Design Method of Quasi-Orthogonal Nonlinear Frequency Modulation Waveform in MIMO Radar

Through the greedy optimization algorithm, the quasi-orthogonal nonlinear frequency modulation waveform in multi-input multi-output radar is designed, which solves the problems of high design complexity and limited waveform freedom in the existing technology, and realizes low sidelobe and efficient radar waveform design , suitable for MIMO radar detection.

CN114966551BActive Publication Date: 2024-11-08AEROSPACE INFORMATION RES INST CAS
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202210515431.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-05-12
Publication Date
2024-11-08
Estimated Expiration
2042-05-12

AI Technical Summary

Technical Problem

It is difficult for the prior art to design a large number of quasi-orthogonal nonlinear frequency modulation waveforms in multi-input multi-output radars, and most NLFM waveform models only consider unidirectional time-frequency relationships, which limits the degree of freedom and design complexity of the waveform.

Method used

The greedy optimization algorithm is used to design the quasi-orthogonal nonlinear frequency modulation waveform in multi-input multi-output radar. By constructing a model with the peak of autocorrelation side lobes and cross-correlation peaks as optimization goals, the generalization ability and full permutation encoding of Fourier coefficients are used. Matrix, decomposition design problems are to minimize two subproblems of autocorrelation sidelobe and cross-correlation peaks, and optimize the time width, time frequency trend and Fourier coefficient of the sub-waveform.

Benefits of technology

A low side lobe quasi-orthogonal waveform design at a given bandwidth and total time width is realized, reducing the design complexity, and obtaining lower autocorrelation side lobe peaks and cross-correlation peaks. It is suitable for MIMO radar detection, and the waveform is Flexible adjustment of quantity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114966551B_ABST
    Figure CN114966551B_ABST
Patent Text Reader

Abstract

The disclosed embodiment discloses a design method of a quasi-orthogonal nonlinear frequency modulation waveform in a multi-input multi-output radar, the method comprising: determining the number of the quasi-orthogonal nonlinear frequency modulation waveforms to be designed based on the number of transceivers of the multi-input multi-output radar; constructing a quasi-orthogonal nonlinear frequency modulation waveform model, the quasi-orthogonal nonlinear frequency modulation waveform model comprising a plurality of sub-waveforms of the same frequency, the sub-waveforms having reciprocating rather than unidirectional frequency modulation; taking the autocorrelation sidelobe peak and the cross-correlation peak as the optimization targets of two optimization sub-problems respectively, calculating the key parameters of the quasi-orthogonal nonlinear frequency modulation waveform model using a greedy optimization algorithm, and determining the quasi-orthogonal nonlinear frequency modulation waveform to be designed based on the key parameters; the key parameters include the number of sub-waveforms of the quasi-orthogonal radar waveform, the time width of the sub-waveform, the time-frequency trend of the sub-waveform, and the Fourier coefficient of the sub-waveform. The technical solution can design a low-sidelobe quasi-orthogonal RNLFM waveform set with different TBPs and different IRWs under different waveform numbers.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of radar detection and imaging technology, and in particular to a design method for a quasi-orthogonal nonlinear frequency modulation waveform in a multi-input multi-output radar. Background Art

[0002] Waveform diversity has received extensive attention in the past two decades. Good orthogonality of radar waveforms can effectively improve the system performance of applications such as MIMO radar and SAR. Orthogonality can usually be obtained by distributing multiple waveforms into different domains, such as time division multiplexing, frequency division multiplexing, code division multiplexing, etc.

[0003] The goal of quasi-orthogonal radar waveform design is to generate a set of waveforms with low autocorrelation function (ACF) sidelobes and low cross-correlation function (CCF) in the same frequency band. In order to design quasi-orthogonal radar waveforms with good characteristics, four main types of waveform models have been proposed, including phase-coded (PC) waveforms, waveforms based on up-down chirps, orthogonal frequency division multiplexing (OFDM) chirped waveforms, and nonlinear frequency modulation (NLFM) waveforms with S-shaped time-frequency relationships.

[0004] The NLFM waveform with an S-shaped time-frequency relationship has three main advantages. First, the NLFM waveform naturally has a constant envelope and high spectral efficiency, which is very suitable for transmission on actual systems. Moreover, the NLFM waveform has an S-shaped time-frequency relationship, so low sidelobes of the ACF can be obtained without sacrificing the signal-to-noise ratio. In addition, the NLFM waveform has a large number of adjustable parameters, which has the potential to generate a large number of quasi-orthogonal waveforms. Although the main lobe of the ACF of the NLFM waveform will be slightly broadened, this small resolution sacrifice is acceptable when much lower ACF and CCF can be obtained. Typically, ASP and CCP are used to measure the ACF and CCF of quasi-orthogonal waveforms, respectively. The difficulty in designing quasi-orthogonal radar waveforms is to minimize ASP and CCP while maximizing the number of waveforms.

[0005] Dr. Jin (JIN G, DENG Y, WANG R, et al. Mitigating range ambiguities with advanced nonlinear frequency modulation waveform [J]. IEEE Geoscience and Remote Sensing Letters, 2019, 16 (8): 1230-1234.) uses generalized piecewise linear functions and enhanced Lagrangian genetic algorithms to optimize NLFM waveforms. Dr. Hague (HAGUE D A. Generating waveform families using multi-tone sinusoidal frequency modulation [C] / / Generating waveform families using multi-tone sinusoidal frequency modulation. 2020 IEEE International Radar Conference (RADAR). IEEE: 946-951.) generates quasi-orthogonal waveforms based on multi-tone sinusoidal frequency modulation waveform models. However, the above two works only give the design results of two quasi-orthogonal waveforms. Dr. Gao (GAO C, TEH KC, LIUA. Piecewise nonlinear frequency modulation waveform for MIMO radar [J]. IEEE Journal of Selected Topics in Signal Processing, 2016, 11 (2): 379-390.) proposed a three-segment segmented NLFM waveform. Dr. Roy (ROY A, NEMADE HB, BHATTACHARJEE R. Radar waveform diversity using nonlinear chirp with improved sidelobe level performance [J]. AEU-International Journal of Electronics and Communications, 2021, 136: 153768.) studied the design process of discrete frequency coding using NLFM. In the articles of Dr. Gao and Dr. Roy, the objective function is a combination of the ASP and CCP of all quasi-orthogonal waveforms. As the number of quasi-orthogonal waveforms increases, the complexity of the objective function will also increase rapidly, making the quasi-orthogonal waveform design problem difficult to solve.In addition, most NLFM waveform models, such as positive-negative NLFM waveform or segmented NLFM waveform, only consider the one-way time-frequency relationship in a given subband, which limits the freedom of the waveform. Summary of the invention

[0006] The disclosed embodiment provides a method for designing a quasi-orthogonal nonlinear frequency modulation waveform in a multiple-input multiple-output radar.

[0007] In a first aspect, an embodiment of the present disclosure provides a method for designing a quasi-orthogonal nonlinear frequency modulation waveform in a multiple-input multiple-output radar, which includes:

[0008] Determining the number of the quasi-orthogonal nonlinear frequency modulation waveforms to be designed based on the number of transceivers of the multiple-input multiple-output radar;

[0009] Constructing a quasi-orthogonal nonlinear frequency modulation waveform model, wherein the quasi-orthogonal nonlinear frequency modulation waveform model includes a plurality of sub-waveforms of the same frequency, wherein the sub-waveforms have reciprocating rather than unidirectional frequency modulation;

[0010] The autocorrelation sidelobe peak and the cross-correlation peak are respectively used as the optimization targets of two optimization sub-problems, and the key parameters of the quasi-orthogonal nonlinear frequency modulation waveform model are calculated using a greedy optimization algorithm. The quasi-orthogonal nonlinear frequency modulation waveform to be designed is determined based on the key parameters; the key parameters include the number of sub-waveforms of the quasi-orthogonal radar waveform, the time width of the sub-waveform, the time-frequency trend of the sub-waveform and the Fourier coefficients of the sub-waveform.

[0011] Furthermore, the autocorrelation sidelobe peak and the cross-correlation peak are respectively used as optimization targets of two optimization sub-problems, and the key parameters of the quasi-orthogonal nonlinear frequency modulation waveform model are calculated using a greedy optimization algorithm, including:

[0012] Constructing a first objective function with the time width of the sub-waveform and the Fourier coefficient of the sub-waveform as parameters; the optimization target of the first objective function is the autocorrelation sidelobe peak value;

[0013] Constructing a second objective function with the time width of the sub-waveform and the time-frequency trend of the sub-waveform as parameters; the optimization target of the second objective function is the cross-correlation peak value;

[0014] The time width of the sub-waveform, the time-frequency trend of the sub-waveform and the Fourier coefficients of the sub-waveform are calculated by using a greedy optimization algorithm, a first objective function and a second objective function; the bandwidth of each sub-waveform is the same.

[0015] Further, the time width of the sub-waveform, the time-frequency trend of the sub-waveform, and the Fourier coefficients of the sub-waveform are obtained by solving the greedy optimization algorithm, the first objective function, and the second objective function, including:

[0016] Initialize the time width matrix of the sub-waveform, the time-frequency trend matrix of the sub-waveform, and the number of iterations i and j;

[0017] Determine the time-frequency trend matrix of the i-th iteration; wherein each row in the time-width matrix of the i-th iteration is 2 i+1 One of the possible cases of full arrangement, corresponding to 2 i+1 The time-frequency trend of one of the quasi-orthogonal radar waveforms;

[0018] Optimizing the time width matrix of the i-th iteration; wherein the time width matrix of the i-th iteration is determined based on the time width matrix optimized in the i-1-th iteration;

[0019] Optimize the Fourier coefficient matrix of the i-th iteration; where k time widths are The quasi-orthogonal radar waveform is used as a reference waveform, and after optimizing the Fourier coefficients of the reference waveform, the Fourier coefficients of the reference waveform closest to the time width of each sub-waveform are determined as the Fourier coefficients of the sub-waveform;

[0020] If 2 i+1 ≥W, the optimization ends; otherwise, set i=i+1 and repeat the above steps.

[0021] In a second aspect, an embodiment of the present disclosure provides a design device for a quasi-orthogonal nonlinear frequency modulation waveform in a multiple-input multiple-output radar, which includes:

[0022] A determination module, configured to determine the number of the quasi-orthogonal nonlinear frequency modulation waveforms to be designed based on the number of transceivers of the multiple-input multiple-output radar;

[0023] A construction module is configured to construct a quasi-orthogonal nonlinear frequency modulation waveform model, wherein the quasi-orthogonal nonlinear frequency modulation waveform model includes a plurality of sub-waveforms with the same frequency, and the sub-waveforms have reciprocating rather than unidirectional frequency modulation;

[0024] The calculation module is configured to use the autocorrelation sidelobe peak and the cross-correlation peak as the optimization targets of two optimization sub-problems respectively, calculate the key parameters of the quasi-orthogonal nonlinear frequency modulation waveform model by using a greedy optimization algorithm, and determine the quasi-orthogonal nonlinear frequency modulation waveform to be designed based on the key parameters; the key parameters include the number of sub-waveforms of the quasi-orthogonal radar waveform, the time width of the sub-waveform, the time-frequency trend of the sub-waveform, and the Fourier coefficient of the sub-waveform.

[0025] Furthermore, the calculation module includes:

[0026] A first construction submodule is configured to construct a first objective function with the time width of the sub-waveform and the Fourier coefficient of the sub-waveform as parameters; the optimization target of the first objective function is the autocorrelation sidelobe peak value;

[0027] A second construction submodule is configured to construct a second objective function with the time width of the sub-waveform and the time-frequency trend of the sub-waveform as parameters; the optimization target of the second objective function is the cross-correlation peak value;

[0028] The solver module is configured to use a greedy optimization algorithm, a first objective function and a second objective function to solve and obtain the time width of the sub-waveform, the time-frequency trend of the sub-waveform and the Fourier coefficient of the sub-waveform; the bandwidth of each sub-waveform is the same.

[0029] Furthermore, the solver module is implemented as follows:

[0030] Initialize the time width matrix of the sub-waveform, the time-frequency trend matrix of the sub-waveform, and the number of iterations i and j;

[0031] Determine the time-frequency trend matrix of the i-th iteration; wherein each row in the time-width matrix of the i-th iteration is 2 i+1 One of the possible cases of full arrangement, corresponding to 2 i+1 The time-frequency trend of one of the quasi-orthogonal radar waveforms;

[0032] Optimizing the time width matrix of the i-th iteration; wherein the time width matrix of the i-th iteration is determined based on the time width matrix optimized in the i-1-th iteration;

[0033] Optimize the Fourier coefficient matrix of the i-th iteration; where k time widths are The quasi-orthogonal radar waveform is used as a reference waveform, and after optimizing the Fourier coefficients of the reference waveform, the Fourier coefficients of the reference waveform closest to the time width of each sub-waveform are determined as the Fourier coefficients of the sub-waveform;

[0034] If 2 i+1 ≥W, the optimization ends; otherwise, set i=i+1 and repeat the above steps.

[0035] The functions can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the functions.

[0036] In one possible design, the structure of the above-mentioned device includes a memory and a processor, the memory is used to store one or more computer instructions that support the above-mentioned device to execute the above-mentioned corresponding method, and the processor is configured to execute the computer instructions stored in the memory. The above-mentioned device may also include a communication interface for the above-mentioned device to communicate with other devices or communication networks.

[0037] In a third aspect, an embodiment of the present disclosure provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method described in any one of the above aspects.

[0038] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium for storing computer instructions used by any of the above-mentioned devices, and when the computer instructions are executed by a processor, they are used to implement the method described in any of the above-mentioned aspects.

[0039] In a fifth aspect, an embodiment of the present disclosure provides a computer program product, which includes computer instructions, and when the computer instructions are executed by a processor, they are used to implement the method described in any of the above aspects.

[0040] The technical solution provided by the embodiments of the present disclosure may have the following beneficial effects:

[0041] The beneficial effects of the present disclosure include, but are not limited to, the following:

[0042] 1. Aiming at the application requirements of MIMO radar detection, a waveform with reciprocating nonlinear frequency modulation, namely RNLFM waveform, is designed. RNLFM waveform has both the low sidelobe characteristics of NLFM and the high degree of freedom brought by reciprocating frequency modulation, and can use the generalization ability of Fourier coefficients to simplify the waveform optimization process.

[0043] 2. A greedy optimization method is proposed to design quasi-orthogonal RNLFM waveforms. The greedy optimization method effectively decomposes the quasi-orthogonal RNLFM waveform design problem into two optimization problems, minimizing the ASP and CCP values ​​respectively, reducing the complexity of the problem.

[0044] 3. The designed quasi-orthogonal RNLFM waveform is used for MIMO radar detection, and lower ASP and CCP values ​​are obtained. When the MIMO radar is 2-transmit and 2-receive, the worst ASP and CCP values ​​of the quasi-orthogonal RNLFM waveform are -33.2dB and -28.1dB respectively.

[0045] 4. According to the number of transmitters and receivers and system requirements of different MIMO radars, the RNLFM waveform parameters can be flexibly adjusted to design a low-sidelobe quasi-orthogonal RNLFM waveform set with different TBPs and IRWs under different numbers of waveforms.

[0046] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Other features, objectives and advantages of the present disclosure will become more apparent through the following detailed description of non-limiting embodiments in conjunction with the accompanying drawings. In the accompanying drawings:

[0048] Figure 1 A flow chart showing a method for designing a quasi-orthogonal nonlinear frequency modulation waveform in a multiple-input multiple-output radar according to an embodiment of the present disclosure;

[0049] Figure 2 An example schematic diagram of a frequency-time curve of an RNLFM waveform model according to an embodiment of the present disclosure is shown;

[0050] Figure 3 A schematic diagram of a greedy optimization algorithm flow according to an embodiment of the present disclosure is shown;

[0051] Figure 4 A schematic diagram showing an autocorrelation function of a quasi-orthogonal RNLFM waveform according to an embodiment of the present disclosure is shown;

[0052] Figure 5 A schematic diagram showing a cross-correlation function of a quasi-orthogonal RNLFM waveform according to an embodiment of the present disclosure;

[0053] Figure 6 A frequency-time curve diagram of a quasi-orthogonal RNLFM waveform according to an embodiment of the present disclosure is shown;

[0054] Figure 7 The present invention is a schematic diagram of the structure of an electronic device suitable for implementing a method for designing a quasi-orthogonal nonlinear frequency modulation waveform in a multiple-input multiple-output radar according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0055] Hereinafter, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily implement them. In addition, for the sake of clarity, parts not related to the description of the exemplary embodiments are omitted in the accompanying drawings.

[0056] In the present disclosure, it should be understood that terms such as "include" or "have" are intended to indicate the presence of features, numbers, steps, behaviors, components, parts, or a combination thereof disclosed in the specification, and do not exclude the possibility that one or more other features, numbers, steps, behaviors, components, parts, or a combination thereof exist or are added.

[0057] It should also be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present disclosure may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0058] The present disclosure discloses a quasi-orthogonal waveform design method with reciprocating nonlinear frequency modulation (RNLFM). In this method, a RNLFM waveform model with low sidelobe characteristics and high degrees of freedom is first established, and then a greedy optimization method is proposed to decompose the design problem of the quasi-orthogonal RNLFM waveform into two sub-problems of minimizing the autocorrelation sidelobe peak and the cross-correlation peak. Among them, the optimization of the autocorrelation sidelobe peak utilizes the generalization of the nonlinear frequency modulation signal based on the Fourier series, while the optimization of the cross-correlation peak is achieved through a fully permuted coding matrix and an optimally designed time-width matrix. Therefore, the greedy optimization method can efficiently solve the optimal RNLFM waveform set in the current iteration, which is also a subset of the optimal quasi-orthogonal RNLFM waveform set under a larger number of waveforms.

[0059] The details of the embodiments of the present disclosure are described in detail below through specific examples.

[0060] In order to solve the problem of angular glint in the detection target, the embodiment of the present disclosure adopts N T Send N R The number of quasi-orthogonal RNLFM waveforms that need to be designed is N. T ×N R .

[0061] In the embodiment of the present disclosure, N is set based on actual needs. T Send N R The bandwidth and total time width of the received MIMO radar are designed by greedy optimization method. The number is N T ×N R The designed quasi-orthogonal RNLFM waveform set is applied to N T Send N RThe target detection of the MIMO radar is carried out by using the matched filtering, and the results of the matched filtering are calculated to obtain the optimized autocorrelation sidelobe peak (ASP) and cross-correlation peak (CCP). It can be understood that the quasi-orthogonal waveform design is to minimize its autocorrelation and cross-correlation sidelobes, that is, ASP and CCP, under the conditions of given bandwidth and total time width. Matched filtering is the detection distance. After designing multiple quasi-orthogonal RNLFM waveforms and obtaining the optimized results of ASP and CCP, the worst ASP and CCP can be used as the measurement indicators of the designed quasi-orthogonal RNLFM waveform.

[0062] Figure 1 A flow chart showing a method for designing a quasi-orthogonal nonlinear frequency modulation waveform in a multiple-input multiple-output radar according to an embodiment of the present disclosure is shown. Figure 1 As shown, the design method of the quasi-orthogonal nonlinear frequency modulation waveform in the multiple-input multiple-output radar includes the following steps:

[0063] In step S101, the number of the quasi-orthogonal nonlinear frequency modulation waveforms to be designed is determined based on the number of transceivers of the MIMO radar;

[0064] In step S102, a quasi-orthogonal nonlinear frequency modulation waveform model is constructed, wherein the quasi-orthogonal nonlinear frequency modulation waveform model includes a plurality of sub-waveforms with the same frequency, and the sub-waveforms have reciprocating rather than unidirectional frequency modulation;

[0065] In step S103, the autocorrelation sidelobe peak and the cross-correlation peak are respectively used as optimization targets of two optimization sub-problems, and the key parameters of the quasi-orthogonal nonlinear frequency modulation waveform model are calculated using a greedy optimization algorithm. The quasi-orthogonal nonlinear frequency modulation waveform to be designed is determined based on the key parameters; the key parameters include the number of sub-waveforms of the quasi-orthogonal radar waveform, the time width of the sub-waveform, the time-frequency trend of the sub-waveform, and the Fourier coefficients of the sub-waveform.

[0066] The above method steps are described in detail below.

[0067] The present disclosure proposes a RNLFM waveform model composed of multiple sub-NLFM waveforms, presenting reciprocating rather than unidirectional frequency modulation. There are four key parameters for the design of quasi-orthogonal waveforms, namely the number of sub-NLFM waveforms, the time width matrix, the coding matrix and the Fourier coefficient matrix. The RNLFM waveform has both the low sidelobe characteristics of NLFM and the high degree of freedom of reciprocating frequency modulation, so it has the potential to produce a quasi-orthogonal waveform with low autocorrelation sidelobe peak (ASP) and low cross-correlation peak (CCP). The greedy optimization method further decomposes the design problem of the quasi-orthogonal RNLFM waveform into two optimization problems of lower complexity, namely minimizing ASP and CCP respectively. Specifically, ASP is optimized by utilizing the generalization ability of Fourier coefficients, and CCP is optimized by full permutation of the coding matrix and careful design of the time width matrix. Therefore, the greedy optimization method can effectively solve the optimal waveform set of the current iteration, which is also a subset of the optimal RNLFM waveform set in the case of a larger number of waveforms.

[0068] The following first introduces the RNLFM waveform model, then introduces the greedy optimization method based on it, and finally gives the simulation results.

[0069] Figure 2 FIG. 1 is a schematic diagram showing an example of a frequency-time curve of an RNLFM waveform model according to an embodiment of the present disclosure. The RNLFM waveform s(t) is composed of multiple sub-NLFMs of the same frequency, with reciprocating frequency modulation. Assuming that there are M( Figure 2 The mth sub-NLFM can be expressed as s(m,t) (m=1,2,…,M). Assume that the total time width of the RNLFM waveform s(t) is T s , the time width and phase of each sub-NLFM are T m and φ(m,t), then And the central moment of each sub-NLFM (Let T0 = 0). In the case of a rectangular window, s(m, t) can be written as follows:

[0070]

[0071] The phase term φ(m,tt of the sub-NLFM m ) is shown in the following equation (2), using the NLFM waveform representation based on Fourier series, where B is the bandwidth, K m,n represents the nth Fourier coefficient of the mth sub-NLFM, where N is the number of Fourier coefficients.

[0072]

[0073] The generalization ability of the Fourier coefficients of the NLFM waveform is described below to simplify the subsequent waveform optimization. This generalization ability is reflected in that the optimized Fourier coefficients of the NLFM waveform under a given time-bandwidth product (TBP) can be applied to other NLFM waveforms with a smaller TBP to achieve similar low sidelobes. For example, assume that the TBP of the NLFM waveform is BT s = 3200, and K s represents the optimized Fourier coefficients obtained through a two-step optimization framework (XIE Q, ZENG H, MO Z, et al. A Two-Step Optimization Framework for Low Sidelobe NLFM Waveform Using Fourier Series[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19: 1-5). Then the optimized Fourier coefficients K s are applied to other NLFM waveforms with a TBP smaller than BT s . As shown in Table 1, the peak-to-sidelobe level ratio (PSLR) of the NLFM waveform gradually increases as the TBP decreases. This is because using K s for waveforms with a TBP smaller than BT s reduces the phase change, thereby reducing the nonlinearity of the waveform. However, it can be seen that when 700 < TBP < 3200, the PSLR is less than -32 dB. Therefore, the same optimized Fourier coefficients K s can also maintain the low sidelobe characteristics of the waveform within a relatively wide TBP range. Therefore, it is not necessary to optimize the Fourier coefficients for each sub-NLFM, which can improve the efficiency of the following optimization process.

[0074] Table 1 Generalization ability of the Fourier coefficients of the NLFM waveform

[0075] TBP 200 700 1200 1700 2200 2700 3200 PSLR -30.9 -32.9 -33.2 -33.4 -33.3 -33.5 -33.5

[0076] At a given total time width T s and bandwidth B, there are mainly four types of key parameters for designing the RNLFM waveform: the number M of sub-NLFM waveforms, the time widths T1, T2, …, T M of the sub-NLFM waveforms, the time-frequency trends c1, c2, …, c M (where c m = 1 or -1 indicates that the time-frequency trend of the m-th sub-NLFM waveform is rising or falling), and the Fourier coefficients K1, K2, …, K M of the sub-NLFM waveforms.

[0077] For the design of W quasi-orthogonal RNLFM waveforms, the above four key parameters are expanded as follows:

[0078] M w Describes the number of sub-NLFM waveforms of the w-th RNLFM.

[0079] The time width matrix T, where T wm Represents the duration of the mth sub-NLFM waveform of the wth RNLFM.

[0080] The encoding matrix C, where C wm Represents the time-frequency trend of the m-th sub-NLFM waveform of the w-th RNLFM.

[0081] The Fourier coefficient matrix K, where K wm Represents the Fourier coefficients of the m-th sub-NLFM waveform of the w-th RNLFM.

[0082] RNLFM waveform optimization

[0083] The disclosed embodiment simplifies the quasi-orthogonal RNLFM waveform optimization problem, and then proposes a greedy optimization method to solve the problem. The core idea of ​​the greedy optimization method is to design the optimal quasi-orthogonal RNLFM waveform in each iteration. The current iterative optimization process is based on the quasi-orthogonal RNLFM waveform optimized in the previous iteration, and iteratively generates more optimal quasi-orthogonal RNLFM waveforms.

[0084] Quasi-orthogonal RNLFM waveform optimization problem

[0085] In order to reduce the complexity of the problem, the design problem of quasi-orthogonal RNLFM waveform is decomposed into two sub-problems. The objective functions of the two sub-problems are set as F1 and F2, corresponding to the lowest ASP value and the lowest CCP value respectively.

[0086] Subproblem 1——Objective function F1

[0087] definition is the M containing the w-th RNLFM w The Fourier coefficient vector of the sub-NLFM waveform. Similarly, let As shown in formula (3), the objective function F1 is the minimum value of the W ASP values ​​of the W RNLFM waveforms, that is, the worst ASP value. w The ASP level of the RNLFM waveform is determined by its spectrum, which mainly depends on K w and T w .

[0088]

[0089] T is obtained by optimizing the objective function F2 in subproblem 2. In order to minimize F1, the generalization ability of Fourier coefficients is used to determine K according to T. Assume that k NLFM waveforms with increasing time width are selected as reference waveforms, The optimized Fourier coefficients obtained through the two-step optimization framework are K ref =(K ref,1 ,K ref,2 ,…,K ref,k ), where ref,n The time width of the corresponding reference NLFM waveform is In equation (4), the impulse response width (IRW) constraint is introduced to control the loss of resolution, where the IRW is measured using the normalized 3 dB main lobe width.

[0090] b0 is the set normalized IRW, and b is the normalized IRW of the waveform (for a linear frequency modulation waveform, b=0.886).

[0091]

[0092] stb<b0(b=IRW(K ref ,T ref )) (4)

[0093] Since the Fourier coefficients have generalization ability, the sub-NLFM waveform selects the optimized Fourier coefficients of the reference waveform closest to its TBP. The bandwidth of each sub-NLFM is the same, so it is equivalent to determining K by finding the optimized Fourier coefficients corresponding to the reference waveform with the closest time width to the sub-NLFM, as shown in equation (5).

[0094]

[0095] Subproblem 2 - Objective function F2

[0096] The objective function F2 is the minimum value of the W(W-1) / 2 CCP values ​​of the W RNLFM waveforms, that is, the worst CCP value. Here, F2 is optimized by fully permuting the code matrix C and carefully designing the time width matrix T.

[0097]

[0098] Greedy Optimization Method

[0099] Table 2 Greedy optimization process of F2

[0100]

[0101] Greedy thinking means making the best choice at each step, which is specifically reflected in solving the optimal quasi-orthogonal RNLFM waveform of the current iteration. As the optimization proceeds, based on the current optimal quasi-orthogonal RNLFM waveform, more optimal quasi-orthogonal RNLFM waveforms are gradually iterated and generated. The flowchart of the greedy optimization method for RNLFM waveform design is shown in the figure. Figure 3 As shown, its main steps include the following 5 steps. Table 2 gives the pseudo code of step 2 and step 3 in the greedy optimization method.

[0102] Step 1: Initialization

[0103] Assume that the number of quasi-orthogonal RNLFM waveforms to be designed is W. Set the initial value of the encoding matrix C (0) The first and second rows of are set to 1 and -1 respectively, that is, the initial values ​​of the time-frequency trend are one positive and one negative. (0) The first and second rows of are set to 1. The iteration numbers i and j are initialized to 0. The time width matrix The first element t1 is also set to 0.

[0104] Step 2: Generate the encoding matrix C for the i-th iteration (i)

[0105] In order to improve the orthogonality between RNLFM waveforms, the full permutation of 1 and -1 is used to generate the coding matrix C. (i) , where each bit can take the value of 1 or -1. That is, for the i-th iteration, the encoding matrix C (i) We can use the number with the value of 1 or -1 in the i+1 position to generate it in all possible orders, C (i) Each row of corresponds to the time-frequency trend of a RNLFM waveform. w sub-NLFM waveforms, each sub-NLFM waveform has a time-frequency trend of rising or falling, corresponding to 1 or -1 in. For the i-th iteration, M w =i+1, so there are 2 i+1 In order to meet the design requirements of W RNLFM waveforms, M w ≥log2W-1. C (i) The iterative representation of is shown in rows 9 and 11 of Table 2.

[0106] Step 3: Optimize the time width matrix T of the i-th iteration (i)

[0107] T (i) The optimization process utilizes C (i) The good orthogonality provided reduces the complexity, that is, the time width vectors of the RNLFM waveforms corresponding to two rows with completely opposite time-frequency trends are kept consistent.(i) The iteration representation of T is shown in row 13 of Table 2. (i) The update of each row consists of two parts, one of which is the t that needs to be optimized. j , and the other part is the optimized T (i-1) . With C (i) Correspondingly, the optimized T (i-1) To generate T (i) , so the optimal quasi-orthogonal RNLFM waveform for the i-th iteration is generated based on the result of the i-1-th iteration. j The solution is achieved through sequential quadratic programming. Where j is the number of iterations of optimizing T, t j Represents the elements in the time-width matrix T during the j-th iteration.

[0108] Step 4: Optimize the Fourier coefficient matrix K of the i-th iteration (i)

[0109] Select k time widths as The NLFM waveform of is used as the reference waveform, and k is the number of reference waveforms. After the Fourier coefficients of the reference waveform are optimized, the Fourier coefficients of the reference waveform with the closest time width are selected for each sub-NLFM waveform of each RNLFM waveform.

[0110] Step 5: Iteration

[0111] If 2 i+1 ≥W, the optimization is completed, otherwise let i=i+1.

[0112] Simulation Results

[0113] Next, for N T ×N R = N of 4, 8 and 16 T Send N R The target detection problem of MIMO radar is simulated, and the waveform design of 4, 8 and 16 quasi-orthogonal RNLFM is simulated, and the worst ASP and CCP values ​​under the corresponding conditions are calculated. Set the total time width T of the RNLFM waveform s The and bandwidth B are 8μs and 400MHz respectively. According to the results in Table 1, two NLFM waveforms with TBP of 200 and 3200 are selected as reference waveforms (k=2). Under the condition of setting the normalized IRWb of the waveform to 1.1, the PSLR of the two reference waveforms is optimized to -33dB using a two-step optimization framework. Then, the quasi-orthogonal RNLFM waveform is generated by the greedy optimization method.

[0114] Figure 4 and Figure 5The ACF and CCF of four RNLFM waveforms (denoted as RNLFM1, RNLFM2, RNLFM3 and RNLFM4) are shown respectively. Figure 4 In the figure, RNLFM1 and RNLFM2 have the same time width vector but opposite coding vector, so they have the same ACF. Similarly, the ACF of RNLFM3 and RNLFM4 is also the same. When the number of waveforms W = 4, the worst ASP value and CCP value are -33.2dB and -28.1dB respectively; when W = 8, the worst ASP value and CCP value are -33.0dB and -26.4dB respectively; when the number of waveforms W = 16, the worst ASP value and CCP value are -31.4dB and -23.0dB respectively. It can be seen that in N T ×N R = N of 4, 8 and 16 T Send N R In the target detection problem of MIMO radar, the optimized quasi-orthogonal RNLFM waveform can achieve a lower ASP value (<-31.4dB) and a lower CCP value (<-23.0dB). Therefore, the quasi-orthogonal RNLFM under the greedy optimization method can ensure as low autocorrelation and cross-correlation sidelobes as possible while expanding the number of waveforms. The greedy optimization method is to continuously expand the quasi-orthogonal RNLFM waveform set through iteration, so the quasi-orthogonal RNLFM waveform set with a smaller number of waveforms is a subset of the quasi-orthogonal RNLFM waveform set with a larger number of waveforms, such as Figure 6 shown.

[0115] The beneficial effects of the present disclosure include, but are not limited to, the following:

[0116] 1. Aiming at the application requirements of MIMO radar detection, a waveform with reciprocating nonlinear frequency modulation, namely RNLFM waveform, is designed. RNLFM waveform has both the low sidelobe characteristics of NLFM and the high degree of freedom brought by reciprocating frequency modulation, and can use the generalization ability of Fourier coefficients to simplify the waveform optimization process.

[0117] 2. A greedy optimization method is proposed to design quasi-orthogonal RNLFM waveforms. The greedy optimization method effectively decomposes the quasi-orthogonal RNLFM waveform design problem into two optimization problems, minimizing the ASP and CCP values ​​respectively, reducing the complexity of the problem.

[0118] 3. The designed quasi-orthogonal RNLFM waveform is used for MIMO radar detection, and lower ASP and CCP values ​​are obtained. When the MIMO radar is 2-transmit and 2-receive, the worst ASP and CCP values ​​of the quasi-orthogonal RNLFM waveform are -33.2dB and -28.1dB respectively.

[0119] 4. According to the number of transmitters and receivers and system requirements of different MIMO radars, the RNLFM waveform parameters can be flexibly adjusted to design a low-sidelobe quasi-orthogonal RNLFM waveform set with different TBPs and IRWs under different numbers of waveforms.

[0120] The following are embodiments of the apparatus of the present disclosure, which can be used to execute embodiments of the method of the present disclosure.

[0121] According to a design device for a quasi-orthogonal nonlinear frequency modulation waveform in a multi-input multi-output radar according to an embodiment of the present disclosure, the device can be implemented as part or all of an electronic device through software, hardware, or a combination of both. The design device for a quasi-orthogonal nonlinear frequency modulation waveform in a multi-input multi-output radar includes:

[0122] A determination module, configured to determine the number of the quasi-orthogonal nonlinear frequency modulation waveforms to be designed based on the number of transceivers of the multiple-input multiple-output radar;

[0123] A construction module is configured to construct a quasi-orthogonal nonlinear frequency modulation waveform model, wherein the quasi-orthogonal nonlinear frequency modulation waveform model includes a plurality of sub-waveforms with the same frequency, and the sub-waveforms have reciprocating rather than unidirectional frequency modulation;

[0124] The calculation module is configured to use the autocorrelation sidelobe peak and the cross-correlation peak as the optimization targets of two optimization sub-problems respectively, calculate the key parameters of the quasi-orthogonal nonlinear frequency modulation waveform model by using a greedy optimization algorithm, and determine the quasi-orthogonal nonlinear frequency modulation waveform to be designed based on the key parameters; the key parameters include the number of sub-waveforms of the quasi-orthogonal radar waveform, the time width of the sub-waveform, the time-frequency trend of the sub-waveform, and the Fourier coefficient of the sub-waveform.

[0125] In an optional implementation of this embodiment, the calculation module includes:

[0126] A first construction submodule is configured to construct a first objective function with the time width of the sub-waveform and the Fourier coefficient of the sub-waveform as parameters; the optimization target of the first objective function is the autocorrelation sidelobe peak value;

[0127] A second construction submodule is configured to construct a second objective function with the time width of the sub-waveform and the time-frequency trend of the sub-waveform as parameters; the optimization target of the second objective function is the cross-correlation peak value;

[0128] The solver module is configured to use a greedy optimization algorithm, a first objective function and a second objective function to solve and obtain the time width of the sub-waveform, the time-frequency trend of the sub-waveform and the Fourier coefficient of the sub-waveform; the bandwidth of each sub-waveform is the same.

[0129] In an optional implementation of this embodiment, the solver module is implemented as follows:

[0130] Initialize the time width matrix of the sub-waveform, the time-frequency trend matrix of the sub-waveform, and the number of iterations i and j;

[0131] Determine the time-frequency trend matrix of the i-th iteration; wherein each row in the time-width matrix of the i-th iteration is 2 i+1 One of the possible cases of full permutation, corresponding to 2 i+1 The time-frequency trend of one of the quasi-orthogonal radar waveforms;

[0132] Optimizing the time width matrix of the i-th iteration; wherein the time width matrix of the i-th iteration is determined based on the time width matrix optimized in the i-1-th iteration;

[0133] Optimize the Fourier coefficient matrix of the i-th iteration; where k time widths are The quasi-orthogonal radar waveform is used as a reference waveform, and after optimizing the Fourier coefficients of the reference waveform, the Fourier coefficients of the reference waveform closest to the time width of each sub-waveform are determined as the Fourier coefficients of the sub-waveform;

[0134] If 2 i+1 ≥W, the optimization ends; otherwise, set i=i+1 and repeat the above steps.

[0135] Figure 7 The present invention is a schematic diagram of the structure of an electronic device suitable for implementing a method for designing a quasi-orthogonal nonlinear frequency modulation waveform in a multiple-input multiple-output radar according to an embodiment of the present disclosure.

[0136] like Figure 7 As shown, the electronic device 700 includes a processing unit 701, which can be implemented as a processing unit such as a CPU, a GPU, an FPGA, and an NPU. The processing unit 701 can perform various processes in the embodiments of any of the above methods of the present disclosure according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage part 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the electronic device 700 are also stored. The processing unit 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0137] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, etc.; an output section 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, a modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as needed, so that a computer program read therefrom is installed into the storage section 708 as needed.

[0138] In particular, according to an embodiment of the present disclosure, any method in the above referenced embodiment of the present disclosure may be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program tangibly contained on a machine-readable medium, and the computer program includes a program code for executing any method in the embodiment of the present disclosure. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 709, and / or installed from a removable medium 711.

[0139] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the road map or block diagram can represent a part of a module, program segment or code, and the part of the module, program segment or code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a different order from the order marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0140] The units or modules involved in the embodiments described in the present disclosure may be implemented by software or hardware. The units or modules described may also be arranged in a processor, and the names of these units or modules do not constitute limitations on the units or modules themselves in some cases.

[0141] As another aspect, the present disclosure further provides a computer-readable storage medium, which may be a computer-readable storage medium included in the device described in the above embodiment; or a computer-readable storage medium that exists independently and is not assembled into the device. The computer-readable storage medium stores one or more programs, and the programs are used by one or more processors to execute the method described in the present disclosure.

[0142] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present disclosure is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other.

Claims

1. A design method for quasi-orthogonal nonlinear frequency modulation waveform in a multiple-input multiple-output radar, wherein: include: Determining the number of the quasi-orthogonal nonlinear frequency modulation waveforms to be designed based on the number of transceivers of the multiple-input multiple-output radar; Constructing a quasi-orthogonal nonlinear frequency modulation waveform model, wherein the quasi-orthogonal nonlinear frequency modulation waveform model includes a plurality of sub-waveforms of the same frequency, wherein the sub-waveforms have reciprocating rather than unidirectional frequency modulation; The autocorrelation sidelobe peak and the cross-correlation peak are respectively used as optimization targets of two optimization sub-problems, and the key parameters of the quasi-orthogonal nonlinear frequency modulation waveform model are calculated using a greedy optimization algorithm, and the quasi-orthogonal nonlinear frequency modulation waveform to be designed is determined based on the key parameters; the key parameters include the number of sub-waveforms of the quasi-orthogonal nonlinear frequency modulation waveform, the time width of the sub-waveform, the time-frequency trend of the sub-waveform, and the Fourier coefficients of the sub-waveform; The autocorrelation sidelobe peak and the cross-correlation peak are respectively used as the optimization targets of the two optimization sub-problems, and the key parameters of the quasi-orthogonal nonlinear frequency modulation waveform model are calculated using a greedy optimization algorithm, including: Constructing a first objective function with the time width of the sub-waveform and the Fourier coefficient of the sub-waveform as parameters; the optimization target of the first objective function is the autocorrelation sidelobe peak value; Constructing a second objective function with the time width of the sub-waveform and the time-frequency trend of the sub-waveform as parameters; the optimization target of the second objective function is the cross-correlation peak value; The time width of the sub-waveform, the time-frequency trend of the sub-waveform and the Fourier coefficient of the sub-waveform are obtained by using a greedy optimization algorithm, a first objective function and a second objective function; the bandwidth of each sub-waveform is the same; The greedy optimization algorithm, the first objective function and the second objective function are used to calculate the time width of the sub-waveform, the time-frequency trend of the sub-waveform and the Fourier coefficients of the sub-waveform, including: Initialize the time width matrix of the sub-waveform, the time-frequency trend matrix of the sub-waveform, and the number of iterations i and j; Determine the time-frequency trend matrix of the i-th iteration; wherein each row in the time-width matrix of the i-th iteration is One of the possible cases of all permutations, corresponding to The time-frequency trend of one of the quasi-orthogonal radar waveforms; Optimizing the time width matrix of the i-th iteration; wherein the time width matrix of the i-th iteration is determined based on the time width matrix optimized in the i-1-th iteration; Optimize the Fourier coefficient matrix of the i-th iteration; where k time widths are The quasi-orthogonal radar waveform is used as a reference waveform, and after optimizing the Fourier coefficients of the reference waveform, the Fourier coefficients of the reference waveform closest to the time width of each sub-waveform are determined as the Fourier coefficients of the sub-waveform; like , then the optimization ends; otherwise, Then, repeat the above steps.

2. A design device for quasi-orthogonal nonlinear frequency modulation waveform in a multiple-input multiple-output radar, wherein: include: A determination module, configured to determine the number of the quasi-orthogonal nonlinear frequency modulation waveforms to be designed based on the number of transceivers of the multiple-input multiple-output radar; A construction module is configured to construct a quasi-orthogonal nonlinear frequency modulation waveform model, wherein the quasi-orthogonal nonlinear frequency modulation waveform model includes a plurality of sub-waveforms with the same frequency, and the sub-waveforms have reciprocating rather than unidirectional frequency modulation; A calculation module is configured to use the autocorrelation sidelobe peak and the cross-correlation peak as optimization targets of two optimization sub-problems respectively, calculate the key parameters of the quasi-orthogonal nonlinear frequency modulation waveform model by using a greedy optimization algorithm, and determine the quasi-orthogonal nonlinear frequency modulation waveform to be designed based on the key parameters; the key parameters include the number of sub-waveforms of the quasi-orthogonal nonlinear frequency modulation waveform, the time width of the sub-waveform, the time-frequency trend of the sub-waveform, and the Fourier coefficient of the sub-waveform; Wherein, the computing module comprises: A first construction submodule is configured to construct a first objective function with the time width of the sub-waveform and the Fourier coefficient of the sub-waveform as parameters; the optimization target of the first objective function is the autocorrelation sidelobe peak value; A second construction submodule is configured to construct a second objective function with the time width of the sub-waveform and the time-frequency trend of the sub-waveform as parameters; the optimization target of the second objective function is the cross-correlation peak value; A solver module is configured to use a greedy optimization algorithm, a first objective function and a second objective function to solve and obtain the time width of the sub-waveform, the time-frequency trend of the sub-waveform and the Fourier coefficient of the sub-waveform; the bandwidth of each sub-waveform is the same; The solver module is implemented as follows: Initialize the time width matrix of the sub-waveform, the time-frequency trend matrix of the sub-waveform, and the number of iterations i and j; Determine the time-frequency trend matrix of the i-th iteration; wherein each row in the time-width matrix of the i-th iteration is One of the possible cases of all permutations, corresponding to The time-frequency trend of one of the quasi-orthogonal radar waveforms; Optimizing the time width matrix of the i-th iteration; wherein the time width matrix of the i-th iteration is determined based on the time width matrix optimized in the i-1-th iteration; Optimize the Fourier coefficient matrix of the i-th iteration; where k time widths are The quasi-orthogonal radar waveform is used as a reference waveform, and after optimizing the Fourier coefficients of the reference waveform, the Fourier coefficients of the reference waveform closest to the time width of each sub-waveform are determined as the Fourier coefficients of the sub-waveform; like , then the optimization ends; otherwise, Then, repeat the above steps.

3. An electronic device, wherein: The method comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the method of claim 1.

4. A computer-readable storage medium having computer instructions stored thereon, wherein: When the computer instructions are executed by a processor, the method according to claim 1 is implemented.

5. A computer program product comprising computer instructions, wherein: When the computer instructions are executed by a processor, the method according to claim 1 is implemented.

Citation Information

Patent Citations

  • Method and device for generating orthogonal nonlinear frequency modulation signal

    CN109343059A

  • Method and system for determining nonlinearity of LFM pulse signal frequency modulation in radar equipment

    CN109490857A

  • Low-sidelobe nonlinear frequency modulation waveform optimization method based on Fourier series

    CN113625230A

  • Method for range ambiguity suppression based on multi-degree-of-freedom frequency modulation signal

    US20190346556A1