Compromise design method of communication perception integrated waveform

Abstract

The invention relates to the technical field of wireless communication, and discloses a compromise design method for a communication-sensing integrated waveform, which comprises the following steps: acquiring a communication channel matrix at a transmitting end; splitting the total power according to the communication perception performance compromise coefficient to obtain power for optimizing perception and power for optimizing communication; maximizing the reachable sum rate by taking the optimized sensing power as a constraint to obtain a first frequency domain auto-covariance matrix; minimizing a Cramer-Rao lower bound by taking optimized communication power as a constraint to obtain a second time domain auto-covariance matrix; establishing an optimization problem of nonlinear programming; and combining the first frequency domain auto-covariance matrix and the second time domain auto-covariance matrix in a frequency domain to obtain a final transmitting signal frequency domain auto-covariance matrix, and deriving to obtain a final transmitting signal frequency domain precoding matrix by using a water injection algorithm. According to the method, the communication performance and the radar sensing performance are improved, and the technical problems that an existing waveform design method is imperfect in optimization and insufficient in consideration are solved.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a compromise design method for a communication-sensing integrated waveform. Background Art

[0002] The large-scale commercial use of the fifth generation mobile communication system (5G) is accelerating the transformation of the economy and society towards digitalization, networking, and intelligence, and is driving the network into the era of the Internet of Everything. However, due to the explosive growth of communication devices, wireless spectrum resources are becoming increasingly tight. In order to meet the demand for additional spectrum resources, industry and academia are exploring how to integrate perception functions represented by traditional radars into communication signals, so that communication and perception can share the same spectrum resources.

[0003] At present, the research direction of waveform design for communication perception integration is mainly divided into two modes. The first mode is called coexistence, in which the radar and communication transceiver, although sharing the same platform, will regard each other as interference. Researchers use multiplexing technologies such as space division multiplexing, time division multiplexing, frequency division multiplexing, etc. to reduce the interference between the two subsystems. However, these methods have an obvious disadvantage, that is, the two systems cannot work in the same domain, which will cause a huge waste of resources.

[0004] The second mode is called collaborative design, also known as DFRC (Dual Functional Radar Communication). This method designs modern integrated technologies such as signals, waveforms, and coding into a complete joint unit. Compared with coexistence waveforms, DFRC waveforms have a higher degree of integration and are therefore more promising for research in both academia and industry. However, most designs of DFRC waveforms use a single-path flat fading channel model, which is not applicable to the actual complex multipath fading channel environment. Orthogonal frequency division multiplexing (OFDM), due to its robustness to multipath fading, is one of the key technologies in 5G mobile communications. In addition, OFDM waveforms have also been shown to be equally useful for radar detection. Therefore, research on the use of OFDM waveforms for communication-perception integration is growing.

[0005] However, most of these studies do not take advantage of the collaboration between receiving antennas, but instead process the signals received by each antenna separately, thus losing the diversity gain between antennas. In addition, for the performance trade-off between communication and perception, most studies use relatively complex optimization methods that cannot meet the low latency requirements of 5G communications. Summary of the invention

[0006] The present invention provides a compromise design method for a communication-perception integrated waveform, which improves communication performance and radar perception performance, so as to solve the technical problems of imperfect optimization and insufficient consideration of existing waveform design methods.

[0007] The present invention provides a compromise design method for a communication-sensing integrated waveform, based on a communication-sensing integrated system, wherein the communication-sensing integrated system includes a transmitting base station, a single multi-antenna user and a sensing target; the method includes:

[0008] S1. Determine parameter data of the communication sensing integrated system, calculate the antenna steering vector, and obtain the communication channel matrix at the transmitting end;

[0009] S2. Setting the total power of the transmitted signal, and determining a communication perception performance compromise coefficient, splitting the total power according to the communication perception performance compromise coefficient to obtain a power for optimizing perception and a power for optimizing communication;

[0010] S3, taking the optimized perceived power as a constraint, establishing a communication optimization problem, maximizing the reachability and rate of the communication system, and obtaining the corresponding frequency domain autocovariance matrix of the transmitted signal through a water filling algorithm, recorded as the first frequency domain autocovariance matrix;

[0011] S4. Taking the power of the optimized communication as a constraint, a perception optimization problem is established to minimize the Cramer-Laur lower bound of the detection target angle estimation, and obtain the corresponding time domain autocovariance matrix of the transmitted signal, which is recorded as the second time domain autocovariance matrix;

[0012] S5. According to the first frequency domain autocovariance matrix and the second time domain autocovariance matrix, a nonlinear programming optimization problem is established with maximizing the reachable sum rate as the optimization goal and the total power of the transmitted signal as the constraint condition;

[0013] S6, using the optimization problem of the nonlinear programming described in the semi-positive definite programming algorithm, and according to the optimization result, merging the first frequency domain autocovariance matrix and the second time domain autocovariance matrix in the frequency domain to obtain a final transmit signal frequency domain autocovariance matrix;

[0014] S7. According to the final frequency domain autocovariance matrix of the transmitted signal, a water filling algorithm is used to derive a final frequency domain precoding matrix of the transmitted signal.

[0015] Furthermore, the step S1 specifically includes:

[0016] Determine the parameter data of the communication sensing integrated system; wherein the parameter data includes the number of transmitting antennas, the distance between adjacent antennas, the radar carrier frequency, the number of receiving antennas, and the power of the base station transmitting signal;

[0017] The transmitter sends a pilot signal, and the receiver calculates the channel matrix through the received signal and the pilot signal, and feeds it back to the transmitter through the control link. In the frequency domain, the channel matrix corresponding to the kth subcarrier is H(k), and the signal at the communication receiving end is expressed as:

[0018] Y(k)=H(k)X(k)+Z(k)

[0019] =H(k)W(k)S(k)+Z(k),k=0,1,...,N-1

[0020] in, is the frequency domain channel matrix corresponding to the kth subcarrier, is a covariance matrix The frequency domain Gaussian noise vector, is of rank N R The identity matrix of ;

[0021] In terms of radar perception, the echo signal received by the receiving end after the transmitting target is reflected can be expressed as:

[0022] y r (n) = ζa(θ)a H (θ)x(n)+z r (n) = ζA(θ)x(n) + z r (n)

[0023] Among them, ζ is the reflection coefficient, which contains the radar cross section information of the target, z r (n) is the Gaussian noise vector, A(θ) = a(θ)a(θ) H , θ is the target’s arrival angle, and the guidance vector a(θ) can be expressed as:

[0024]

[0025] where the center of the ULA antenna is chosen as the reference point and Δ is the normalized antenna spacing.

[0026] Furthermore, the step S2 specifically includes:

[0027] The total power of the transmitted signal is split to optimize perception and communication respectively:

[0028]

[0029] Among them, P T is the total power of the transmitted signal, ρ is the communication perception performance trade-off coefficient, P1 is the power used to optimize perception, and P2 is the power used to optimize communication.

[0030] Furthermore, the step S3 specifically includes:

[0031] Taking P1 in step S2 as the power constraint and maximizing the reachability and rate of the communication system as the goal, according to the calculation formula for maximizing reachability and transmission, the communication optimization problem is established as follows:

[0032]

[0033] Constraints:

[0034]

[0035] The optimization problem is solved by the traditional water filling algorithm, and the results are recorded as the precoding matrix W(k) and the power allocation matrix P S,1 (k), according to the traditional water injection algorithm, W1(k) is expressed as:

[0036] W1(k)=V1(k)

[0037] in, is the singular value decomposition of H(k), and the solution to the optimization problem is the first frequency domain autocovariance matrix R f,1 (k) has:

[0038] Furthermore, the maximum achievable sum rate of the communication system can be expressed as:

[0039]

[0040] Wherein, max{·} represents the maximum value, Δf is the subcarrier spacing, B is the bandwidth of the OFDM system, det(·) represents the determinant of the matrix, and I represents the identity matrix.

[0041] Furthermore, the step S4 specifically includes:

[0042] Taking P2 in step S2 as the power constraint and minimizing the Cramer-Rao lower bound of the arrival angle estimation of the detected target as the goal, according to the Cramer-Rao lower bound calculation formula, the perception optimization problem is established as follows:

[0043]

[0044] Constraint 1:

[0045] tr(R)=P2

[0046] Constraint 2:

[0047] R=R H

[0048] This problem is a convex problem, and the solution to this problem can be easily obtained using convex optimization tools, which is denoted as the second time-domain autocovariance matrix R2.

[0049] Furthermore, the Cramer-Rao lower bound calculation formula is:

[0050]

[0051] Among them, μ is a constant vector, and the derivative of the intermediate variable is tr(·) is the matrix trace operator, which converts A(θ)=a(θ)a(θ) H Substituting into the above equation, we get:

[0052]

[0053] in, is the derivative of a(θ).

[0054] Furthermore, the step S5 specifically includes:

[0055] The first frequency domain autocovariance matrix R f,1 (k) is combined with the second time domain autocovariance matrix R2, and according to the corresponding relationship of the autocovariance matrix in the time-frequency domain, R2 is distributed to the frequency domain according to the set ratio to generate the second frequency domain autocovariance matrix, which is then combined with R f,1 (k) corresponding addition;

[0056] Let the second frequency domain autocovariance matrix R f (k) = ω k N 2 R d ,in, And k ≥0; Taking maximizing communication reachability and transmission rate as the optimization goal, the optimization problem established by taking the allocation ratio coefficient of R2 in the frequency domain as the optimization object is as follows:

[0057]

[0058] Constraint 1:

[0059]

[0060] Constraint 2:

[0061] ω k ≥0

[0062] Where ω=[ω0,ω1,...,ω N-1 ] T ,This problem is a nonlinear optimization problem and is solved by the ,semi-positive definite programming method.

[0063] Furthermore, the step S6 specifically includes:

[0064] The optimization objective function is defined as:

[0065]

[0066] The constraint function is defined as:

[0067]

[0068] g k (ω)=-ω k =-e k ω,k=0,1,...,N-1

[0069] Optimization problem:

[0070]

[0071] Constraint 1:

[0072] v(ω)=0

[0073] Constraint 2:

[0074] g k (ω)≤0,k=0,1,...,N-1

[0075] Furthermore, the nonlinear function f(ω) can be obtained by (m) The Taylor expansion is transformed into a quadratic function, and the objective function is given by the following formula:

[0076]

[0077] Define η = ω-ω (m) , the above optimization problem is further transformed into:

[0078]

[0079] Constraint 1:

[0080] 1η=1-1ω (m)

[0081] Constraint 2:

[0082] -e k η≤e k ω (m) ,k=0,1,...,N-1

[0083] The above optimization problem is a standard quadratic programming problem about η. It is solved by Lagrangian operator method. The optimal ω is substituted into R f,2 (k) = ω k N 2 R2, solve the optimal R f,2 (k).

[0084] Furthermore, the step S7 specifically includes:

[0085] The water injection algorithm is used to obtain the precoding matrix W1(k) with the best communication performance under the constraint of power limit P1, and the power allocation matrix P S,1 (k), and the corresponding first frequency domain autocovariance matrix R f,1 (k), and the relationship between them is

[0086] The second frequency domain autocovariance matrix R f,2 (k) Satisfy:

[0087]

[0088] Among them, W2(k), P S,2 (k) corresponds to W1(k) of the above relationship, P S,1 (k);

[0089] because Setting P S,2 (k) is:

[0090]

[0091] Define P°(k)=(P S,1 (k)) 1 / 2 , substitute into the above two formulas, and R f,2 (k) can be derived as:

[0092]

[0093] Assume that N R <N T , R f,2 The sizes of (k), W2(k), and P°(k) are N respectively. T ×N T , N T ×N R and N R ×N T Therefore, according to the above derived formula, the approximate solution of the communication precoding matrix W2(k) is derived as follows:

[0094]

[0095] Where Q(k)D(k)Q -1 (k) = ρ / (1-ρ)R f,2 (k) is R f,2 (k) eigenvalue decomposition;

[0096] Finally, the frequency domain precoding matrix in the optimization target is:

[0097] The present invention also provides a compromise design device for a communication-sensing integrated waveform, based on a communication-sensing integrated system, the communication-sensing integrated system comprising a transmitting base station, a single multi-antenna user and a sensing target; the device comprises:

[0098] An acquisition module, used to determine parameter data of the communication sensing integrated system, calculate the antenna's steering vector, and acquire a communication channel matrix at the transmitting end;

[0099] A splitting module, used to set the total power of the transmitted signal and determine the communication perception performance compromise coefficient, and split the total power according to the communication perception performance compromise coefficient to obtain the power for optimizing perception and the power for optimizing communication;

[0100] A first establishment module is used to establish a communication optimization problem with the optimized perceived power as a constraint, maximize the reachability and rate of the communication system, and obtain a corresponding frequency domain autocovariance matrix of the transmitted signal through a water injection algorithm, which is recorded as a first frequency domain autocovariance matrix;

[0101] The second establishment module is used to establish a perception optimization problem with the power of the optimized communication as a constraint, minimize the Cramer-Laur lower bound of the detection target angle estimation, and obtain the corresponding transmission signal time domain autocovariance matrix, which is recorded as the second time domain autocovariance matrix;

[0102] A third establishment module is used to establish an optimization problem of nonlinear programming based on the first frequency domain autocovariance matrix and the second time domain autocovariance matrix, with maximizing the reachable sum rate as the optimization goal and the total power of the transmitted signal as the constraint condition;

[0103] A merging module, used for using the optimization problem of the nonlinear programming described in the semi-positive definite programming algorithm, and merging the first frequency domain autocovariance matrix and the second time domain autocovariance matrix in the frequency domain according to the optimization result to obtain a final frequency domain autocovariance matrix of the transmitted signal;

[0104] The derivation module is used to derive the final frequency domain precoding matrix of the transmitted signal according to the final frequency domain autocovariance matrix of the transmitted signal by using a water injection algorithm.

[0105] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0106] The present invention also provides a computer-readable storage medium on which a computer program is stored, and the computer program implements the steps of the above method when executed by a processor.

[0107] The beneficial effects of the present invention are:

[0108] The present invention regards the Cramer-Rao bound of the perception system and the reachability and transmission rate of the communication system as performance indicators, studies the performance trade-off between the communication system and the perception system, and proposes corresponding optimization problems and solution algorithms. In the waveform design method of minimizing the Cramer-Rao bound and maximizing the reachability and transmission rate, the present invention proposes an algorithm based on water-injected sequential quadratic programming to solve the nonlinear programming optimization problem, derives the covariance matrix and precoding matrix of the transmitted signal, and can have good communication performance and perception performance at the same time, while greatly improving the communication performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0109] Figure 1 It is a flow chart of the compromise design method of the communication-sensing integrated waveform of the present invention.

[0110] Figure 2 It is a structural diagram of the communication perception integration system in the present invention.

[0111] Figure 3 The figure is a schematic diagram comparing the relationship between the reachability and rate and the signal-to-noise ratio of the method of the present invention and other strategies.

[0112] Figure 4 It is a schematic diagram comparing the relationship between the reachability and rate and the change of the compromise factor under the method of the present invention and other strategies.

[0113] Figure 5 It is a schematic diagram of the change of the overall objective of the method of the present invention as the number of iterations increases.

[0114] Figure 6 The final radar beam diagrams of different tradeoff factors in the method of the present invention.

[0115] Figure 7 It is a structural schematic diagram of the compromise design device of the communication-sensing integrated waveform of the present invention.

[0116] Figure 8 It is a schematic diagram of the internal structure of the computer device in the present invention.

[0117] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0118] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0119] Based on the radar communication integrated waveform design method with performance compromise, a single-user multi-antenna model is established to address the problem that existing studies have not utilized the collaborative cooperation between receiving antennas, and reachable transmission is used as the performance indicator of the communication system. The present invention focuses on the performance trade-off between the radar system and the communication system. In the waveform design method of minimizing the Cramer-Rao bound and maximizing the reachable transmission, the corresponding nonlinear programming optimization problem is proposed. The present invention proposes a new algorithm based on water-injected sequential quadratic programming to derive the covariance matrix and precoding matrix of the transmitted signal. Finally, MATLAB is used to verify and analyze the performance of the communication perception integrated system.

[0120] like Figure 1 As shown, the present invention provides a compromise design method for a communication-sensing integrated waveform, based on a communication-sensing integrated system, the communication-sensing integrated system includes a transmitting base station, a single multi-antenna user and a sensing target; the method includes:

[0121] S1. Determine parameter data of the integrated communication and sensing system, calculate the antenna's steering vector, and obtain a communication channel matrix at the transmitting end.

[0122] like Figure 2 As shown, the communication and perception integrated system includes a transmitting base station, a single multi-antenna user and a perception target. T =20, the transmitting antenna array is a uniform linear array, the spacing between adjacent antennas is 0.5 wavelength, the radar carrier frequency is 5.7GHz, the number of subcarriers used by OFDM is N=16, the detection target is a point target, and the direction is 0°. The number of communication target receiving antennas is N R = 10. The power of the base station's transmitted signal is P T The limit is 1, and the signal-to-noise ratio at the communication receiving end is 20dB. The communication channel satisfies the standard complex Gaussian distribution.

[0123] The signal transmitted by the base station meets the following requirements:

[0124]

[0125] in,(·) T represents the matrix transpose, N is the number of discrete Fourier transform points, x i (n) is the signal transmitted by the i-th antenna in the time domain, which can be expressed as:

[0126]

[0127] Among them, X i (k) is the frequency domain signal corresponding to the k-th subcarrier of the ith antenna, and e is the natural base.

[0128] The frequency domain transmitted signal vector corresponding to the kth subcarrier can be expressed as:

[0129] X(k)=W(k)S(k) (3)

[0130] in, is the precoding matrix, is the sending information flow vector.

[0131] The autocovariance matrix of the transmitted signal in the time domain can be expressed as:

[0132]

[0133] Where E{·} represents the expectation of a random variable, (·) H is the conjugate transpose operator, the diagonal matrix is the power of the i-th information stream at the k-th subcarrier, and diag{·} is a diagonal operator. The frequency domain covariance matrix of the transmitted signal at the k-th subcarrier is defined as have:

[0134] R f (k) = E{X(k)X * (k)}=W(k)P S (k)W H (k) (5)

[0135] It is related to the time domain autocovariance matrix as follows:

[0136]

[0137] The transmitter sends a pilot signal, and the receiver calculates the channel matrix through the received signal and the pilot signal, and feeds it back to the transmitter through the control link. In the frequency domain, the channel matrix corresponding to the kth subcarrier is H(k), and the signal at the communication receiving end can be expressed as:

[0138]

[0139] in, is the frequency domain channel matrix corresponding to the kth subcarrier, is a covariance matrix The frequency domain Gaussian noise vector, is of rank N R For the communication system in the example, the maximum achievable sum rate of the system can be expressed as:

[0140]

[0141] Wherein, max{·} represents the maximum value, Δf is the subcarrier spacing, B is the bandwidth of the OFDM system, det(·) represents the determinant of the matrix, and I represents the identity matrix.

[0142] In terms of radar perception, the echo signal received by the receiving end after the transmitting target is reflected can be expressed as:

[0143] y r (n) = ζa(θ)a H (θ)x(n)+z r (n) = ζA(θ)x(n) + z r (n) (9)

[0144] Among them, ζ is the reflection coefficient, which contains the radar cross section information of the target, z r (n) is the Gaussian noise vector, A(θ) = a(θ)a(θ) H , θ is the target’s arrival angle, and the guidance vector a(θ) can be expressed as:

[0145]

[0146] where the center of the ULA antenna is chosen as the reference point and Δ is the normalized antenna spacing.

[0147] The Cramer-Rao lower bound for angle estimation can be expressed as:

[0148]

[0149] Among them, μ is a constant vector, which is related to the radar perception signal-to-noise ratio, and the derivative of the intermediate variable tr(·) is the matrix trace operator. H Substituting into the above equation, we get:

[0150]

[0151] in, is the derivative of a(θ).

[0152] S2. Set the total power of the transmitted signal and determine the communication perception performance compromise coefficient, split the total power according to the communication perception performance compromise coefficient to obtain the power for optimizing perception and the power for optimizing communication.

[0153] The total transmit power is split into two parts for optimization of perception and communication:

[0154]

[0155] Among them, P Tis the total power of the transmitted signal, ρ is the trade-off factor, P1 is the power used to optimize perception, and P2 is the power used to optimize communication.

[0156] S3. Taking the optimized perceived power as a constraint, establish a communication optimization problem to maximize the reachability and rate of the communication system, and obtain the corresponding frequency domain autocovariance matrix of the transmitted signal through a water injection algorithm, which is recorded as the first frequency domain autocovariance matrix.

[0157] Taking P1 in step S2 as the power constraint condition and maximizing the reachability and transmission rate of the communication system as the goal, according to the maximization reachability and transmission calculation formula (8), the optimization problem is established as follows:

[0158] Optimization Problem 1:

[0159]

[0160] Constraints:

[0161]

[0162] Optimization problem 1 can be solved by the traditional water filling algorithm, and the results are recorded as the precoding matrix W(k) and the power allocation matrix P S,1 (k), according to the traditional water injection algorithm, W1(k) is given by:

[0163] W1(k)=V1(k) (16)

[0164] in, is the singular value decomposition of H(k). The solution to the corresponding optimization problem 1 is the first frequency domain autocovariance matrix R f,1 (k) has:

[0165]

[0166] S4. Taking the power of the optimized communication as a constraint, establish a perception optimization problem, minimize the Cramer-Laur lower bound of the detection target angle estimation, and obtain the corresponding time domain autocovariance matrix of the transmitted signal, which is recorded as the second time domain autocovariance matrix.

[0167] Taking P2 in step S2 as the power constraint condition and minimizing the Cramer-Rao lower bound of the arrival angle estimation of the detected target as the goal, according to the Cramer-Rao lower bound calculation formula (12), the optimization problem is established as follows:

[0168] Optimization Problem 2:

[0169]

[0170] Constraint 1:

[0171] tr(R)=P2 (19)

[0172] Constraint 2:

[0173] R=R H (20)

[0174] This problem is a convex problem, and the solution to this problem can be easily obtained using convex optimization tools, which is denoted as the second time-domain autocovariance matrix R2.

[0175] S5. According to the first frequency domain autocovariance matrix and the second time domain autocovariance matrix, a nonlinear programming optimization problem is established with maximizing the achievable sum rate as the optimization goal and the total power of the transmitted signal as the constraint condition.

[0176] Step S3 obtains the first frequency domain autocovariance matrix R with communication performance as the optimization target f,1 (k), step S4 obtains the second time domain autocovariance matrix R2 with the perception performance as the optimization target. In order to realize the integrated system, they need to be placed in the same domain for merging. According to the corresponding relationship of the autocovariance matrix in the time-frequency domain of formula (6), R2 can be allocated to the frequency domain according to a certain proportion to generate the second frequency domain autocovariance matrix, and then combined with R f,1 (k) Add the corresponding items.

[0177] Let the second frequency domain autocovariance matrix R f (k) = ω k N 2 R d ,in, And k ≥0, is the allocation ratio coefficient. In this example, the optimization goal is to maximize the communication achievable transmission rate, and the allocation ratio coefficient of R2 in the frequency domain is the optimization object to establish the optimization problem as follows:

[0178] Optimization Problem 3:

[0179]

[0180] Constraint 1:

[0181]

[0182] Constraint 2:

[0183] ω k ≥0 (23)

[0184] Where ω=[ω0,ω1,...,ω N-1 ] T ,This problem is a nonlinear optimization problem and can be solved by semi-positive definite programming method.

[0185] S6. The optimization problem of the nonlinear programming described in the semi-positive definite programming algorithm is adopted, and according to the optimization result, the first frequency domain autocovariance matrix and the second time domain autocovariance matrix are merged in the frequency domain to obtain the final frequency domain autocovariance matrix of the transmitted signal.

[0186] First, define the optimization objective function as:

[0187]

[0188] The constraint function is defined as:

[0189]

[0190] Optimization Problem 4:

[0191]

[0192] Constraint 1:

[0193] v(ω)=0 (27)

[0194] Constraint 2:

[0195] g k (ω)≤0,k=0,1,...,N-1 (28)

[0196] Furthermore, the nonlinear function f(ω) can be obtained by (m) The Taylor expansion is transformed into a quadratic function, and the objective function is given by the following formula:

[0197]

[0198] Define η = ω-ω (m) , the optimization problem 4 is further transformed into:

[0199] Optimization Problem 5:

[0200]

[0201] Constraint 1:

[0202] 1η=1-1ω (m) (31)

[0203] Constraint 2:

[0204] -e k η≤e k ω (m) ,k=0,1,...,N-1 (32)

[0205] Optimization problem 5 is a standard quadratic programming problem about η, which can be solved using the Lagrangian operator method. The iterative steps to solve this problem are summarized in Table 1:

[0206] Table 1

[0207]

[0208]

[0209] After solving this problem, substitute the optimal ω into R f,2 (k) = ω k N 2 R2 can solve the optimal R f,2 (k).

[0210] S7. According to the final frequency domain autocovariance matrix of the transmitted signal, a water filling algorithm is used to derive a final frequency domain precoding matrix of the transmitted signal.

[0211] According to step S3, the precoding matrix W1(k) and the power allocation matrix P1 that optimize the communication performance under the constraint of the power limit P1 can be obtained by the water injection algorithm. S,1 (k), and the corresponding first frequency domain autocovariance matrix R f,1 (k), and the relationship between them is shown in formula (17).

[0212] Similarly, the second frequency domain autocovariance matrix R f,2 (k) It should meet the following requirements:

[0213]

[0214] Where W2(k), P S,2 (k) corresponds to W1(k) in formula (17), P S,1 (k).

[0215] because In this example, set P S,2 (k) is:

[0216]

[0217] Define P°(k)=(P S,1 (k)) 1 / 2 , substitute formula (34) into formula (33), and R f,2 (k) can be derived as:

[0218]

[0219] Note that in this invention, it is assumed that N R <N T, R f,2 The sizes of (k), W2(k), and P°(k) are N respectively. T ×N T , N T ×N R and N R ×N T Therefore, according to formula (35), the approximate solution of the communication precoding matrix W2(k) can be derived as:

[0220]

[0221] Where Q(k)D(k)Q -1 (k) = ρ / (1-ρ)R f,2 (k) is R f,2 Eigenvalue decomposition of (k)

[0222] Finally, the frequency domain precoding matrix in the optimization objective is designed as:

[0223]

[0224] The present invention uses MATLAB to verify that the achievable transmission rate of the system is maximized.

[0225] As Figure 3 As shown in Figure 2, the user's achievable sum rate increases with the increase of the signal-to-noise ratio. When the signal-to-noise ratio is fixed, the larger the ρ, the higher the achievable sum rate. By adopting the solution of the precoding matrix design proposed in equation (37), the achievable rate only experiences a slight performance loss compared to the ideal optimal result.

[0226] like Figure 4 As shown, with the increase of the compromise coefficient, the achievable rate of the waveform design strategy proposed by the present invention continues to increase. In addition, with the increase of the compromise factor, the communication performance of the strategy continues to improve.

[0227] like Figure 5 As shown, it can be seen that for each compromise factor, the objective function decreases as the number of iterations increases, and the termination condition is almost met within 40 iterations. For different compromise factors. As the compromise factor increases, the convergence curve decreases faster and the objective function approaches a smaller value. This verifies the feasibility of the optimization scheme proposed in the present invention, and a good performance compromise can be achieved.

[0228] like Figure 6 As shown in FIG. 1 , when the compromise factor is equal to 0, the performance compromise waveform design strategy proposed by the present invention obtains the same beam pattern as the "strict" beam pattern. At the same time, as the compromise factor increases, the height of the waveform main lobe becomes lower and lower, while the height of the side lobe becomes higher and higher, and the mismatch between the obtained beam pattern and the "strict" beam pattern begins to increase.

[0229] Therefore, the present invention can clearly sense and distinguish the echoes of different moving targets while communicating with multi-antenna user targets, which not only greatly improves the user reachability and transmission rate, but also accelerates the convergence speed of the algorithm.

[0230] like Figure 7 As shown, the present invention also provides a compromise design device for a communication-sensing integrated waveform, based on a communication-sensing integrated system, the communication-sensing integrated system includes a transmitting base station, a single multi-antenna user and a sensing target; the device includes:

[0231] Acquisition module 1, used to determine parameter data of the communication sensing integrated system, calculate the steering vector of the antenna, and obtain the communication channel matrix at the transmitting end;

[0232] Splitting module 2, used to set the total power of the transmitted signal and determine the communication perception performance compromise coefficient, and split the total power according to the communication perception performance compromise coefficient to obtain the power for optimizing perception and the power for optimizing communication;

[0233] A first establishment module 3 is used to establish a communication optimization problem with the optimized perceived power as a constraint, maximize the reachability and rate of the communication system, and obtain a corresponding frequency domain autocovariance matrix of the transmitted signal through a water injection algorithm, which is recorded as a first frequency domain autocovariance matrix;

[0234] The second establishment module 4 is used to establish a perception optimization problem with the power of the optimized communication as a constraint, minimize the Cramer-Laur lower bound of the detection target angle estimation, and obtain the corresponding transmission signal time domain autocovariance matrix, which is recorded as the second time domain autocovariance matrix;

[0235] A third establishment module 5 is used to establish a nonlinear programming optimization problem based on the first frequency domain autocovariance matrix and the second time domain autocovariance matrix, with maximizing the reachable sum rate as the optimization goal and the total power of the transmitted signal as the constraint condition;

[0236] A merging module 6 is used to adopt the optimization problem of the nonlinear programming described in the semi-positive definite programming algorithm, and according to the optimization result, merge the first frequency domain autocovariance matrix and the second time domain autocovariance matrix in the frequency domain to obtain the final transmission signal frequency domain autocovariance matrix;

[0237] The derivation module 7 is used to derive the final frequency domain precoding matrix of the transmitted signal according to the final frequency domain autocovariance matrix of the transmitted signal by using a water filling algorithm.

[0238] In one embodiment, the acquisition module 1 specifically includes:

[0239] Determine the parameter data of the communication sensing integrated system; wherein the parameter data includes the number of transmitting antennas, the distance between adjacent antennas, the radar carrier frequency, the number of receiving antennas, and the power of the base station transmitting signal;

[0240] The transmitter sends a pilot signal, and the receiver calculates the channel matrix through the received signal and the pilot signal, and feeds it back to the transmitter through the control link. In the frequency domain, the channel matrix corresponding to the kth subcarrier is H(k), and the signal at the communication receiving end is expressed as:

[0241] Y(k)=H(k)X(k)+Z(k)

[0242] =H(k)W(k)S(k)+Z(k),k=0,1,...,N-1

[0243] in, is the frequency domain channel matrix corresponding to the kth subcarrier, is a covariance matrix The frequency domain Gaussian noise vector, I NR is of rank N R The identity matrix of ;

[0244] In terms of radar perception, the echo signal received by the receiving end after the transmitting target is reflected can be expressed as:

[0245] y r (n) = ζa(θ)a H (θ)x(n)+z r (n) = ζA(θ)x(n) + z r (n)

[0246] Among them, ζ is the reflection coefficient, which contains the radar cross section information of the target, z r (n) is the Gaussian noise vector, A(θ) = a(θ)a(θ) H , θ is the target’s arrival angle, and the guidance vector a(θ) can be expressed as:

[0247]

[0248] where the center of the ULA antenna is chosen as the reference point and Δ is the normalized antenna spacing.

[0249] In one embodiment, the splitting module 2 specifically includes:

[0250] The total power of the transmitted signal is split to optimize perception and communication respectively:

[0251]

[0252] Among them, P Tis the total power of the transmitted signal, ρ is the communication perception performance trade-off coefficient, P1 is the power used to optimize perception, and P2 is the power used to optimize communication.

[0253] In one embodiment, the first establishing module 3 specifically includes:

[0254] Taking P1 in step S2 as the power constraint and maximizing the reachability and rate of the communication system as the goal, according to the calculation formula for maximizing reachability and transmission, the communication optimization problem is established as follows:

[0255]

[0256] Constraints:

[0257]

[0258] The optimization problem is solved by the traditional water filling algorithm, and the results are recorded as the precoding matrix W(k) and the power allocation matrix P S,1 (k), according to the traditional water injection algorithm, W1(k) is expressed as:

[0259] W1(k)=V1(k)

[0260] in, is the singular value decomposition of H(k), and the solution to the optimization problem is the first frequency domain autocovariance matrix R f,1 (k) has:

[0261] In one embodiment, the maximum achievable sum rate of the communication system can be expressed as:

[0262]

[0263] Wherein, max{·} represents the maximum value, Δf is the subcarrier spacing, B is the bandwidth of the OFDM system, det(·) represents the determinant of the matrix, and I represents the identity matrix.

[0264] In one embodiment, the second establishing module 4 specifically includes:

[0265] Taking P2 in step S2 as the power constraint and minimizing the Cramer-Rao lower bound of the arrival angle estimation of the detected target as the goal, according to the Cramer-Rao lower bound calculation formula, the perception optimization problem is established as follows:

[0266]

[0267] Constraint 1:

[0268] tr(R)=P2

[0269] Constraint 2:

[0270] R=R H

[0271] This problem is a convex problem, and the solution to this problem can be easily obtained using convex optimization tools, which is denoted as the second time-domain autocovariance matrix R2.

[0272] In one embodiment, the Cramer-Rao lower bound calculation formula is:

[0273]

[0274] Among them, μ is a constant vector, and the derivative of the intermediate variable is tr(·) is the matrix trace operator, which converts A(θ)=a(θ)a(θ) H Substituting into the above equation, we get:

[0275]

[0276] in, is the derivative of a(θ).

[0277] In one embodiment, the third establishing module 5 specifically includes:

[0278] The first frequency domain autocovariance matrix R f,1 (k) is combined with the second time domain autocovariance matrix R2, and according to the corresponding relationship of the autocovariance matrix in the time-frequency domain, R2 is distributed to the frequency domain according to the set ratio to generate the second frequency domain autocovariance matrix, which is then combined with R f,1 (k) corresponding addition;

[0279] Let the second frequency domain autocovariance matrix R f (k) = ω k N 2 R d ,in, And k ≥0; Taking maximizing communication reachability and transmission rate as the optimization goal, the optimization problem established by taking the allocation ratio coefficient of R2 in the frequency domain as the optimization object is as follows:

[0280]

[0281] Constraint 1:

[0282]

[0283] Constraint 2:

[0284] ω k ≥0

[0285] Where ω=[ω0,ω1,...,ω N-1 ] T,This problem is a nonlinear optimization problem and is solved by the ,semi-positive definite programming method.

[0286] In one embodiment, the merging module 6 specifically includes:

[0287] The optimization objective function is defined as:

[0288]

[0289] The constraint function is defined as:

[0290]

[0291] g k (ω)=-ω k =-e k ω,k=0,1,...,N-1

[0292] Optimization problem:

[0293]

[0294] Constraint 1:

[0295] v(ω)=0

[0296] Constraint 2:

[0297] g k (ω)≤0,k=0,1,...,N-1

[0298] Furthermore, the nonlinear function f(ω) can be obtained by (m) The Taylor expansion is transformed into a quadratic function, and the objective function is given by the following formula:

[0299]

[0300] Define η = ω-ω (m) , the above optimization problem is further transformed into:

[0301]

[0302] Constraint 1:

[0303] 1η=1-1ω (m)

[0304] Constraint 2:

[0305] -e k η≤e k ω (m) ,k=0,1,...,N-1

[0306] The above optimization problem is a standard quadratic programming problem about η. It is solved by Lagrangian operator method. The optimal ω is substituted into R f,2 (k) = ω k N 2 R2, solve the optimal R f,2 (k).

[0307] In one embodiment, the derivation module 7 specifically includes:

[0308] The water injection algorithm is used to obtain the precoding matrix W1(k) with the best communication performance under the constraint of power limit P1, and the power allocation matrix P S,1 (k), and the corresponding first frequency domain autocovariance matrix R f,1 (k), and the relationship between them is

[0309] The second frequency domain autocovariance matrix R f,2 (k) Satisfy:

[0310]

[0311] Among them, W2(k), P S,2 (k) corresponds to W1(k) of the above relationship, P S,1 (k);

[0312] because Setting P S,2 (k) is:

[0313]

[0314] Define P°(k)=(P S,1 (k)) 1 / 2 , substituting into the above two formulas, and Rf,2(k) can be derived as:

[0315]

[0316] Assume that N R <N T , R f,2 The sizes of (k), W2(k), and P°(k) are N respectively. T ×N T , N T ×N R and N R ×N T Therefore, according to the above derived formula, the approximate solution of the communication precoding matrix W2(k) is derived as follows:

[0317]

[0318] Where Q(k)D(k)Q -1 (k) = ρ / (1-ρ)R f,2 (k) is the eigenvalue decomposition of Rf,2(k);

[0319] Finally, the frequency domain precoding matrix in the optimization target is:

[0320] The above modules are used to execute the corresponding steps in the compromise design method of the above communication-aware integrated waveform. The specific implementation method thereof is described in the above method embodiment and will not be repeated here.

[0321] like Figure 8 As shown, the present invention also provides a computer device, which can be a server, and its internal structure can be as shown in Figure 8 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store all data required for the process of the compromise design method of the communication-aware integrated waveform. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the compromise design method of the communication-aware integrated waveform is implemented.

[0322] Those skilled in the art will understand that Figure 8 The structure shown in is merely a block diagram of a portion of the structure related to the present application solution and does not constitute a limitation on the computer device to which the present application solution is applied.

[0323] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the compromise design method for any one of the above-mentioned communication-awareness integrated waveforms is implemented.

[0324] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0325] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0326] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A compromise design method for a communication-sensing integrated waveform, characterized in that: Based on a communication-sensing integrated system, the communication-sensing integrated system includes a transmitting base station, a single multi-antenna user and a sensing target; the method includes: S1. Determine parameter data of the communication sensing integrated system, calculate the antenna steering vector, and obtain the communication channel matrix at the transmitting end; S2. Setting the total power of the transmitted signal, and determining a communication perception performance compromise coefficient, splitting the total power according to the communication perception performance compromise coefficient to obtain a power for optimizing perception and a power for optimizing communication; S3, taking the optimized perceived power as a constraint, establishing a communication optimization problem, maximizing the reachability and rate of the communication system, and obtaining the corresponding frequency domain autocovariance matrix of the transmitted signal through a water filling algorithm, recorded as the first frequency domain autocovariance matrix; S4. Taking the power of the optimized communication as a constraint, a perception optimization problem is established to minimize the Cramer-Laur lower bound of the detection target angle estimation, and obtain the corresponding time domain autocovariance matrix of the transmitted signal, which is recorded as the second time domain autocovariance matrix; S5. According to the first frequency domain autocovariance matrix and the second time domain autocovariance matrix, a nonlinear programming optimization problem is established with maximizing the reachable sum rate as the optimization goal and the total power of the transmitted signal as the constraint condition; S6, using the optimization problem of the nonlinear programming described in the semi-positive definite programming algorithm, and according to the optimization result, merging the first frequency domain autocovariance matrix and the second time domain autocovariance matrix in the frequency domain to obtain a final transmit signal frequency domain autocovariance matrix; S7. According to the final frequency domain autocovariance matrix of the transmitted signal, a water filling algorithm is used to derive a final frequency domain precoding matrix of the transmitted signal.

2. The compromise design method for communication-awareness integrated waveform according to claim 1, characterized in that: The step S1 specifically includes: Determine the parameter data of the communication sensing integrated system; wherein the parameter data includes the number of transmitting antennas, the distance between adjacent antennas, the radar carrier frequency, the number of receiving antennas, and the power of the base station transmitting signal; The transmitter sends a pilot signal, and the receiver calculates the channel matrix through the received signal and the pilot signal, and feeds it back to the transmitter through the control link. In the frequency domain, the channel matrix corresponding to the kth subcarrier is H(k), and the signal at the communication receiving end is expressed as: Y(k)=H(k)X(k)+Z(k) =H(k)W(k)S(k)+Z(k),k=0,1,...,N-1 in, is the frequency domain channel matrix corresponding to the kth subcarrier, is a covariance matrix The frequency domain Gaussian noise vector, is of rank N R The identity matrix of ; In terms of radar perception, the echo signal received by the receiving end after the transmitting target is reflected can be expressed as: y r (n)=ζa(θ)a H (θ)x(n)+z r (n)=ζA(θ)x(n)+z r (n) Among them, ζ is the reflection coefficient, which contains the radar cross section information of the target, z r (n) is the Gaussian noise vector, A(θ) = a(θ)a(θ) H , θ is the target’s arrival angle, and the guidance vector a(θ) can be expressed as: where the center of the ULA antenna is chosen as the reference point and Δ is the normalized antenna spacing.

3. The compromise design method for communication-awareness integrated waveform according to claim 2, characterized in that: The step S2 specifically includes: The total power of the transmitted signal is split to optimize perception and communication respectively: Among them, P T is the total power of the transmitted signal, ρ is the communication perception performance trade-off coefficient, P1 is the power used to optimize perception, and P2 is the power used to optimize communication.

4. The compromise design method for communication-awareness integrated waveform according to claim 3 is characterized in that: The step S3 specifically includes: Taking P1 in step S2 as the power constraint and maximizing the reachability and rate of the communication system as the goal, according to the calculation formula for maximizing reachability and transmission, the communication optimization problem is established as follows: Constraints: The optimization problem is solved by the traditional water filling algorithm, and the results are recorded as the precoding matrix W(k) and the power allocation matrix P S,1 (k), according to the traditional water injection algorithm, W1(k) is expressed as: W1(k)=V1(k) in, is the singular value decomposition of H(k), and the solution to the optimization problem is the first frequency domain autocovariance matrix R f,1 (k) has:

5. The compromise design method for communication-awareness integrated waveform according to claim 4 is characterized in that: The maximum achievable sum rate of the communication system can be expressed as: Wherein, max{·} represents the maximum value, Δf is the subcarrier spacing, B is the bandwidth of the OFDM system, det(·) represents the determinant of the matrix, and I represents the identity matrix.

6. The compromise design method for communication-awareness integrated waveform according to claim 3, characterized in that: The step S4 specifically includes: Taking P2 in step S2 as the power constraint condition and minimizing the Cramer-Rao lower bound of the arrival angle estimation of the detected target as the goal, according to the Cramer-Rao lower bound calculation formula, the perception optimization problem is established as follows: Constraint 1: tr(R)=P2 Constraint 2: R=R H This problem is a convex problem, and the solution to this problem can be easily obtained using convex optimization tools, which is denoted as the second time-domain autocovariance matrix R2.

7. The compromise design method for communication-awareness integrated waveform according to claim 6, characterized in that: The Cramer-Rao lower bound calculation formula is: Among them, μ is a constant vector, and the derivative of the intermediate variable is tr(·) is the matrix trace operator, which converts A(θ)=a(θ)a(θ) H Substituting into the above equation, we get: in, is the derivative of a(θ).

8. The compromise design method for communication-awareness integrated waveform according to claim 7, characterized in that: The step S5 specifically includes: The first frequency domain autocovariance matrix R f,1 (k) is combined with the second time domain autocovariance matrix R2, and according to the corresponding relationship of the autocovariance matrix in the time-frequency domain, R2 is distributed to the frequency domain according to the set ratio to generate the second frequency domain autocovariance matrix, which is then combined with R f,1 (k) corresponding addition; Let the second frequency domain autocovariance matrix R f (k) = ω k N 2 R d ,in, And k ≥0; Taking maximizing communication reachability and transmission rate as the optimization goal, the optimization problem established by taking the allocation ratio coefficient of R2 in the frequency domain as the optimization object is as follows: Constraint 1: Constraint 2: oh k ≥0 Where ω=[ω0,ω1,...,ω N-1 ] T ,This problem is a nonlinear optimization problem and is solved by the ,semi-positive definite programming method.

9. The compromise design method for communication-awareness integrated waveform according to claim 8, characterized in that: The step S6 specifically includes: The optimization objective function is defined as: The constraint function is defined as: g k (ω)=-ω k =-e k ω,k=0,1,...,N-1 Optimization problem: Constraint 1: v(ω)=0 Constraint 2: g k (ω)≤0,k=0,1,...,N-1 Furthermore, the nonlinear function f(ω) can be obtained by (m) The Taylor expansion is transformed into a quadratic function, and the objective function is given by the following formula: Define η = ω-ω (m) , the above optimization problem is further transformed into: Constraint 1: 1st=1-1h (m) Constraint 2: -e k η≤e k oh (m) ,k=0,1,...,N-1 The above optimization problem is a standard quadratic programming problem about η. It is solved by Lagrangian operator method. The optimal ω is substituted into R f,2 (k) = ω k N 2 R2, solve the optimal R f,2 (k).

10. The compromise design method for communication-awareness integrated waveform according to claim 9, characterized in that: The step S7 specifically includes: The water injection algorithm is used to obtain the precoding matrix W1(k) with the best communication performance under the constraint of power limit P1, and the power allocation matrix P S,1 (k), and the corresponding first frequency domain autocovariance matrix R f,1 (k), and the relationship between them is The second frequency domain autocovariance matrix R f,2 (k) Satisfy: Among them, W2(k), P S,2 (k) corresponds to W1(k) of the above relationship, P S,1 (k); because Setting P S,2 (k) is: Define P°(k)=(P S,1 (k)) 1 / 2 , substituting into the above two formulas, and Rf,2(k) can be derived as: Among them, assuming that N R <N T , R f,2 The sizes of (k), W2(k), and P°(k) are N respectively. T ×N T , N T ×N R and N R ×N T Therefore, according to the above derived formula, the approximate solution of the communication precoding matrix W2(k) is derived as follows: Where Q(k)D(k)Q -1 (k) = ρ / (1-ρ)R f,2 (k) is the eigenvalue decomposition of Rf,2(k); Finally, the frequency domain precoding matrix in the optimization target is:

Images