; renew Update by solving the following optimization problem : The SU BER constraint is transformed into an objective function, and a SU receiver performance optimization problem under PU performance constraints is established, namely: in , make , and using equations The objective function is rewritten as: in ;definition The constraints are rewritten as follows: in ; Therefore, the subproblem of optimizing the RIS reflection coefficient is expressed as: Where max represents the optimization operation, and diag represents the diagonal of the matrix; the rank-1 constraint is relaxed to The relaxed problem becomes a convex SDP problem, which is solved using the interior point method. Based on the optimal solution of the relaxation problem, an approximate optimal solution to the original problem is obtained using the Gaussian randomization method.

2. The design method for a smart reflector-assisted secure symbiotic radio system according to claim 1, characterized in that: In step 1, the channel between PT and RIS is The channel between PT and PU is The channel between PT and SU is The channel between RIS and PU is The channel between RIS and SU is , Representing the complex field, the channel is in The RIS reflection coefficients are known at the PT end and follow quasi-static block fading during the coherence time; the RIS reflection coefficients are expressed as: in For the first Reflection angle of each reflecting unit The signals received by PU and SU are then represented as follows: in, Indicates Hermitian transpose; This is the PT transmit precoding vector; It is the RIS reflection coefficient correlation matrix; It is the transmitted signal at PT, RIS through By controlling the parameters of the reflection channel, symbol transmission can be achieved; and The mean is zero and the power is Additive white Gaussian noise.

3. The design method for a smart reflector-assisted secure symbiotic radio system according to claim 2, characterized in that: In step 1, the PU end uses a maximum likelihood ML detector to decode the master information. and secondary information ,Right now: in Representing vectors The estimated value; let Thus, PU decoding under the ML scheme is obtained. The average signal-to-noise ratio (SNR) is: 。 4. The design method for a smart reflector-assisted secure symbiotic radio system according to claim 3, characterized in that: In step 1, the SU end uses a hypothesis testing-based receiving scheme to decode the secondary information. : To introduce a hypothesis-test-based SU receiver scheme, we assume that the RIS uses binary phase-shift keying modulation (BPSK), thus... ;according to Equal to 0 or 1, receive signal Represented as and Two assumptions, namely For unknown PU signals Approximating it as a Gaussian distribution with zero mean and 1 variance at the SU end, we obtain... Follows a Gaussian distribution in: will with The relevant received signal is represented as a vector. ;when When they are independent, The elements are also independent of each other, at this time ; In ML, The detection is achieved through the following likelihood ratio test: in The energy to receive signals, Indicates received signal In the assumption The probability density function PDF is as follows; based on the above tests, when the relative value is large At that time, the closed-form solution for the bit error rate (BER) is: in , and Based on the above detection process, SU does not require demodulation. .

Images