A parameter estimation method and device of an airborne distributed MIMO radar system and a medium

Abstract

The embodiment of the application discloses a kind of parameter estimation method, device and medium of airborne distributed MIMO radar system, belong to radar signal processing technical field;Including: the system model of airborne distributed MIMO radar is constructed;The echo signal received by receiver airborne platform is matched filtering and Doppler processing, and the rough estimation of the distance and Doppler shift of target is obtained;According to the rough estimation of distance and Doppler shift, the maximum likelihood function of echo signal is solved by one-dimensional search of time delay and Doppler shift respectively, and the optimal estimation of time delay and Doppler shift is obtained respectively.The method can give the mathematical model of airborne distributed MIMO radar system, based on maximum likelihood algorithm and the rough prior information of time delay-Doppler shift, the target parameter is accurately estimated by one-dimensional search of parameter, and the operation amount of parameter estimation is reduced.

Description

TECHNICAL FIELD

[0001] The embodiment of the application relates to the technical field of radar signal processing, in particular to a parameter estimation method and device of an airborne distributed MIMO radar system and a medium. BACKGROUND

[0002] Radar (radar) is used to detect target information by using the reflection principle of electromagnetic waves. At the transmitting end, the radar detects a target of interest by transmitting electromagnetic waves, and after a short time delay, the receiving end receives the echo from the target reflection. The echo signal contains key information about the target, such as distance, angle and speed, etc. The receiving antenna of the radar sends the received reflected wave to the subsequent signal processing device for signal processing operation, and some information about the target can be extracted.

[0003] The concept of multiple-input and multiple-output (MIMO) originates from wireless communication technology. The idea of this technology is that the signal transmitting and receiving ends are both arranged with multiple antennas, so that different transceiver antennas produce multiple independent channels, greatly improving the information transmission efficiency and reducing the bit error rate. In the field of wireless communication, MIMO technology has shown outstanding ability in suppressing interference and overcoming channel fading.

[0004] With the development of the traditional radar technology being basically perfect, the MIMO radar has shown great advantages in improving target detection capability, parameter estimation accuracy and anti-interference, etc. SUMMARY

[0005] Therefore, the embodiment of the application expects to provide a parameter estimation method and device of an airborne distributed MIMO radar system and a medium. The mathematical model of the airborne distributed MIMO radar system can be given, and when the time delay and Doppler shift parameters are estimated, the target parameters are accurately estimated by one-dimensional search of the parameters based on the maximum likelihood algorithm and the rough prior information of the range-Doppler shift, so that the operation amount of the parameter estimation is reduced.

[0006] The technical scheme of the embodiment of the application is as follows:

[0007] In a first aspect, the embodiment of the application provides a parameter estimation method of an airborne distributed MIMO radar system, comprising:

[0008] A system model of the airborne distributed MIMO radar is constructed, the system includes at least two transmitting airborne platforms, at least two receiving airborne platforms and at least one target; wherein each transmitting airborne platform and each receiving airborne platform carries a centralized MIMO antenna;

[0009] performing matched filtering and Doppler processing on the echo signal received by the airborne platform based on the system model to obtain a rough estimation of the distance and Doppler shift of the target;

[0010] solving a maximum likelihood function of the echo signal through one-dimensional search of time delay and Doppler shift respectively based on the rough estimation of the distance and Doppler shift to obtain optimal estimations of the time delay and Doppler shift respectively.

[0011] In a second aspect, an embodiment of the present application provides a device for parameter estimation of an airborne distributed MIMO radar system, the device comprising: a system model construction part, a rough estimation part, and an optimal estimation part; wherein,

[0012] the system model construction part is configured to construct a system model of the airborne distributed MIMO radar, the system comprising at least two transmitting airborne platforms, at least two receiving airborne platforms, and at least one target; wherein each of the transmitting airborne platforms and each of the receiving airborne platforms carries a centralized MIMO antenna;

[0013] the rough estimation part is configured to perform matched filtering and Doppler processing on the echo signal received by the airborne platform based on the system model to obtain a rough estimation of the distance and Doppler shift of the target;

[0014] the optimal estimation part is configured to solve a maximum likelihood function of the echo signal through one-dimensional search of time delay and Doppler shift respectively based on the rough estimation of the distance and Doppler shift to obtain optimal estimations of the time delay and Doppler shift respectively.

[0015] In a third aspect, an embodiment of the present application provides a computer storage medium, characterized in that the computer storage medium stores a program for parameter estimation of an airborne distributed MIMO radar system, and the program for parameter estimation of the airborne distributed MIMO radar system, when executed by at least one processor, implements the steps of the method for parameter estimation of the airborne distributed MIMO radar system according to the first aspect.

[0016] The embodiment of the application provides a parameter estimation method, device and medium of an airborne distributed MIMO radar system; a system model of the airborne distributed MIMO radar is constructed, and theoretical analysis basis of parameter estimation of the airborne distributed MIMO radar system is given; rough estimation of distance and Doppler frequency shift of the target is obtained by performing matching filtering and Doppler processing on echo signals, and is used as prior information for accurate estimation of time delay and Doppler frequency shift; a one-dimensional search is performed on the time delay and the Doppler frequency shift based on the prior estimation of the target distance and the Doppler frequency, and the parameter value of the maximum likelihood function is the optimal estimation value. Compared with two-dimensional search, the two one-dimensional searches based on the prior information reduce the operation amount and complexity of the search. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 The system model of the airborne distributed MIMO radar provided by the embodiment of the application is shown in the figure;

[0018] Figure 2 The parameter estimation method of the airborne distributed MIMO radar system provided by the embodiment of the application is shown in the figure;

[0019] Figure 3 The coordinate system of the airborne distributed MIMO radar system provided by the embodiment of the application is shown in the figure;

[0020] Figure 4 The rough distance estimation of the airborne distributed MIMO radar system provided by the embodiment of the application is shown in the figure;

[0021] Figure 5 The rough distance-Doppler frequency shift estimation of the airborne distributed MIMO radar system provided by the embodiment of the application is shown in the figure;

[0022] Figure 6 The root mean square error of the one-time delay optimal estimation of the target of the airborne distributed MIMO radar system provided by the embodiment of the application is shown in the figure;

[0023] Figure 7 The root mean square error of the two-time delay optimal estimation of the target of the airborne distributed MIMO radar system provided by the embodiment of the application is shown in the figure;

[0024] Figure 8 The root mean square error of the three-time delay optimal estimation of the target of the airborne distributed MIMO radar system provided by the embodiment of the application is shown in the figure;

[0025] Figure 9 The root mean square error of the one-Doppler frequency shift optimal estimation of the target of the airborne distributed MIMO radar system provided by the embodiment of the application is shown in the figure;

[0026] Figure 10A root mean square error diagram of optimal estimation of target two Doppler frequency shift of an airborne distributed MIMO radar system provided by an embodiment of the present application is shown in the figure;

[0027] Figure 11 A root mean square error diagram of optimal estimation of target three Doppler frequency shift of an airborne distributed MIMO radar system provided by an embodiment of the present application is shown in the figure;

[0028] Figure 12 A device diagram of parameter estimation of an airborne distributed MIMO radar system provided by an embodiment of the present application is shown in the figure;

[0029] Figure 13 A hardware structure diagram of a computing device provided by an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0030] The terms "first", "second" in the embodiments of the present application are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features.

[0031] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application.

[0032] According to the arrangement mode of the radar antenna in space, the MIMO radar can be divided into centralized MIMO radar and distributed MIMO radar. The antennas of the centralized MIMO radar are close to each other. The transmitting antennas of the centralized MIMO radar transmit multiple orthogonal signals, and the receiving end separates each transmitting signal by using the matched filtering technology through long-time coherent accumulation of the echo. The centralized MIMO radar increases the system degree of freedom through waveform diversity, and obtains better target parameter estimation precision, energy utilization rate, clutter suppression ability, etc. The antennas of the distributed MIMO radar are far away from each other, and the signals received by each receiving antenna can be considered as multiple independent target scattering echo signals. The distributed MIMO radar obtains high-quality target detection precision and parameter estimation performance through spatial diversity gain. The centralized MIMO radar cannot overcome the problem of target detection failure caused by the fluctuation of the target radar cross section (RCS), while the diversity gain of the distributed MIMO radar can solve the problem of target detection failure, and make up for the defects of the centralized MIMO radar.

[0033] Therefore, a new airborne distributed MIMO radar system which integrates the advantages of centralized MIMO radar and distributed MIMO radar has obvious advantages in survivability, anti-interference, target detection, parameter estimation and the like. The classical estimation method for airborne parameter estimation can directly use matching filtering and Doppler processing technology to achieve, and there are also maximum likelihood method, adaptive gradient descent method based on maximum likelihood method and the like. Although the matching filtering and Doppler processing technology can obtain time delay and Doppler shift information, the information is rough and the accuracy is relatively low. The maximum likelihood method is used to jointly estimate the time delay and Doppler shift, and although the optimal estimation of the parameters can be obtained, the method needs to perform two-dimensional parameter search and the operation amount is large.

[0034] The embodiment of the present application proposes a parameter estimation method of the airborne distributed MIMO radar system for the system, and the method comprises the following steps:

[0035] S201: a system model of the airborne distributed MIMO radar is constructed, the system comprises at least two transmitting airborne platforms, at least two receiving airborne platforms and at least one target; wherein each transmitting airborne platform and each receiving airborne platform carries a centralized MIMO antenna;

[0036] S202: based on the system model, the echo signal received by the airborne platform is subjected to matching filtering and Doppler processing to obtain rough estimation of the distance and Doppler shift of the target;

[0037] S203: according to the rough estimation of the distance and Doppler shift, the maximum likelihood function of the echo signal is solved through one-dimensional search of the time delay and Doppler shift respectively to obtain optimal estimation of the time delay and Doppler shift respectively.

[0038] It should be noted that the airborne distributed MIMO radar system deploys multiple airborne platforms which are far apart. The airborne platform can be a flying vehicle carrying an antenna array, and the antenna array is a centralized MIMO antenna. The antenna on the airborne platform can be a transmitting antenna, a receiving antenna or a transceiving antenna. The position of the airborne platform can be adjusted according to the target state, and the airborne platforms meet the layout requirements of the distributed MIMO. As shown in the example, Figure 1 The airborne distributed MIMO radar system comprises transmitting airborne platforms TX1 and TX2 for transmitting signals, receiving airborne platforms RX1 and RX2 for receiving signals and one target. The transmitting airborne platforms TX1 and TX2 run at speeds V T1 and V T2 respectively, and the receiving airborne platforms RX1 and RX2 run at speeds V R1 and V R2The target moves at a speed V. The airborne platforms keep a relatively stable speed and position around the target. The antennas on each transmitting and receiving airborne platform are uniform linear arrays with equal intervals. The transmitting waveforms of the antennas on the transmitting airborne platforms TX1 and TX2 are orthogonal to each other, i.e., S (1,1) (t), …, S (1,P) (t), S (2,1) (t), …, S (2,P) (t) are orthogonal to each other, so the transmitting antennas are also centralized MIMO antennas. The relative motion between one or more targets to be detected by the radar system and the airborne platforms causes Doppler frequency shift of the echo signals. The transmitting antennas of the transmitting airborne platforms transmit signals with waveforms orthogonal to each other, and after being reflected by the target, the receiving antennas on the receiving airborne platforms RX1 and RX2 receive echo signals reflected by the target at multiple angles. After processing the echo signals, the time delay and Doppler frequency shift parameter estimates of the target are obtained to support the identification of the position and speed of the target in the next step.

[0039] The processing of the echo signals includes matching filtering of the echo signals received by the receiving antennas of each receiving airborne platform to extract the signals of each transmitting antenna, so that the rough distance information of the target can be obtained. The matching filtering of the echo signals can be implemented through time domain convolution operation, or through frequency domain multiplication and then conversion into time domain signals. The conversion between the time domain signals and the frequency domain signals can be completed through fast Fourier transform (FFT) and inverse fast Fourier transform (IFFT). After the matching filtering, the signals are subjected to fast Fourier transform along the slow time with respect to the Doppler frequency to obtain rough Doppler frequency shift information.

[0040] The maximum likelihood estimation method belongs to a kind of Bayesian estimation method, and the maximum likelihood estimation method is asymptotically optimal estimation. Generally, maximum likelihood estimation is first based on the parameter to be estimated to construct a likelihood function, and the parameter value that makes the likelihood function reach extreme value is the optimal estimation of the parameter to be estimated. When maximum likelihood function is used for time delay-doppler frequency shift parameter estimation, the method for solving maximum likelihood function can use time delay-doppler two-dimensional grid search, and the disadvantage is that the operation amount is relatively large. In order to reduce the operation amount of two-dimensional parameter search, the embodiment of the present application utilizes the rough prior information of target distribution in distance-doppler plane, and decomposes one two-dimensional search problem into two one-dimensional search problems, so as to reduce the operation amount of search. According to the rough information of target distribution in distance-doppler frequency shift plane obtained by matching filtering and doppler processing, the rough value of one of distance or doppler frequency shift is substituted into the likelihood function, wherein the distance rough information can be converted into time delay parameter and substituted into the likelihood function. Then, dynamic search is carried out in the set range of the other parameter, that is, one-dimensional parameter search is completed. When the maximum value of the likelihood function is obtained, the value of time delay or doppler frequency shift corresponding to the maximum value is the corresponding parameter estimation value. The same method is used when searching the other parameter. Therefore, the parameter estimation of time delay and doppler frequency shift is converted into two one-dimensional searches of the likelihood function, so as to reduce the operation amount of parameter search and the complexity of parameter estimation.

[0041] For Figure 2 The technical solutions shown in the description, in some implementations, a system model of the airborne distributed MIMO radar is constructed, comprising:

[0042] The transmission signal model of one transmitter airborne platform of the airborne distributed MIMO radar system is constructed as Wherein, s (k,p) (k, p, t) represents the transmission signal of the pth transmission antenna on the kth transmitter airborne platform, and P represents the number of transmission antennas on one transmitter airborne platform.

[0043] The discrete reception signal model of one antenna of one receiver airborne platform of the airborne distributed MIMO radar system is constructed as shown in formula (1):

[0044]

[0045] Wherein, y (l,q),m (n) represents the echo signal of the nth fast time point of the mth pulse signal transmitted by the transmitter airborne platform received by the qth antenna on the lth receiver airborne platform, K represents the number of transmitter airborne platforms, P represents the number of transmission antennas on one transmitter airborne platform, E represents the total energy of the transmission signal, Z represents the number of targets, and ξ (k,l)denotes the target complex reflection coefficient on the propagation path of the target at the kth transmitter airborne platform and the lth receiver airborne platform, τ (k,p)(l,q) denotes the delay of the signal transmitted by the pth antenna of the kth transmitter airborne platform, reflected by the target, and received by the qth antenna of the lth receiver airborne platform, d T denotes the interval between the antennas, θ T,k denotes the visibility angle of the target with respect to the kth transmitter airborne platform, θ R,l denotes the visibility angle of the target with respect to the lth receiver airborne platform, λ denotes the wavelength of the transmitted signal, f c denotes the carrier frequency center frequency, PRI denotes the pulse signal repetition interval, f (k,p)(l,q) denotes the Doppler shift of the signal transmitted by the pth antenna of the kth transmitter airborne platform, reflected by the target, and received by the qth antenna of the lth receiver airborne platform, w denotes the phase error of the signal transmitted by the pth antenna of the kth transmitter airborne platform, reflected by the target, and received by the qth antenna of the lth receiver airborne platform, w k,m,(l,q) (n) is a complex Gaussian white noise with zero mean;

[0046] The received signal matrix of the lth receiver airborne platform is obtained as shown in equation (2):

[0047] y l,m (n) = [y (l,1),m (n), y (l,2),m (n), y (l,Q),m (n)] T (2),

[0048] where Q denotes the number of antennas on one receiver airborne platform;

[0049] The total received signal matrix of all receiver airborne platforms is obtained as shown in equation (3):

[0050] y = [y 1,1 [1], y 1,1 [2], … y 1,1 [N], …, y 1,M [N], …, y L,M [N]] T (3),

[0051] where M denotes the number of received pulse signals, L denotes the number of receiver airborne platforms, and N denotes the number of snapshots per pulse signal.

[0052] Specifically, in constructing the system model, we assume there are Z targets, K transmitting airborne platforms, and L receiving airborne platforms. The k-th (k = 1, 2, ..., K) transmitting station has P centralized antennas, and the l-th (l = 1, 2, ..., L) receiving station has Q centralized antennas. The distance between the transmitting antennas is d. P The distance between the receiving antennas is d Q In this embodiment of the invention, d is set. P =d Q =d T The value is equal to half the wavelength of the transmitted signal, where the wavelength is λ. The positions of the transceiver platform, antenna, and target are defined using a Cartesian plane coordinate system. For example... Figure 3 As shown, the origin of the coordinate system is the first antenna node of the airborne platform, the X-axis is the straight line formed by the antenna nodes of the airborne platform, and the axis perpendicular to this line is designated as the Y-axis. θ T,k θ represents the visible viewing angle of the target relative to the k-th airborne transmitter platform. Since each airborne transmitter uses a centralized MIMO antenna, the visible viewing angle of the target relative to the transmitter antennas within the same airborne platform is approximately equal. R,l This indicates that the target's visible field of view relative to the l-th receiving station is approximately equal to the target's visible field of view relative to antennas within the same receiving airframe.

[0053] It should be noted that, in order to separate the transmitted waveforms of each platform, the transmitted waveforms of each platform must be orthogonal to each other. Secondly, in order to extract the transmission degrees of freedom of each transmitter at the receiving end, the transmitted waveforms within each transmitter platform are generally required to be orthogonal to each other; therefore, the transmitting antennas are also centralized MIMO antennas. The signal received by each antenna on the receiving platform is the superposition of the signals transmitted by each transmitting antenna on each transmitter platform after reflection from the target. The transmitted signals of each transmitting antenna are orthogonal to each other, and the orthogonality of the signals is still satisfied even when the time delays of the signals arriving at the receiving antenna are different.

[0054] It should also be noted that orthogonal linear frequency modulated (LFM) signals can be used as the transmitted signal to achieve better waveform isolation. Besides LFM signals, nonlinear frequency modulated (NLFM) signals can also be used. Synthesizing NLFM signal waveforms is a complex process, and approximation methods are often used in practice. A classic approach is to design the waveform based on the autocorrelation function of the design signal and the principle of lingering phase.

[0055] For matched filtering and Doppler processing of the echo signal received by the receiving airborne platform, in detail, after each receiving antenna receives the echo signal, it first performs matched filtering to extract the signals from different transmitting antennas. By matching filtering the echo and the transmitted signal, each antenna on each receiving airborne platform uses KP filters, resulting in a total of KP outputs. Matched filtering can be described in the time domain as a convolution process, as shown in equation (4):

[0056]

[0057] The relationship between the impulse response function h(t) and the reference signal can be expressed as h(t) = s * (-t).

[0058] Matched filtering can be implemented not only in the time domain but also in the frequency domain. Frequency domain matched filtering can be described as shown in equations (5) and (6):

[0059] Y out (f)=S(f)H(f) (5)

[0060] y out (t) = IFFT(Y) out (6).

[0061] For Doppler processing of echo signals, specifically, the radar transmits a periodic pulse sequence. Every M pulses in the fast-time sample vector set N can be viewed as a two-dimensional matrix y[n,m]. The dimension containing the pulse count is the slow-time axis, and the slow-time sampling frequency is essentially the pulse repetition frequency. After performing matched filtering on the range dimension of the echo, a Fast Fourier Transform (FFT) is performed along the slow time dimension on the Doppler dimension (i.e., the pulse dimension), which is equivalent to performing an FFT in the Doppler dimension. This allows the Doppler frequency shift information of the target to be extracted.

[0062] for Figure 2 In some implementations of the technical solution shown, based on the coarse estimates of the distance and Doppler frequency shift, the maximum likelihood functions of the echo signal are solved by one-dimensional search of the time delay and Doppler frequency shift, respectively, to obtain the optimal estimates of the time delay and Doppler frequency shift, including:

[0063] Define the time delay-Doppler frequency shift joint parameters Let η be the product of the target's complex reflection coefficient and phase error, which is the first parameter, and let η be the second parameter. The expression for the time delay and Doppler shift matrix of the z-th target is:

[0064]

[0065]

[0066] A first frequency domain log-likelihood function of the echo signal is constructed, as shown in equation (7):

[0067]

[0068] where Y = [Y (1,1)1 [1], Y (1,1)1 [2], …, Y (1,1)1 [N], … Y (l,q),m (u), …, Y (L,Q)1 [N], …, Y (L,Q)M [N]] T represents a matrix form of echoes of all paths in the frequency domain,

[0069]

[0070] wherein represents a noise in the form of the frequency domain, obeying a Gaussian distribution, M represents the number of pulses;

[0071] solving an estimated value of the second parameter according to the first frequency domain log-likelihood function;

[0072] substituting the estimated value of the second parameter into the first frequency domain log-likelihood function to obtain a second frequency domain log-likelihood function containing only the first parameter;

[0073] based on the rough estimation of the distance, performing one-dimensional search in a Doppler shift search range, so that the parameter value making the second frequency domain log-likelihood function maximum is the optimal estimation of the Doppler shift;

[0074] based on the rough estimation of the Doppler shift, performing one-dimensional search in a time delay search range, so that the parameter value making the second frequency domain log-likelihood function maximum is the optimal estimation of the time delay.

[0075] It should be noted that when the likelihood function is used to estimate the parameter, the time domain likelihood function can also be used, and the frequency domain likelihood function can also be used. The embodiment of the present application estimates the second parameter η first, substitutes the estimated value of the second parameter η into the first frequency domain log-likelihood function, and the likelihood estimation of the time delay-Doppler shift is converted into a generalized likelihood estimation. When the time delay and the Doppler shift are estimated, the second parameter η does not need to be searched, so that the grid search dimension of the subsequent estimation problem is reduced.

[0076] For the above implementation manner, in some examples, the solving of the estimated value of the second parameter according to the frequency domain log-likelihood function comprises:

[0077] simplifying the first frequency domain log-likelihood function to obtain an estimated expression of the second parameter as shown in equation (8):

[0078]

[0079] wherein, denotes taking the real part;

[0080] By taking the derivative of the simplified frequency domain log-likelihood function, the parameter value corresponding to the derivative being 0 is the estimation value of the second parameter, as shown in equation (9):

[0081]

[0082] It should be noted that after simplifying the first frequency domain log-likelihood function (7), the following equation can be obtained:

[0083]

[0084] Discarding the irrelevant terms in equation (10), the estimation expression of the second parameter shown in equation (9) can be obtained.

[0085] For the above implementation manner, in some examples, the estimation value of the second parameter is substituted into the first frequency domain log-likelihood function to obtain a second frequency domain log-likelihood function containing only the first parameter, including:

[0086] The estimation value of the second parameter is substituted into the log-likelihood function, as shown in equation (11):

[0087]

[0088] Discarding the irrelevant terms in the above equation, the second frequency domain log-likelihood function containing only the first parameter is shown in equation (12):

[0089]

[0090] The rough estimation of the Doppler shift is substituted into equation (12), and the time delay search range is set to The interval is The value that makes the log-likelihood function maximum in the time delay search range is the optimal estimation value of the time delay.

[0091] The rough estimation of the time delay is substituted into equation (12), and the Doppler shift search range is set to The interval is The value that makes the log-likelihood function maximum in the Doppler shift search range is the optimal estimation value of the Doppler shift; wherein PRF represents the pulse repetition frequency, and

[0092] It should be noted that for the optimal estimation of time delay-Doppler shift, a two-dimensional search can be used. The search range of time delay and Doppler shift parameters is set, and the likelihood function is substituted into the search range, and when the likelihood function is maximum, the corresponding time delay and Doppler shift are the accurate estimated values. At this time, the estimated value is the most accurate, but the operation amount is also extremely large. Based on the prior estimation of target distance and Doppler shift, one-dimensional search is performed on time delay and Doppler shift, so that the parameter value of the maximum likelihood function is the optimal estimation value. Compared with two-dimensional search, two one-dimensional searches based on prior information reduce the operation amount and complexity of the search.

[0093] For Figure 2 The technical solutions shown in the drawings, in some embodiments, further comprise:

[0094] Converting the maximum likelihood function into a Vandermonde matrix form;

[0095] The one-dimensional search is realized by performing fast Fourier transform on the maximum likelihood function in the Vandermonde matrix form.

[0096] For the above example, after obtaining the second frequency domain log-likelihood function, the part inside the absolute value symbol in formula (12) is converted into a Vandermonde matrix form, as shown in formula (13):

[0097]

[0098] The one-dimensional search is performed in the Doppler shift search range, including performing one-dimensional search by fast Fourier transform to obtain the optimal estimation of Doppler shift;

[0099] The one-dimensional search is performed in the time delay search range, including performing one-dimensional search by fast Fourier transform to obtain the optimal estimation of Doppler shift.

[0100] It should be noted that the search for optimal estimation by fast Fourier transform operation can further reduce the operation complexity. When two one-dimensional searches are used, for Doppler shift, the operation complexity is MxA τ For time delay, the operation complexity is NxA f Wherein A τ and A f respectively represent the number of search grid points of time delay and Doppler shift. Therefore, the final complexity of time delay-Doppler shift estimation is: NxA τ +MxA fWhen the fast Fourier transform operation is adopted, in the time delay dimension, the FFT calculation complexity of N points is NlogN, in the Doppler dimension, the FFT calculation complexity of M points is MlogM, therefore the operation complexity of the FFT of the Vandermonde matrix is MNlog2N+NMlog2M=MNlog2(MN). Obviously, the operation complexity of the two one-dimensional parameter searches is greater than the operation complexity of the two FFTs.

[0101] Based on the foregoing technical solution, the embodiment of the application performs simulation experiments on parameter estimation of an airborne distributed MIMO radar system, and simulation parameters are as follows:

[0102] The speed of light c=3×10 8 m / s, the signal carrier frequency f c =3GHz, the signal bandwidth B=10MHz, the pulse width T p =10uus, the number of transmitter airborne platforms K=3, the number of transmitting antennas in each transmitter airborne platform is P=3, and the sampling frequency f s =50MHz. According to the relationship between the sampling time and the sampling frequency, the sampling time interval t s =1 / f s =0.02us. The signal-to-noise ratio SNR=20dB, the number of point targets Z=3, the number of receiver airborne platforms L=3, the number of antennas of each receiver airborne platform Q=3, the nominal distance between the transmitter airborne platforms and the receiver airborne platforms is equal, and the distance is d=500m. It is assumed that the first antenna of the first airborne platform is at the origin position, and the coordinates of the second and third platforms are (500, 0) and (1000, 0) respectively. In order to simplify the model, it is assumed that the antennas of each airborne platform are both transmitting antennas and receiving antennas. The positions of the three targets relative to the first airborne platform are and The angles of the three targets relative to the origin of the rectangular coordinate system are and The antenna spacing of each airborne platform is d T =λ / 2. The three targets move at a uniform speed, the tangential speed is equal and is 10m / s, and the radial speed is also 10m / s. The airborne platforms move at a uniform speed, the speed in the horizontal coordinate axis is 10m / s, and the speed in the vertical coordinate axis is 10m / s. The SNR simulation range is set to-20dB-20dB, and the complex reflection coefficients of the targets are The phase error φ is a random number, the number of pulses M=500, the pulse repetition interval PRI=40us, and the number of Monte Carlo experiments is 1000. The root mean square error of the time delay-Doppler frequency shift estimation is used as a parameter for measuring the estimation performance, which is defined as shown in formula (14) and formula (15):

[0103]

[0104]

[0105] Where Ω represents the number of Monte Carlo experiments, This represents the estimated time delay for the i-th experiment. This represents the estimated Doppler frequency shift value of the i-th Monte Carlo experiment.

[0106] Figure 4 This diagram illustrates a rough range estimation for an airborne distributed MIMO radar system provided in an embodiment of the present invention. Figure 5 This diagram illustrates a rough estimation of the range-Doppler frequency shift for an airborne distributed MIMO radar system provided in an embodiment of the present invention. Figure 6 This diagram illustrates the root mean square error of the target-time delay optimal estimation of the airborne distributed MIMO radar system provided in an embodiment of the present invention. Figure 7 This diagram illustrates the root mean square error of the optimal estimation of target two time delay in an airborne distributed MIMO radar system provided in an embodiment of the present invention. Figure 8 This diagram illustrates the root mean square error of the optimal estimation of the three time delays of the target in the airborne distributed MIMO radar system provided in an embodiment of the present invention. Figure 9 This diagram illustrates the root mean square error of the optimal target-Doppler frequency shift estimation for an airborne distributed MIMO radar system provided in an embodiment of the present invention. Figure 10 This diagram illustrates the root mean square error of the optimal estimation of the second-Doppler frequency shift of a target in an airborne distributed MIMO radar system provided in an embodiment of the present invention. Figure 11 This diagram illustrates the root mean square error of the optimal estimation of the three-Doppler frequency shift of a target in an airborne distributed MIMO radar system provided in an embodiment of the present invention.

[0107] from Figure 4 Simulation results show that after matched filtering the target echo, three very narrow time pulses appear, allowing for the separation of the three target positions. The curves in the figure, from left to right, represent the positions of the first, second, and third targets, with distances from the first radar of approximately 999m, 1500m, and 2001m, respectively. Because the power of each transmitting antenna is equal, the amplitude of the compressed pulses is the same. In the range dimension, signals at distances unrelated to the target are compressed in amplitude. Therefore, the simulation results effectively verify the effectiveness of matched filtering in extracting target position information, thus improving range resolution.

[0108] from Figure 5 Simulation results show that performing matched filtering followed by pulse Doppler processing on the output yields range-Doppler information. From... Figure 5As can be seen, the abscissa corresponds to the distance information of the first target to the third target in turn, and the ordinate corresponds to the Doppler of the first target to the third target, which is about 200MHz, -75MHz and -75MHz in turn, and the error range of the true values 200MHz, -73.21MHz and -73.21MHz is within the ideal range. The matching filter output and the Doppler processing result provide rough prior information for the next step of time delay and Doppler shift estimation.

[0109] Figure 6 、 Figure 7 and Figure 8 The three figures respectively represent the echo delay estimation values of three different targets using the two-dimensional search and one-dimensional search methods, the signal is transmitted by the first transmitting platform and the first transmitting antenna, and the first receiving platform receives the first receiving antenna, Figure 9 、 Figure 10 and Figure 11 The three figures respectively represent the Doppler shift estimation values of three different targets, and the relationship between the echo delay estimation RMSE and the SNR is given in the figure. The abscissa represents the signal-to-noise ratio range of the simulation setting, and the ordinate represents the estimated RMSE. The simulation results show that for each target, the estimation results of the two methods are similar, and with the increase of the signal-to-noise ratio, the RMSE of the echo delay estimation corresponding to the target gradually decreases.

[0110] Based on the same inventive concept of the foregoing technical solutions, see Figure 12 which shows a device 120 for estimating parameters of an airborne distributed MIMO radar system according to an embodiment of the application, the device 120 comprises: a system model construction part 1201, a rough estimation part 1202, and an optimal estimation part 1203; wherein,

[0111] The system model construction part 1201 is configured to construct a system model of the airborne distributed MIMO radar, the system comprising at least two transmitting airborne platforms, at least two receiving airborne platforms and at least one target; wherein each of the transmitting airborne platforms and each of the receiving airborne platforms carries a centralized MIMO antenna;

[0112] The rough estimation part 1202 is configured to perform matching filtering and Doppler processing on the echo signal received by the airborne platform based on the system model, to obtain rough estimation of the distance and Doppler shift of the target;

[0113] The optimal estimation part 1203 is configured to solve the maximum likelihood function of the echo signal through one-dimensional search of time delay and Doppler shift respectively according to the rough estimation of the distance and Doppler shift, to obtain optimal estimation of the time delay and Doppler shift respectively.

[0114] It should be noted that, for the above device, the specific implementation of the functions configured by each "part" can be referred to the foregoing Figure 2 The implementation manners and examples of the corresponding steps in the parameter estimation method of the airborne distributed MIMO radar system shown in FIG. 8 are not described herein again.

[0115] It can be understood that, in the embodiment, the "part" can be a part of circuit, a part of processor, a part of program or software, etc., and of course can also be a unit, and can also be a module or non-modular.

[0116] In addition, each component in the embodiment can be integrated in a processing unit, or each unit can exist physically independently, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software function module.

[0117] When the integrated unit is realized in the form of a software function module and is not sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the embodiment can be embodied in the form of a software product, and the computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform all or part of the steps of the method described in the embodiment. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0118] Therefore, the embodiment provides a computer storage medium, which stores an airborne distributed MIMO radar system parameter estimation program. The airborne distributed MIMO radar system parameter estimation program is executed by at least one processor to implement the steps of the airborne distributed MIMO radar system parameter estimation method described in the foregoing technical solution.

[0119] According to the airborne distributed MIMO radar system parameter estimation device 120 and the computer storage medium, referring to the foregoing description of the airborne distributed MIMO radar system parameter estimation device 120 and the computer storage medium, the specific implementation of the functions configured by each "part" can be referred to the foregoing Figure 13Fig. 12 is a schematic diagram showing a specific hardware structure of a computing device 130 capable of implementing the parameter estimation device 120 of the airborne distributed MIMO radar system according to the embodiments of the present application. The computing device 130 can be a wireless device, a mobile or cellular phone (including so-called smart phone), a personal digital assistant (PDA), a video game console (including a video display, a mobile video game device, a mobile video conferencing unit), a laptop computer, a desktop computer, a television set-top box, a tablet computing device, an electronic book reader, a fixed or mobile media player, etc. The computing device 130 includes a communication interface 1301, a memory 1302 and a processor 1303, which are coupled together by a bus system 1304. It can be understood that the bus system 1304 is used to realize the connection communication between these components. The bus system 1304 includes, in addition to a data bus, a power supply bus, a control bus and a status signal bus. However, for the purpose of clear illustration, all kinds of buses are marked as the bus system 1304 in Fig. 12. Among them, Figure 13

[0120] The communication interface 1301 is used for receiving and sending signals in the process of transceiving information with other external network elements.

[0121] The memory 1302 is used for storing a computer program capable of running on the processor 1303.

[0122] The processor 1303 is used for executing the steps of the unknown parameter identification method of the satellite with large inertia rotating load in the foregoing technical solutions when running the computer program, which will not be repeated here.

[0123] ​It is appreciated that the memory 1302 in the embodiments of the present application can be a volatile memory or a nonvolatile memory, or can include both volatile and nonvolatile memory. Among them, the nonvolatile memory can be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically EPROM (EEPROM), or a flash memory. The volatile memory can be a Random Access Memory (RAM) used as an external cache. By way of example, and not limitation, many forms of RAM can be used, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct Rambus RAM (DRRAM). The memory 1302 of the system and method described herein is intended to include, without being limited to, these and any other suitable types of memory.

[0124] The processor 1303 can be an integrated circuit chip logic circuit, or an instruction in a form of software. In implementation process, each step of the above method can be completed by the integrated logic circuit or the instruction in the form of software in the processor 1303. The processor 1303 can be a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. Each method, step and logical block diagram disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as a hardware code processor to execute, or a combination of hardware and software modules in the code processor to execute. The software module can be located in the random access memory, the flash memory, the read only memory, the programmable read only memory or the electrically erasable programmable memory, the register or other mature storage medium in the art. The storage medium is located in the memory 1302, and the processor 1303 reads the information in the memory 1302 and combines the hardware to complete the steps of the above method.

[0125] It can be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit can be implemented within one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, micro-controllers, microprocessors, other electronic units designed to perform the functions described herein, or a combination thereof.

[0126] For software implementation, the techniques described herein can be implemented with modules (e.g., procedures, functions, and so on) that perform the functions described herein. The software codes can be stored in the memory and executed by the processor. The memory can be implemented within the processor or external to the processor.

[0127] Specifically, the processor 1303 is further configured to execute the parameter estimation method of the airborne distributed MIMO radar system in the foregoing technical solutions when the computer program is executed, and details are not described herein.

[0128] It can be understood that the exemplary technical solutions of the parameter estimation device 120 of the airborne distributed MIMO radar system and the computing device 130 and the technical solutions of the parameter estimation method of the airborne distributed MIMO radar system belong to the same concept, and therefore, the details of the technical solutions of the parameter estimation device 120 of the airborne distributed MIMO radar system and the computing device 130 that are not described in detail can be referred to the description of the technical solutions of the parameter estimation method of the airborne distributed MIMO radar system. The present application does not repeat them.

[0129] It should be noted that the technical solutions disclosed in the embodiments of the present application can be combined arbitrarily without conflict.

[0130] The above describes only a specific embodiment of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for parameter estimation of an airborne distributed MIMO radar system, characterized in that, The method comprises the following steps: constructing a system model of a distributed MIMO radar borne by at least two transmitter platforms, at least two receiver platforms and at least one target, wherein each of the transmitter platforms and each of the receiver platforms is provided with a centralized MIMO antenna; performing matched filtering and Doppler processing on echo signals received by the platforms based on the system model to obtain rough estimates of the distance and Doppler frequency shift of the target; solving a maximum likelihood function of the echo signals through one-dimensional search of time delay and Doppler frequency shift respectively based on the rough estimates of the distance and Doppler frequency shift to obtain optimal estimates of the time delay and Doppler frequency shift respectively; The method further comprises: converting the maximum likelihood function into a Vandermonde matrix form; performing fast Fourier transform on the maximum likelihood function in the Vandermonde matrix form to realize the one-dimensional search; The step of constructing the system model of the distributed MIMO radar borne by at least two transmitter platforms, at least two receiver platforms and at least one target comprises: The transmit signal model of one transmitter airborne platform of the airborne distributed MIMO radar is constructed as wherein, denotes the transmit signal of the k th transmit antenna on the p th transmitter airborne platform, P denotes the number of transmit antennas on one transmitter airborne platform; The first l The first q The discrete received signal model of the first antenna of the first receiver airborne platform of the airborne distributed MIMO radar is shown in equation (1): （1） in, Indicates the first l The first receiver onboard platform q The antenna received the first transmission from the airborne platform. m The first pulse signal n A fast time point echo signal K Indicates the number of airborne launch platforms. P This indicates the number of transmit antennas on a transmitter platform. E This represents the total energy of the transmitted signal. Z Indicates the target quantity. Indicates the target is in the first place. k The first airborne launch platform and the first l The target complex reflection coefficient along the propagation path of an airborne receiving platform. Indicates by the first k The first launch airborne platform p The first antenna transmits the signal, which is reflected by the target and then transmitted by the second antenna. l The first receiver airborne platform q The delay of the signal received by each antenna, Indicates the spacing between antennas, This represents the visible field of view of the target relative to the k-th airborne launching platform. Indicates the target relative to the first l The viewpoint of an airborne receiving platform Indicates the wavelength of the transmitted signal. Indicates the carrier frequency. PRI Indicates the repetition interval of the pulse signal. Indicates by the first k The first launch airborne platform p The first antenna transmits the signal, which is reflected by the target and then transmitted by the second antenna. l The first receiver airborne platform q Doppler frequency shift of the signal received by each antenna Indicates by the first k The first launch airborne platform p The first antenna transmits the signal, which is reflected by the target and then transmitted by the second antenna. l The first receiver airborne platform q The phase error of the signal received by each antenna. It is complex Gaussian white noise with zero mean; Get the l The received signal matrix of each airborne receiving platform is shown in equation (2): （2）， wherein Q represents the number of antennas on one receiver platform; obtaining a total received signal matrix of all receiver platforms as shown in formula (3): （3）， wherein, M represents the number of received pulse signals, L represents the number of receiver carrier platforms, N represents the number of snaps per pulse signal; The step of solving the maximum likelihood function of the echo signals through one-dimensional search of time delay and Doppler frequency shift respectively based on the rough estimates of the distance and Doppler frequency shift to obtain optimal estimates of the time delay and Doppler frequency shift respectively comprises: Defining delay-doppler shift joint parameters for the first parameter and the product of the target complex reflection coefficient and phase error for the second parameter, wherein the delay and Doppler shift matrix expression for the first target is z ​ ； constructing a first frequency domain log-likelihood function of the echo signals as shown in formula (4): （4） wherein represents the matrix form of the echoes of all paths in the frequency domain, wherein denotes the noise in the frequency domain form, which is subject to Gaussian distribution, , M denotes the number of pulses; solving an estimated value of the second parameter based on the first frequency domain log-likelihood function; substituting the estimated value of the second parameter into the first frequency domain log-likelihood function to obtain a second frequency domain log-likelihood function containing only the first parameter; performing one-dimensional search within a Doppler frequency shift search range based on the rough estimate of the distance to make the parameter value at which the second frequency domain log-likelihood function is maximum as the optimal estimate of the Doppler frequency shift; performing one-dimensional search within a time delay search range based on the rough estimate of the Doppler frequency shift to make the parameter value at which the second frequency domain log-likelihood function is maximum as the optimal estimate of the time delay.

2. The method of claim 1, wherein, The step of solving the estimated value of the second parameter based on the frequency domain log-likelihood function comprises: simplifying the first frequency domain log-likelihood function to obtain an estimated expression of the second parameter as shown in formula (5): （5） wherein represents a real number; solving the parameter value corresponding to the derivative being 0 by derivation of the simplified frequency domain log-likelihood function, and the parameter value is the estimated value of the second parameter as shown in formula (6): （6）。 3. The method of claim 2, wherein, The step of substituting the estimated value of the second parameter into the first frequency domain log-likelihood function to obtain a second frequency domain log-likelihood function containing only the first parameter comprises: substituting the estimated value of the second parameter into the log-likelihood function as shown in formula (7): （7）； discarding irrelevant terms in the above formula to obtain a second frequency domain log-likelihood function containing only the first parameter as shown in formula (8): （8）。 4. The method of claim 3, wherein, The step of converting the maximum likelihood function into a Vandermonde matrix form comprises: converting the part within the absolute value symbol in the second frequency domain log-likelihood function shown in formula (8) into a Vandermonde matrix form as shown in formula (9): （9）。 5. An apparatus for parameter estimation of an airborne distributed MIMO radar system, the apparatus comprising: The system model construction part, the rough estimation part, and the optimal estimation part; wherein The system model construction part is configured to construct a system model of the airborne distributed MIMO radar, the system including at least two transmitting airborne platforms, at least two receiving airborne platforms, and at least one target; wherein each of the transmitting airborne platforms and each of the receiving airborne platforms carries a centralized MIMO antenna; The rough estimation part is configured to, based on the system model, perform matched filtering and Doppler processing on echo signals received by the receiving airborne platforms to obtain rough estimations of distances and Doppler frequency shifts of the target; The optimal estimation part is configured to, according to the rough estimations of the distances and Doppler frequency shifts, respectively solve a maximum likelihood function of the echo signals through one-dimensional searches of time delays and Doppler frequency shifts to respectively obtain optimal estimations of the time delays and Doppler frequency shifts; The system model construction part is further configured to: A transmit signal model of an airborne platform for constructing an airborne distributed MIMO radar is as follows: ,in, Indicates the first k The first on the launch airborne platform p The transmitted signal from each transmitting antenna, P This indicates the number of transmitting antennas on a single airborne transmitting platform; The first l The first q The discrete received signal model of the first antenna of the first receiver airborne platform of the airborne distributed MIMO radar is shown in equation (1): （1） in, Indicates the first l The first receiver onboard platform q The antenna received the first transmission from the airborne platform. m The first pulse signal n A fast time point echo signal K Indicates the number of airborne launch platforms. P This indicates the number of transmit antennas on a transmitter platform. E This represents the total energy of the transmitted signal. Z Indicates the target quantity. Indicates the target is in the first place. k The first airborne launch platform and the first l The target complex reflection coefficient along the propagation path of an airborne receiving platform. Indicates by the first k The first launch airborne platform p The first antenna transmits the signal, which is reflected by the target and then transmitted by the second antenna. l The first receiver airborne platform q The delay of the signal received by each antenna, Indicates the spacing between antennas, This represents the visible field of view of the target relative to the k-th airborne launching platform. Indicates the target relative to the first l The viewpoint of an airborne receiving platform Indicates the wavelength of the transmitted signal. Indicates the carrier frequency. PRI Indicates the repetition interval of the pulse signal. Indicates by the first k The first launch airborne platform p The first antenna transmits the signal, which is reflected by the target and then transmitted by the second antenna. l The first receiver airborne platform q Doppler frequency shift of the signal received by each antenna Indicates by the first k The first launch airborne platform p The first antenna transmits the signal, which is reflected by the target and then transmitted by the second antenna. l The first receiver airborne platform q The phase error of the signal received by each antenna. It is complex Gaussian white noise with zero mean; Get the l The received signal matrix of each airborne receiving platform is shown in equation (2): （2）， wherein Q represents a number of antennas on one receiving airborne platform; A total received signal matrix of all receiving airborne platforms is obtained as shown in equation (3): （3）， wherein, M represents the number of received pulse signals, L represents the number of receiver carrier platforms, N represents the number of snaps per pulse signal; The optimal estimation part is further configured to: convert the maximum likelihood function into a Vandermonde matrix form; perform a fast Fourier transform on the maximum likelihood function in the Vandermonde matrix form to implement the one-dimensional search; Defining delay-doppler shift joint parameters for the first parameter and the product of the target complex reflection coefficient and phase error for the second parameter, wherein the delay and Doppler shift matrix expression for the first target is z ​ ； construct a first frequency domain log-likelihood function of the echo signals as shown in equation (4): （4） wherein represents the matrix form of the echoes of all paths in the frequency domain, wherein represents the noise in the frequency domain form, obeying a Gaussian distribution, , M represents the number of pulses; solve an estimation value of the second parameter according to the first frequency domain log-likelihood function; substitute the estimation value of the second parameter into the first frequency domain log-likelihood function to obtain a second frequency domain log-likelihood function containing only the first parameter; perform a one-dimensional search within a Doppler frequency shift search range based on the rough estimation of the distance, so that a parameter value making the second frequency domain log-likelihood function maximum is taken as an optimal estimation of the Doppler frequency shift; perform a one-dimensional search within a time delay search range based on the rough estimation of the Doppler frequency shift, so that a parameter value making the second frequency domain log-likelihood function maximum is taken as an optimal estimation of the time delay.

6. A computer storage medium, characterized in that The computer storage medium stores a program for parameter estimation of an airborne distributed MIMO radar system, and the program for parameter estimation of the airborne distributed MIMO radar system, when executed by at least one processor, implements steps of the parameter estimation method of the airborne distributed MIMO radar system according to any one of claims 1 to 4.

Images