Joint multipath mitigation method based on conformal mimo radar waveform and array-transmitting weighting

Abstract

The application discloses a kind of based on conformal MIMO radar waveform and array transceiver weighting joint multipath suppression method, belong to MIMO radar multi-channel waveform design and signal receiving processing field.The present application utilizes environmental information or prior information of multipath signal to carry out waveform and transceiver joint weighting multipath suppression, improve the identification accuracy of multipath component, and fully suppress the undesired multi-path signal, the present application can significantly improve target detection accuracy under indoor equal intensity multipath condition;The present application starts from the radar waveform design and array design angle, and the complexity is low, and real-time is good;The present application fully suppresses multipath echo, and improves output signal-to-noise ratio by multi-channel coherence superposition, and anti-interference performance is obviously improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of MIMO radar multi-channel waveform design and signal receiving processing, and particularly relates to a method for joint multipath suppression based on conformal MIMO radar waveform and array transmitting-receiving weighting. BACKGROUND

[0002] Multiple-input multiple-output (MIMO) technology is a technique that uses multiple transmitting and receiving antennas to increase the performance of a communication system. It was originally widely used in the field of communication, but was later introduced into the fields of radar, radio spectrum sensing and others. Conformal array is an antenna arrangement in which antennas are arranged on an irregular array that adapts to the shape of the object surface. Unlike traditional regular arrays, the arrangement of antennas in a conformal array can be adjusted according to a specific curved surface, making it better adapt to the shape of the carrier. Conformal MIMO radar is a radar system that uses multiple transmitting and receiving antennas on a conformal carrier. Compared with traditional single-antenna radars, conformal MIMO radars have some significant advantages. Among them, the conformal feature allows the radar to adapt to irregular curved surfaces, such as aircraft wings, ship hulls or vehicle surfaces. Such a design allows the antennas to be better integrated into the carrier, reducing the impact on the carrier's aerodynamic performance and appearance. At the same time, the antenna layout of the conformal array can better utilize the surface of the carrier, improving the spatial efficiency of the array. On the other hand, conformal MIMO radar improves the performance of the radar system by fully utilizing the advantages of multiple antennas, including spatial resolution, anti-interference ability, detection performance and adaptability to complex environments. This makes conformal MIMO radar have wide potential and application prospects in various application scenarios. For example, conformal MIMO radar can effectively suppress multipath effects by introducing time delay differences between different antennas, improving the radar system's detection and tracking performance of targets. And conformal MIMO radar can reduce the sensitivity to interference on a single path by utilizing multiple antennas to receive signals and performing signal processing at the receiving end. This makes the performance of conformal MIMO radar more robust in complex electromagnetic environments. These advantages provide a technical approach to solving the impact of multipath effects in the field of non-contact vital sign monitoring technology and automatic driving technology. SUMMARY

[0003] In view of the deficiencies of the prior art, the present application discloses a method for joint multipath suppression based on conformal MIMO radar waveform and array transmitting-receiving weighting, to reduce the processing complexity of multipath suppression and improve the output signal-to-noise ratio, thereby improving the anti-interference performance of conformal MIMO radar.

[0004] The technical scheme adopted by the present application is as follows:

[0005] The method for joint multipath suppression based on conformal MIMO radar waveform and array transmitting-receiving weighting comprises the following steps:

[0006] Step S1, input the element positions of the conformal MIMO radar array, the number of receiving units of the conformal MIMO radar array is defined as N, and the number of transmitting units is defined as M;

[0007] Step S2, judge whether there is prior information, if not, go to step S3, otherwise, go to step S4;

[0008] Step S3, obtain the range-angle prior information of the current frame: first, perform joint estimation of the transmit spatial frequency and the receive spatial frequency, then perform multi-path component identification, and obtain the range-angle prior based on the multi-path component identification; then perform step S4 based on the range-angle prior of the current frame.

[0009] Step S4, perform target state parameter estimation on the next frame based on the range-angle prior information of the current frame, which specifically includes: transmit weighting solution of the next frame, transmit waveform parameter optimization of the next frame, and receive space-time filtering based on the waveform parameter optimization result of the current frame, and finally obtain the target state parameter estimation result.

[0010] Further, step S3 specifically includes:

[0011] Step S301, obtain the distance information of the target, after multi-channel coherent matched filtering or dechirp processing of the received N-channel signals at the receiving end, perform constant false alarm rate detection to obtain the distance information of the target;

[0012] Step S302, jointly estimate the transmit spatial frequency and the receive spatial frequency at the distance unit of the target:

[0013]

[0014] Wherein, f T , f R respectively represent the transmit spatial frequency vector and the receive spatial frequency vector, represents the transmit spatial angle, represents the receive spatial angle; a R (·) and a R (·) are the transmit steering vector and the receive steering vector respectively, and the superscript H represents the conjugate transpose of the matrix, represents the sampling covariance matrix of the receiving end estimated according to the MN virtual channel data of the receiving end;

[0015] Step S303, obtain the range-angle prior information: find the peak value in P capon (f T ,f R ) corresponding to the angle set In the angle set, find the set with the same transmit spatial angle and receive spatial angle, that is, a set of This is the distance-angle prior information.

[0016] Further, the step S4 specifically comprises:

[0017] Step S401, based on the radar echo data of the current frame, the transmit weight Target azimuth information Calculate the virtual receiving channel weight of the current frame: Wherein, the superscript q is used to identify the current frame;

[0018] Step S402, according to the transmit waveform parameter s(t; t sh ) of the current frame, the equivalent transmit steering vector of the q+1 frame capable of suppressing multipath components is solved

[0019] Wherein, the equivalent cross-correlation vector Y ss (τ l ,ω l ) represents the value of the Fourier transform of the cross-correlation matrix of the transmit baseband waveform of the lth time delay component at the frequency ω l , and represents the double-path delay of the lth multipath, c represents the speed of light, r l represents the length of the lth multipath, Y ss (τ0, ω0) represents the value of the Fourier transform of the cross-correlation matrix of the transmit baseband waveform of the first time delay component at the frequency ω0, β l represents the reflection coefficient of the lth multipath, 1 M is an M-order all-1 vector, I M represents an M-order unit matrix, and σ represents the power of Gaussian white noise;

[0020] Step S403, solving the transmit weighting parameter of the q+1 frame: Symbol represents point division operation;

[0021] Step S404, taking the cyclic delay t sh as the optimization variable of the transmit waveform parameter s(t; t sh ), substituting the equivalent transmit steering vector of the q+1 frame into the waveform optimization objective function to solve t sh ; and then obtaining the adjusted transmit waveform parameter s(t; t sh ) of the conformal MIMO radar array based on the currently solved t sh , for use in the spatial domain filtering of the next frame (q+1 frame) to obtain the corresponding target parameter;

[0022] The waveform optimization objective function is specifically:

[0023]

[0024] wherein, represents an equivalent transmit steering vector, and γ is a constant normalization factor;

[0025] Preferably, the alternating projection method (ADMM) or genetic algorithm is used to optimize and solve t sh ;

[0026] Step S405, based on the virtual receiving channel weight of the current frame at the receiving end spatial domain filtering is performed on the multi-channel data of the qth frame;

[0027] Step S406, phase and frequency estimation is performed on the spatial domain filtered data to obtain target parameters.

[0028] Further, the expression of Υ ss (τ l , ω l ) is specifically:

[0029]

[0030] wherein, T c is a pulse repetition period or a duration of a frame, s(t) = [s1(t), …, s M (t)] T is a vector representation of the transmit waveform at time t, s m (t) is a baseband complex waveform transmitted by the mth transmit unit, m = 0, …, M-1, and M is the number of transmit units; is a Fourier transform of a time series, ω is an angular frequency parameter of the Fourier transform, is a correlation function between transmit waveforms, and the transmit unit index m' = 0, …, M-1.

[0031] The angular frequency parameter of the Fourier transform is ω = 2πf, wherein f is the frequency of the transmit waveform.

[0032] Further, the constant normalization factor γ is specifically set as:

[0033]

[0034] wherein, μ represents the frequency modulation slope of the LFM waveform, and L represents the number of multipath paths.

[0035] Further, the expression of the transmit waveform parameter s(t; t sh ) is:

[0036]

[0037] wherein s m (t) is the baseband complex waveform transmitted by the mthtransmitting unit, m = 0, …, M-1, M is the number of transmitting units, is the time cyclic shift.

[0038] The technical solutions provided by the present application bring at least the following beneficial effects:

[0039] (1) The present application can use knowledge assistance (environmental information or prior information of multipath signals) to perform waveform and transceiver joint weighting multipath suppression, improve the identification accuracy of multipath components, and fully suppress undesired multipath signals. The traditional independent component analysis (ICA) method suppresses multipath signals by estimating the propagation channel, but is only suitable for weak multipath environments, while the waveform and transceiver weighting joint design method of the radar sensor in the present application still has high detection accuracy under strong multipath conditions such as indoors;

[0040] (2) The existing technology usually uses an autoregressive model to recover signals damaged by mutual interference, but this method is an offline processing method, which has the problems of large amount of calculation and inability to process in real time, while the present application starts from the aspects of radar waveform design and array design, has low complexity and good real-time performance;

[0041] (3) Compared with the traditional method, the present application fully suppresses multipath echoes and improves the output signal-to-noise ratio through multi-channel coherent superposition, and the anti-interference performance is obviously improved. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0043] Figure 1 Configuration diagram of conformal MIMO array.

[0044] Figure 2 Weighted joint multipath suppression processing diagram adopted in the embodiment of the present application.

[0045] Figure 3 Indoor human vital sign monitoring multipath effect diagram in the embodiment of the present application.

[0046] Figure 4 Target at 2m and other three strong multipath interference echo distance distribution diagram in the embodiment of the present application.

[0047] Figure 5 The signal distance distribution diagram after multipath suppression using the method of this embodiment is shown compared to the signal that only receives beamforming processing.

[0048] Figure 6 This is a graph showing the estimated and true values ​​of heart rate and respiration in an embodiment of the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be described in detail and completely below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Generally, the components of the embodiments of the present invention described and shown in the accompanying drawings can be arranged and designed using different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the present invention.

[0050] To facilitate understanding of the embodiments of the present invention, a brief description of the conformal MIMO radar is provided below:

[0051] Assuming the radar is a co-located MIMO radar, such as Figure 1 As shown. A rectangular coordinate system is established with the phase center of the antenna array as the origin. The array consists of M transmitting elements and N receiving elements. The distances from the m-th transmitting element, m = 1, ..., M, and the n-th receiving element, n = 1, ..., N, to the origin can be expressed as follows:

[0052]

[0053]

[0054] Among them, e x e y e z Let x, y, and z represent the unit direction vectors of the x, y, and z axes, respectively. and These represent the three-dimensional rectangular coordinates of the transmitting and receiving array elements, respectively.

[0055] Let the position of the far-field point target be... Where θ represents the pitch angle. Let be the azimuth angle, and r be the distance between the target and the origin. Assume the target moves towards the radar at a constant speed v for a relatively short period. Indicates the direction of the target A unit vector on a coordinate plane can be represented as follows in a rectangular coordinate system:

[0056]

[0057] Let s m (t) is the baseband complex waveform transmitted by the mthtransmitting unit with unit energy, i.e.,

[0058]

[0059] The superscript "*" represents the conjugate transpose operation.

[0060] At the same time, the MIMO waveform also satisfies the orthogonality condition between transmitting waveforms, i.e.,

[0061]

[0062] where T c is the waveform duration, if the linear frequency modulation waveform (LFM) or the frequency modulation continuous waveform (FMCW) is used at the transmitting end, the baseband waveform can be represented as:

[0063]

[0064] where μ m represents the frequency modulation rate of the LFM waveform transmitted by the mthtransmitting element, defined as a rectangular function, limiting the duration of the waveform:

[0065]

[0066] Assuming is the average transmitting power of each transmitting unit, and f0is the reference carrier frequency of each unit, the mthsignal can be represented as Here, the effect of radio frequency impedance is ignored and it is assumed that the terminal impedance is equal to 1. For convenience of representation, the transmitting waveform vector is written in the form of a matrix and a vector: x(t) = [x0(t), x1(t), …, x M-1 (t)] T , (·) T represents the transpose operation of a matrix or a vector, then the receiving signal vector of a far-field point target can be written as:

[0067]

[0068] wherein represents an Mx1-dimensional complex domain, β0represents the complex amplitude factor equivalent to the channel and target reflection coefficient, which represents the attenuation constant of the echo energy in the time-invariant channel. τ(r, v) = 2d(t) / c is the delay caused by the target distance r and the target velocity v, d(t) = r + vt represents the distance of a single target at time t, and c represents the speed of light. and are the transmitting and receiving steering vectors, respectively, which are represented as:

[0069]

[0070]

[0071] Meanwhile, according to the radar equation, the echo amplitude factor of the distance r can be approximately quantified as:

[0072]

[0073] In the formula, G F and G R are the transmitting and receiving antenna gains, generally 1 for the microstrip array antenna, σ T is the target radar cross section (RCS), λ0=c / f0 is the wavelength corresponding to the central carrier frequency f0. φ∈(0, 2π) is a random variable, representing the random phase change of the channel.

[0074] Embodiments of the present application start from the radar waveform and the transmitting and receiving array weighting, and propose a joint multipath suppression method based on the conformal MIMO radar waveform and the array transmitting and receiving weighting, to suppress and eliminate the multipath interference. The specific implementation process of the method of the embodiments of the present application is as follows:

[0075] Step S1, a discrete sampling model of a multipath echo is established:

[0076] According to the basic theory of the conformal MIMO radar, the array configuration of M co-located transmitting and N co-located receiving is adopted, and the echo signal is fast-time sampled, so that the discrete echo signal can be obtained. Assuming that there is a single target in the space, the I fast-time sampling of the qth frame (here, a frame refers to a coherent processing interval or a pulse repetition period) echo is performed, and the discrete time signal obtained is as follows:

[0077]

[0078] Among them, T c represents the duration of a chirp (a pulse repetition period or a frame), T s represents the sampling interval of the fast time. v represents the relative speed between the target and the radar, represents the discrete sampling of the received independent and identically distributed complex Gaussian white noise vector. In a typical multipath environment, L-1 multipaths of a typical target are combined with the direct path. It can be intuitively explained that L point targets of different azimuths and distances share the same target micro-motion information. Assuming that the channel response change can be ignored within the coherent processing interval. Then, the multipath channel impulse response can be represented as:

[0079]

[0080] Among them, δ(t) represents the Dirac function. τl β represents the time delay of the l-th path. l This represents the reflection coefficient of the l-th multipath path.

[0081] The received signal from the multipath channel is then modeled as a convolution of the channel impulse response and the transmitted signal. Therefore, the echo signal of a single target in a multipath environment can be expressed as:

[0082]

[0083] in, Let N represent the N-dimensional real number field.

[0084] Step S2, Virtual multi-channel data generation:

[0085] The N-channel signal received at the receiving end can be represented as:

[0086]

[0087] After down-converting the received N-channel signals and performing multi-channel matched filtering using M orthogonal transmitted waveforms, a virtual receiver matrix of MN channels can be obtained. There are two cases: In the first case, if the radar uses a pulse system, the receiver uses multi-channel coherent matched filtering, meaning the processed virtual channel data can be represented as:

[0088]

[0089] in, ⊙ represents the Kronecker product, and ⊙ represents the Hadama product. N Let represent an N-dimensional vector of all ones. Let c represent the two-way delay of the l-th multipath path, and r be the speed of light. l β represents the length of the l-th multipath path. l Let r represent the reflection coefficient of the l-th multipath path. xx [i-τ l The ] represents the sampling of the autocorrelation function of the transmitted waveform vector at time i, specifically expressed as:

[0090]

[0091] Where s(t) = [s1(t), ..., s M (t)] T This represents the vector representation of the transmitted waveform at time t.

[0092]

[0093] The second scenario: If the transmitter uses a time-division multiplexing frequency-modulated continuous wave system to ensure orthogonality, the baseband waveform is represented as:

[0094]

[0095] The length of one frame at this time is MT c The receiver should adopt the dechirp processing to simplify the complex matched filter structure, and the multi-channel data of the received echo can be expressed as:

[0096]

[0097]

[0098] τ(r, v) = 2d(t) / c is the delay caused by the target distance r and the target velocity v, and d(t) = r + vt represents the distance of a single target at t time. is a rectangular function.

[0099] Step S3, joint estimation of transmit spatial frequency and receive spatial frequency.

[0100] Rewrite the form of transmit and receive steering vectors:

[0101]

[0102] Considering the structure of transmit and receive conformal steering vectors, define The transmit spatial frequency vector and the receive spatial frequency vector in the direction are:

[0103]

[0104] Using the MN virtual channel data of the receiver obtained in the second step, the sampling covariance matrix of the receiver is constructed as:

[0105]

[0106] Using the Capon spectrum principle, the transmit spatial frequency and the receive spatial frequency (transmit angle and receive angle) can be jointly estimated, that is,

[0107]

[0108] Where the angle and represent the one-to-one correspondence with the spatial frequency f R and f T .

[0109] Find the peak value in P capon (f T , f R ) corresponding to the angle set: Find the set of the same transmit spatial angle and receive spatial angle in this set, that is, the set of , which is represented as: The azimuth angle of the real target.

[0110] Step S4, joint multipath suppression processing based on waveform and transceiving weights.

[0111] In order to suppress the multipath components in the environment, it can be expressed as maximizing the target signal energy while minimizing the energy of the multipath components, so the embodiment of the application adopts the way of maximizing the signal-to-interference-and-noise ratio (SINR) function f(w R , w F , s) to suppress the multipath signal, thereby obtaining the maximum detection probability of the target. According to the known prior information of the environment multipath The objective function can thus be expressed as the following optimization problem:

[0112] max f(w R , w F , s)

[0113] w R , w F , s (25)

[0114] wherein the variables respectively represent the received virtual multi-channel weight, the transmitted weight and the multi-channel transmitted baseband waveform matrix, and the SINR function can be expressed as:

[0115]

[0116] wherein z(t) is the received processed virtual multi-channel signal vector, and

[0117]

[0118] If a time division multiplexing frequency modulated continuous wave system is adopted, according to the expression of Fourier transform, the SINR function can be expressed as:

[0119]

[0120] wherein ω l′ = 2πμτ l′ is the intermediate frequency corresponding to different multipath ranges (which can be explained as maximizing the frequency energy of the intermediate frequency signal at the target distance, while minimizing the energy of the intermediate frequency signal at the multipath distance), and it is assumed that the multi-channel transmitted frequency modulated continuous wave signal has the same frequency modulation rate μ. And:

[0121]

[0122] Step S401, receive weight solving:

[0123] According to the minimum variance distortionless response (MVDR) theorem, the receive weight can be expressed as:

[0124]

[0125] wherein, is a sample covariance matrix (SCM), which is usually calculated by a discrete summation method, i.e.: A set of azimuth angles where the real target is located.

[0126] Step S402, range of main lobe multipath suppression (transmit waveform parameter and transmit weighting alternately optimized to solve):

[0127] Substitute formula (30) into the SINR target function of FMCW, which can be written as:

[0128]

[0129] This step only needs to derive the expression of the frequency-modulated continuous wave system. The expression of the pulse system can be obtained by setting w l′ = 0 in the above formula, l' = 0, …, L-1. Intuitively, if you want to maximize the output SINR, you need to minimize the energy output of the multipath signal while ensuring constant gain of the desired signal. It can also be explained that the goal of formula (31) is to find the smallest sidelobe waveform, so according to a large number of researches in the field of FMCW radar waveform design, the embodiment of the present application selects a quasi-orthogonal waveform, a cyclic FMCW (C-FMCW), as the initial baseband waveform, and the cyclic time delay t sh is used as a waveform optimization variable. The C-FMCW waveform has good range sidelobe suppression performance and has little effect on range resolution, which is expressed as:

[0130]

[0131] In the formula, τ max is the maximum echo delay, representing the maximum detection distance. represents the time cyclic shift. Assuming that the signals with a difference greater than the main lobe width of the adaptive beamforming from the direct path arrival angle are removed by spatial filtering through formula (30), it can be approximately considered that the multipath signals within the main lobe width range have the same arrival angle. Therefore, after certain theoretical derivation, it is obtained that the optimization problem (31) can be simplified as:

[0132]

[0133] Considering the coupling characteristics of the transmit weighting and the transmit array steering vector, for the purpose of simplifying the derivation, a new equivalent transmit steering vector is defined as: wherein 1 Mis a M-order all-one vector, and is the cross-correlation vector γ is a constant normalization factor, which does not affect the optimization result, and thus can be ignored.

[0134] Through operation derivation, the following can be obtained:

[0135]

[0136] wherein, is equivalent to the cross-correlation vector, and γ ss (τ l , ω l ) is a value of a Fourier transform of a cross-correlation matrix of a transmit baseband waveform of an lth time delay component at w l frequency, and l represents a time delay of an lth multipath signal, and is denoted as:

[0137]

[0138] wherein, is a correlation function between transmit waveforms. denotes a Fourier transform on a time sequence.

[0139] It is noted that the optimization problem is a minimum variance distortion response problem in structure, and thus a closed-form solution of the original optimization problem is:

[0140]

[0141] wherein, σ represents a Gaussian white noise power. M denotes an M-order unit matrix. Because

[0142] Thus, the transmit weight can be represented as:

[0143] Herein, denotes a point division operation, such as:

[0144] Substitute into the original optimization problem, and an alternating projection method (ADMM) can be used to iteratively optimize the waveform parameter t sh .

[0145] As a possible implementation manner, referring to Figure 2 , the embodiment of the present application provides an implementation step of the method for joint multipath suppression based on conformal MIMO radar waveform and array transceiver weight, which comprises the following steps:

[0146] Step S1, determining an array parameter, and giving a conformal MIMO array element position:

[0147] Step S2, if no prior information is available, then go to step S3, otherwise go to step S4;

[0148] Step S3, obtain the distance-angle prior information of the current frame (qth frame), first perform joint estimation of the transmit spatial frequency and the receive spatial frequency, then perform multi-path component identification, and finally obtain the distance-angle prior; then perform step S4 based on the distance-angle prior of the current frame;

[0149] Step S4, perform target state parameter estimation on the next frame (q+1th frame) based on the distance-angle prior information of the current frame, which specifically includes: virtual receive channel weighting, q+1th frame transmit weighting solving, q+1th frame transmit waveform optimization; and perform receive space-time filtering based on the waveform optimization result of the q+1th frame, and finally obtain the target state parameter estimation result.

[0150] Preferably, step S3 specifically includes:

[0151] Step S301, obtain the distance information of the target, after multi-channel coherent matched filtering or dechirp processing of the N-channel signals received at the receiving end, CFAR detection is performed to obtain the distance information of the target;

[0152] Step S302, jointly estimate the transmit spatial frequency and the receive spatial frequency at the distance unit of the target:

[0153]

[0154] Step S303, obtain the distance-angle prior information: find the peak value in P capon (f T , f R ) and the corresponding angle set Find the set of the same transmit spatial angle and receive spatial angle in this set, i.e. the set of , denoted as to obtain the angle information of the real target.

[0155] Preferably, step S4 specifically includes:

[0156] Step S401, assume that the qth frame of radar echo data is obtained, and the qth frame of transmit weighting According to the target orientation information , calculate the virtual receive channel weighting of the current frame:

[0157] Step S402, according to the transmit waveform parameters s(t; t sh ) of the current frame, obtain the equivalent transmit steering vector of the q+1th frame that can suppress the multi-path component

[0158] wherein:

[0159]

[0160]

[0161] wherein, the first frame of the transmit waveform parameter s(t; t sh ) can be a preset value, and the transmit waveform parameter s(t; t sh ) of the qth frame is obtained based on step S404 optimization;

[0162] Step S403, solving the transmit weight of the q+1th frame,

[0163] Step S404, solving the MIMO waveform parameter optimization problem of the q+1th frame, according to the quasi-orthogonal waveform C-FMCW characteristics, the cyclic delay t sh is taken as the waveform optimization variable s(t; t sh ); and is substituted into the optimization problem to obtain:

[0164]

[0165] The alternating projection method (ADMM) or genetic algorithm is used to optimize and solve the waveform parameter t sh ;

[0166] Step S405, spatial domain filtering, using the receive weight to perform spatial domain filtering on the multi-channel data of the qth frame at the receiving end;

[0167] Step S406, target state parameter estimation, performing phase and frequency estimation on the filtered data to obtain the target parameter.

[0168] Heart rate and respiration monitoring is a key indicator for assessing health, stress, and physical fitness. This technology can remotely monitor isolated family members or hospitals in the case of viral infection or epidemic risk, reducing the likelihood of cross-infection. However, such devices are prone to forgetting to wear or charging, and there are many restrictions for patients with conditions such as "sleep apnea" that require long-term continuous monitoring of vital signs. These limitations have led to the development of non-contact heart rate monitoring technology in some sense, and MIMO radar-based non-contact vital sign monitoring technology has received widespread attention. However, there are some problems in the process of heart rate and respiration monitoring based on MIMO radar. When there are two or more paths from the transmitting antenna to the receiving antenna, multipath effects occur. Multipath depends on the room structure and whether a target is present. Therefore, even small objects moving in the environment will change the multipath reflections, causing positioning errors and vital sign extraction errors, and multipath effects pose a huge challenge to MIMO radar-based non-contact vital sign monitoring.

[0169] Autonomous driving is an advanced vehicle technology designed to enable vehicles to perform driving tasks without human driver intervention. Using sensor data, autonomous driving systems can identify roads, vehicles, pedestrians, traffic signs, and other obstacles, and adaptively perceive and recognize the environment. However, the presence of multipath effects has a serious impact on autonomous driving tasks. For example: multipath effects cause signals to experience different paths during transmission, making it possible for the receiving end to receive multiple signals that have traveled different paths. This can cause the phase of the received signal to superimpose and interfere, resulting in positioning errors. The positioning system may incorrectly calculate the vehicle's position, affecting the accuracy of navigation. In high-speed autonomous driving scenarios, multipath effects can become even more complex. Vehicles may receive multiple reflected signals at different positions, speeds, and directions, making it more difficult for the positioning system to track the vehicle's accurate position, especially in urban environments or areas with a large number of obstacles.

[0170] The method for joint multipath suppression based on conformal MIMO radar waveform and array transceiver weighting provided by the embodiments of the present application can be applied to heart rate and respiration monitoring to achieve indoor human vital sign monitoring tasks; it can also be applied to autonomous driving.

[0171] Taking the application of the embodiments of the present application to human vital sign monitoring tasks in an indoor multipath environment as an example, the specific application implementation process is described as follows:

[0172] A quasi-orthogonal waveform, a cyclic linear continuous frequency modulation wave signal, is used as the initial baseband waveform t sh = 0.03 ms, and the working frequency f0 of the radar system used is 77 GHz, and the signal duration T c= 0.14ms, frame time is 50ms, sweep bandwidth B = 4GHz, the number of transmitting arrays M = 3, the array element spacing is d T = 2λ0= 8mm, the number of receiving arrays N = 4, the array element spacing is d R = 0.5λ0= 2mm.

[0173] Suppose that the target and three multipath components are located in the azimuth plane θ = [10, 11, 13, 21], ignoring the angle difference of the azimuth plane, the distances of the real target and three multipath components are d0= [2, 2.5, 3, 3.5], and the schematic diagram is shown in Figure 3

[0174] The specific implementation steps are as follows:

[0175] (1) Joint estimation of transmitting spatial frequency and receiving spatial frequency;

[0176] The received samples are processed by multi-channel, CFAR (constant false alarm rate) detection is performed on the 4 receiving channels to obtain the distance information of the target. Joint estimation of transmitting spatial frequency and receiving spatial frequency is performed at the distance unit of the target to identify the target signal and the multipath signal, and to give the prior information of distance and angle.

[0177] (2) Multipath suppression or elimination processing;

[0178] Suppose that the qth frame of radar echo data is obtained, according to the prior information of the multipath signal and the optimization problem, the receiving weight w R of the qth frame, the transmitting waveform s(t; t sh ), the transmitting weight w T of the q+1th frame are sequentially solved.

[0179] The spatial domain filtering is performed on the received multi-channel data by using w R to obtain the filtered data, and finally the human vital sign state parameter estimation is performed.

[0180] Figure 4 The echo distance distribution diagram of the target in this embodiment at 2m and other three strong multipath interferences is shown, the amplitude of the real echo of the target is smaller compared with the multipath components, and the real signal cannot be filtered out; Figure 5 The signal distance distribution diagram after multipath suppression of this embodiment is shown, if the echo signal is not processed by the multipath suppression algorithm, the two multipath interferences (i.e. 11° and 13°) of the target in the angle resolution unit cannot be filtered out. In addition, the multipath interference (i.e. 21°) outside the angle resolution unit is not completely filtered out, and its strength is only slightly lower than that of the unprocessed echo. By using the multipath suppression method of this embodiment, the multipath echo can be suppressed, and the output SINR can be improved by multi-channel coherent superposition. Figure 6 ​For the heartbeat and respiration estimation and true value curve of the embodiment, the final mean square error of the calculated respiration and heartbeat is 0.608 and 3.164 respectively, which proves the effectiveness of the method of the embodiment.

[0181] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

[0182] The above only describes some embodiments of the present application. For those skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, which are all within the protection scope of the present application.

Claims

1. A method for joint multipath mitigation based on conformal MIMO radar waveforms and array-transmitting weighting, characterized in that, The method comprises the following steps: Step S1, inputting the positions of array elements of a conformal MIMO radar array, the number of receiving units of the conformal MIMO radar array being defined as N, and the number of transmitting units being defined as M; Step S2, judging whether there is prior information, if not, entering step S3, otherwise, entering step S4; Step S3, acquiring the range-angle prior information of the current frame: firstly, performing joint estimation of transmitting spatial frequency and receiving spatial frequency, then performing multi-path component identification, and acquiring the range-angle prior based on the multi-path component identification; and then performing step S4 based on the range-angle prior of the current frame; Step S4, performing target state parameter estimation on the next frame based on the range-angle prior information of the current frame, specifically including: solving the transmitting weighting of the next frame, optimizing the transmitting waveform parameters of the next frame, and performing receiving space-time filtering based on the optimization result of the waveform parameters of the current frame, and finally obtaining the target state parameter estimation result; Wherein, step S4 specifically includes: Step S401, based on the radar echo data of the current frame, the transmitting weight , target azimuth information Calculate the virtual receiving channel weight of the current frame: ； wherein, represents a sample covariance matrix of the receiving end estimated according to the MN virtual channel data of the receiving end; and are a transmit steering vector and a receive steering vector, respectively, and the superscript H represents a conjugate transpose of a matrix; the superscript q is used to identify a current frame. Step S402, according to the transmitting waveform parameter of the current frame , the equivalent transmitting steering vector of the q+1th frame which can suppress the multipath component is calculated: ； where the equivalent cross-correlation vector , denotes the value of the Fourier transform of the cross-correlation matrix of the transmit baseband waveforms of the th delay component at frequency , denotes the two-way delay of the th multipath path, denotes the speed of light, denotes the length of the th multipath path, denotes the reflection coefficient of the th multipath path, is an M-order all-one vector, denotes an M-order identity matrix, denotes the Gaussian white noise power; Step S403, solving the transmitting weighting parameters of the q+1th frame: wherein the symbol represents a point division operation; Step S404, the cycle delay As the optimization variable of the transmit waveform parameter , the equivalent transmit steering vector of the q+1th frame is substituted into the waveform optimization objective function to solve ; based on the current solved , the adjusted transmit waveform parameter of the conformal MIMO radar array is obtained ; The waveform optimization objective function is specifically: wherein denotes the equivalent transmit steering vector, and γ is a constant normalization factor. Step S405, the receiving end performs virtual receiving channel weight based on the current frame spatially filtering the multi-channel data of the qth frame; Step S406, performing phase and frequency estimation on the data filtered in the space domain to obtain the target parameters.

2. The method of claim 1, wherein, Step S3 specifically includes: Step S301, solving the distance information of the target, after multi-channel coherent matched filtering or dechirp processing is performed on the N-channel signals received at the receiving end, constant false alarm rate detection is performed to obtain the distance information of the target; Step S302, jointly estimating the transmitting spatial frequency and the receiving spatial frequency at the distance unit of the target: wherein , respectively represent a transmit spatial frequency vector and a receive spatial frequency vector, represents a receive spatial angle, represents a transmit spatial angle; Step S303, obtain prior information on distance and angle: find The set of angles corresponding to the peaks in the middle Within this set of angles, find the set where the transmit space angle and the receive space angle are the same, denoted as... This yields the angle information of the actual target location, i.e., the prior information of distance and angle.

3. The method of claim 1, wherein, Solving by using alternate projection method or genetic algorithm optimization .

4. The method of claim 1, wherein, The specific expression is: in, The duration of one pulse repetition cycle or one frame. Let be the vector representation of the transmitted waveform at time t. For the first The baseband complex waveforms transmitted by each transmitting unit, m=0,…,M-1, For the Fourier transform of the time series, The angular frequency parameter of the Fourier transform. It is the correlation function between the transmitted waveforms, and the transmission unit index m'=0,…,M-1.

5. The method of claim 1, wherein, The constant normalization factor γ is specifically set as: where μ denotes a frequency modulation slope of the waveform, L denotes a number of multipath paths, is the duration of one pulse repetition period or one frame.

6. The method of claim 1, wherein, Transmit waveform parameters The expression for the transmit waveform parameters is: wherein is the baseband complex waveform transmitted by the mth transmitting element, m = 0,..., M - 1, is the time cyclic shift.

Images