Terahertz frequency diversity array radar error correction and target localization methods and devices

By acquiring and correcting the echo signals of a terahertz frequency diversity array radar, calculating the ideal steering vector and compensation amount, and combining the MDL criterion and the two-dimensional MUSIC spectral function, the angle-range coupling problem of the terahertz frequency diversity array radar is solved, improving positioning accuracy and robustness, and adapting to complex environments.

CN123066697APending Publication Date: 2026-11-10NAT UNIV OF DEFENSE TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202511341266.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-09-19
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

In existing technologies, terahertz frequency diversity array radars suffer from angle-range coupling problems in the far field, leading to positioning ambiguity. Furthermore, traditional methods are susceptible to array errors in actual target detection, resulting in insufficient positioning accuracy and robustness.

Method used

By acquiring radar echo signals with known range and azimuth, the ideal steering vector and fixed phase error are calculated, compensation is generated, error correction is performed, and target positioning is achieved using the MDL criterion and two-dimensional MUSIC spectral function, thus realizing the correction of array errors and parameter estimation.

Benefits of technology

It achieves distance-angle decoupling, improves parameter estimation accuracy and positioning accuracy, reduces noise interference, enhances computational robustness, adapts to complex environments, and avoids long-term, large-scale pre-training.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123066697A_ABST
    Figure CN123066697A_ABST
Patent Text Reader

Abstract

This invention relates to a method and apparatus for error correction and target localization of terahertz frequency diversity array radar. The method includes: acquiring several sets of radar echo signals and preprocessing them to obtain analytical signals; calculating an ideal steering vector based on prior target information and extracting the fixed phase error of the analytical signals; calculating a compensation amount based on the ideal steering vector and the fixed phase error; correcting the radar echo signals using the compensation amount to obtain calibrated radar echo signals; calculating the covariance matrix of the calibrated radar echo signals and performing subspace decomposition on the covariance matrix to obtain a noise subspace; estimating the number of signal sources according to the MDL criterion; deriving a two-dimensional MUSIC spectrum function based on the estimated number of signal sources through the noise subspace and calculating the two-dimensional MUSIC spectrum of the target; performing peak detection on the two-dimensional MUSIC spectrum, extracting target parameters, and obtaining the final target position information. This invention eliminates the need for extensive pre-training processes and improves positioning accuracy.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of radar target localization technology, and in particular to a method and apparatus for error correction and target localization of a terahertz frequency diversity array radar. Background Technology

[0002] Frequency Diverse Arrays (FDAs) have become a research hotspot in radar, communication, and navigation fields since their inception due to their ability to simultaneously control beam focusing in an angle-range two-dimensional space. Unlike traditional phased arrays, which are only angle-dependent, FDAs apply a tiny frequency offset much smaller than the carrier frequency between array elements, making the beam pattern a joint function of time, angle, and range, thus gaining additional range degrees of freedom. This characteristic not only gives FDAs the ability to achieve automatic beam scanning without relying on phase shifters, but also provides a new approach to suppressing interference sources that are in the same direction as the target but at different distances. It can be used in scenarios such as multi-target localization, multi-task processing, secure communication, and electronic warfare. Especially in the terahertz band, traditional phased arrays rely on phase shifters for automatic beam scanning, but terahertz frequencies are high, and phase shifters are difficult to manufacture and have significant losses, thus limiting their effective application for a long time. Utilizing the non-phase-shifting mechanism of FDAs, automatic electronically scanned radar systems can be realized in the terahertz band.

[0003] However, conventional linear FDA suffers from angle-range coupling problems in the far field, resulting in an S-shaped energy distribution and causing positioning ambiguity. Furthermore, real-world radar systems are subject to amplitude, phase, element position, and mutual coupling errors due to array configurations, rendering traditional methods ineffective when directly applied to practical target detection missions.

[0004] In parameter estimation, subspace-based 2D / 3D MUSIC, matrix pencil (MP), and maximum likelihood (ML) algorithms have been extensively studied. ML offers higher accuracy when snapshots are scarce, but requires global search, and the computational load increases rapidly with the parameter space. While MP-type algorithms avoid subspace search, they separate angle and distance information, limiting performance. Traditional MUSIC algorithms are susceptible to noise and array errors, leading to a rapid decline in estimation performance. In recent years, deep learning has provided new approaches to address these challenges. Models such as convolutional neural networks (CNNs) have shown promise in radar parameter estimation, but their local receptive fields limit the capture of global features, and they require large-scale pre-training, resulting in significant time consumption. Summary of the Invention

[0005] Therefore, it is necessary to provide a method and apparatus for error correction and target localization of terahertz frequency diversity array radar that can improve parameter estimation accuracy and does not require a large amount of pre-training process, in order to address the above-mentioned technical problems.

[0006] A method for error correction and target localization of a terahertz frequency diversity array radar, the method comprising: Several sets of radar echo signals with known range and azimuth are obtained by using a terahertz frequency diversity array. Each set of radar echo signals is preprocessed to obtain an analytical signal. Calculate the ideal steering vector based on the target prior information, and extract the fixed phase error of the analytical signal; calculate the compensation amount based on the ideal steering vector and the fixed phase error; The radar echo signal is corrected for errors by the compensation amount to obtain a calibrated radar echo signal. Calculate the covariance matrix of the calibrated radar echo signal, and perform subspace decomposition on the covariance matrix to obtain the noise subspace; The number of information sources is estimated according to the MDL criterion. Based on the estimated number of information sources, the two-dimensional MUSIC spectrum function is derived through the noise subspace, and the two-dimensional MUSIC spectrum of the target is calculated. Peak detection is performed on the two-dimensional MUSIC spectrum to obtain the final target location information.

[0007] A terahertz frequency diversity array radar error correction and target localization device, the device comprising: The analytical signal acquisition module is used to acquire several sets of radar echo signals with known range and azimuth through a terahertz frequency diversity array, and to preprocess each set of radar echo signals to obtain the analytical signal. The compensation calculation module is used to calculate the ideal steering vector based on the target prior information and extract the fixed phase error of the analytical signal; and to calculate the compensation amount based on the ideal steering vector and the fixed phase error. The calibration module is used to perform error correction on the radar echo signal using the compensation amount to obtain a calibrated radar echo signal. The subspace decomposition module is used to calculate the covariance matrix of the calibrated radar echo signal, and to perform subspace decomposition on the covariance matrix to obtain the noise subspace. The two-dimensional MUSIC spectrum calculation module is used to estimate the number of information sources according to the MDL criterion, derive the two-dimensional MUSIC spectrum function through the noise subspace based on the estimated number of information sources, and calculate the two-dimensional MUSIC spectrum of the target. The target localization module is used to perform peak detection on the two-dimensional MUSIC spectrum to obtain the final target location information.

[0008] Compared with existing technologies, the terahertz frequency diversity array radar error correction and target localization method and apparatus provided by this invention have the following advantages: 1. By directly acquiring radar echo signals with known range and azimuth through a terahertz frequency diversity array, range-angle decoupling is achieved, providing clear parameter dimensions for subsequent positioning.

[0009] 2. Fixed phase error can be extracted and compensation amount generated from several sets of radar echo signals with known range and azimuth, realizing array error correction under small data samples without a large pre-training process, thus improving efficiency; at the same time, error correction compensates for the performance degradation caused by array error, thus improving parameter estimation accuracy.

[0010] 3. After calibration, the radar echo signal can accurately reflect the target information in subsequent calculations, effectively avoiding spectral peak blurring or false peaks caused by errors, and ultimately restoring the unique identification of the target in the two-dimensional MUSIC space, thereby improving positioning accuracy.

[0011] 4. The method proposed in this invention reduces noise interference by correcting the signal, and significantly improves the accuracy of the covariance matrix and the noise subspace. Even in complex scenarios such as electromagnetic interference and multiple target superposition, it can still output stable positioning results, reduce the impact of environmental factors on positioning performance, enhance computational robustness, and adapt to complex environments. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention, and those skilled in the art can obtain other related drawings based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart illustrating the terahertz frequency diversity array radar error correction and target localization method provided in Example 1. Figure 2 This is a schematic diagram of the FDA radar array structure provided in Example 1; Figure 3 This is a schematic diagram of the 50 channels for compensation phase provided in Example 1; Figure 4 This is a schematic diagram of the compensated phase after mean fitting of 50 channels as provided in Example 1; Figure 5 This is a schematic diagram of the compensated phase after median fitting of 50 channels, as provided in Example 1. Figure 6 This is a schematic diagram of the uncompensated MUSIC spectrum provided in Example 1; Figure 7 This is a schematic diagram of the mean-fit correction results provided in Example 1; Figure 8This is a schematic diagram of the correction results using median fitting provided in Example 1; Figure 9 This is a structural block diagram of the terahertz frequency diversity array radar error correction and target positioning device provided in Example 2; Figure 10 This is an internal structural diagram of the computer device provided in Example 3.

[0014] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0016] It is understood that the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0017] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0018] Example 1 Before discussing this solution, we will first explain traditional radar technology.

[0019] Traditional phased array radars control beam pointing via electronic scanning, forming only a fixed, narrow beam pointing within a small angular range at any given time. Multiple-input multiple-output (MIMO) radars, utilizing transmit freedom, can simultaneously form multiple beams. However, like phased array radars, MIMO radars cannot suppress interference with targets at the same azimuth angle but different distances. If the interference and the target of interest have the same azimuth angle but different distances, the positioning performance of the MIMO radar will be significantly reduced. Furthermore, the transmit waveforms of each element in a MIMO radar must be perfectly orthogonal, which is difficult to achieve in practice.

[0020] In contrast, the FDA's beam is dependent not only on azimuth but also on range, thus compensating for the lack of range-pointing resolution in traditional phased array beams and effectively suppressing range-dependent interference. However, there is currently a gap in the development of terahertz phased array radar technology. This is because phased array beam scanning relies on phase shifters, and in the terahertz domain, the increased operating frequency presents challenges to the design and implementation of phase shifters. Furthermore, the power divider and phase shifter losses at the front end of a terahertz centralized phased array amplifier are generally above 20dB, and there is currently no effective technology to reduce them. The slow development and poor performance of terahertz band phase shifters severely limit the beam performance and detection results of terahertz radar. From a functional perspective, frequency diversity array systems can achieve electronic beam scanning without the aid of phase shifters, enabling a non-phase-shifting automatic electronically scanned radar system in the terahertz band.

[0021] Based on this, this embodiment acquires radar echo data of the same target with known range and azimuth using an FDA radar array; extracts valid data from the radar echo data; reassembles the valid data into an eight-channel matrix and performs preprocessing; then calculates the ideal steering vector, extracts phase error from the acquired radar echo data to generate compensation, and finally applies the compensation to correct the data; then calculates the covariance matrix and performs eigenvalue decomposition and subspace division to estimate the number of sources; finally, performs two-dimensional spatial scanning and peak detection to obtain the target's position information.

[0022] like Figure 1 The diagram shows the terahertz frequency diversity array radar error correction and target localization method provided in this embodiment, which includes the following steps: Step 201: Acquire several sets of radar echo signals with known range and azimuth using a terahertz frequency diversity array, preprocess each set of radar echo signals to obtain an analytical signal.

[0023] Step 202: Calculate the ideal steering vector based on the target prior information and extract the fixed phase error of the analytical signal; calculate the compensation amount based on the ideal steering vector and the fixed phase error.

[0024] Step 203: Correct the radar echo signal for errors by using compensation to obtain the calibrated radar echo signal.

[0025] Step 204: Calculate the covariance matrix of the calibrated radar echo signal, and perform subspace decomposition on the covariance matrix to obtain the noise subspace.

[0026] Step 205: Estimate the number of information sources according to the MDL criterion. Based on the estimated number of information sources, derive the two-dimensional MUSIC spectrum function through the noise subspace and calculate the two-dimensional MUSIC spectrum of the target.

[0027] Step 206: Peak detection is performed on the two-dimensional MUSIC spectrum to obtain the final target location information.

[0028] In the specific implementation of step 201, it is mainly based on the terahertz sparse FDA radar with eight transmitters and one receiver. Its array element position and frequency offset design adopt non-uniform coefficients to achieve range-angle decoupling.

[0029] Specifically, such as Figure 2 The image shown is a schematic diagram of the FDA radar array structure. The number of array elements is represented by 8 in this embodiment. The first array element is used as the reference element, and the transmission frequency is... The ideal spacing between each array element is determined by It means that the first The transmission frequency of each array element is , , Non-uniform frequency offset design enhances distance dimension discernibility and avoids distance ambiguity. This is particularly relevant in the far field of space. A reference target is set at the location, and several sets of radar echo data of the same reference target with known range and azimuth are collected through the radar system and host computer software. The number of radar echo data sets collected is determined according to the actual situation; in this embodiment, no less than 50 sets of radar echo data are collected. The sampling rate is set to 250MHz, the duration of a single signal acquisition is 100μs, the pulse period is 2000μs, and the number of PRI samples is 10.

[0030] Then, the radar echo signals of each group are preprocessed, including channel matrix construction, filtering and Hilbert transform.

[0031] Specifically, radar data is transmitted in a fixed-format frame structure, with each frame containing a frame header identifier and valid data. The start position of each frame is accurately located through frame synchronization, and the valid data segments of each group of radar echo signals are extracted.

[0032] The valid data segments are stored in a channel-interleaved manner (Ch1, Ch2, ..., ChM cycle), and need to be separated and reassembled into a channel matrix. Therefore, the valid data segments are separated and reassembled into a channel matrix according to channels, as shown below: ; in, ; In the formula, express OK The channel matrix of the columns, where, It depends on the number of channels; in this embodiment, it is 8. It is the number of data sets stored; express The Middle OK The elements of the column, where and Less than , Integers; This represents the original data vector.

[0033] A bandpass filter is designed to filter the signals of each channel in the channel matrix, obtaining a filtered signal to suppress noise and interference outside the signal bandwidth and improve the signal-to-noise ratio. In this embodiment, an FIR filter is used to ensure linear phase characteristics, expressed as: ; In the formula, Indicates the filter function used; Indicates the cutoff frequency; Represents a rectangular window; Indicates group delay; This indicates the number of snapshots, which is the total number of data points.

[0034] Based on the zero-phase filtering principle, phase distortion is eliminated through forward-backward filtering. ; In the formula, Indicates received signal The discrete representation of , where, For discrete independent variables, take values ; Indicates in Values When, function The value of ; Indicates in Values When, function The value of .

[0035] Then, the Hilbert transform is used to convert the filtered signal into an analytic signal, preserving the positive frequency components for subsequent phase processing. The expression for the analytic signal is: ; In the formula, Indicates an analytical signal; Indicates radar echo signal; Represents the imaginary unit; This represents the Hilbert transform operator.

[0036] The instantaneous phase and instantaneous amplitude of the analytic signal are respectively: ; ; In the formula, express Analyzing the signal at any given moment; express The instantaneous phase of a moment; express The instantaneous amplitude of a moment.

[0037] The system reads 50 sets of radar echo signals in a loop, processes them through the above process to obtain the analyzed signals, and then stores them as a cell array for subsequent processing.

[0038] In the specific implementation of steps 202 and 203, there are two types of key errors in the actual FDA radar system. One is the phase error caused by initial phase deviation of array elements and inconsistency in the phase response of radio frequency channels. Another type is the distance pointing error caused by the positional deviation of the array elements. The purpose of correction is to estimate and compensate for both types of errors so that the phase of the received signal is only relative to the position of the reference target. Relevant. Therefore, the target location will be referenced. Set it as a known quantity and treat it as prior information of the target.

[0039] Ignoring propagation attenuation, the reference target reflection coefficient is defined as... At that time, based on phase error With distance pointing error Construct the first The model of the radar echo signal received by each array element is expressed as follows: ; In the formula, Indicates the first Time-domain representation of the received signal of each array element; Indicates the first The transmission frequency of each array element; Represents the speed of light; Indicates the distance to the reference target.

[0040] Based on the radar echo signal model, the measured signal is acquired, and an ideal model signal is constructed. Then, an optimization function is constructed with the objective of minimizing the difference between the ideal model signal and the measured signal, expressed as: ; In the formula, Indicates the first The measured signals of each array element; Indicates the first Ideal model signal of each array element; To represent time, the time domain handles the independent variable; Indicates the first Phase estimates in the array element model; Indicates the first The distance pointing error estimate in the individual element model.

[0041] Based on the optimization function, the azimuth steering vector and range steering vector under ideal conditions are calculated according to the prior information of the target. The expressions are as follows: ; ; The ideal guidance vector is synthesized from the azimuth guidance vector and the range guidance vector, and the expression is: ; ; In the formula, Represents the set of ideal guidance vectors; Indicates the orientation guidance vector; Indicates the distance steering vector; Indicates the position index of the array element; The base frequency of the array radar transmission is indicated, which is the center frequency of the reference array element. In this embodiment, the first array element is used as the reference. Indicates the spacing between array elements; Represents the channel frequency vector; Indicates the first The ideal guiding vector of each array element; Represents the imaginary unit; It is a set The element in represents the first element. The position index value of each array element; Indicates the first Frequency increment of each array element; Indicates the reference target azimuth angle; Indicates the distance to the reference target; This indicates element-wise multiplication.

[0042] It is worth noting that the azimuth steering vector (angle-dependent) represents the phase delay caused by the path difference when the electromagnetic wave arrives at different array elements. Frequency offset correction term. The primary function is to ensure wavenumber consistency across different frequency channels. The range steering vector (range dependence) represents the phase delay incurred by the signal traveling to and from the reference target.

[0043] To measure the phase deviation of each channel in a real system, it is necessary to extract the instantaneous phase of the analytical signal and compare it with the ideal case.

[0044] First, the instantaneous phase of the analytic signal is extracted, and it is represented as: ; In the formula, Indicates the first The instantaneous phase of the signal received by each array element; Indicates the first The analytical signal of each array element.

[0045] Using the first channel in the channel matrix as a reference, the relative phase difference is calculated based on the extracted instantaneous phase to eliminate common phase terms. The expression for calculating the relative phase difference based on the extracted instantaneous phase is as follows: ; In the formula, Indicates the first The relative phase difference of each array element; This represents the instantaneous phase of the first array element.

[0046] Phase unwrapping is performed based on the relative phase difference to resolve the 2π ambiguity problem and ensure phase continuity, yielding the unfolded phase difference, expressed as: ; in, It is an integer, used to ensure that adjacent phases after unwinding directly satisfy the following relationship: ; In the formula, Represents the first in the discrete processing domain The relative phase after untangling; Indicates the first The relative phase after untangling at each point.

[0047] Then, the phase difference after unwinding is averaged to suppress random noise and extract the fixed phase error, expressed as: ; In the formula, Indicates a fixed phase error; Indicates the number of snapshots.

[0048] Based on the difference between the ideal steering vector calculated above and the fixed phase error, the compensation amount required to eliminate the system error is calculated, and the expression is: ; In the formula, Indicates the amount of compensation; Indicates a fixed phase error; Indicates the first The ideal guiding vector for each array element.

[0049] when When, it indicates that the measured phase is ahead and requires negative compensation; when When the time is 0, it indicates that the measured phase is lagging and positive compensation is required.

[0050] Furthermore, after calculating the compensation amount based on the ideal steering vector and the fixed phase error, the following is also included: Calculate the compensation amount for each group of radar echo signals; perform mean fitting on all compensation amounts; or perform median fitting on all compensation amounts to obtain the global compensation amount; perform error correction on the radar echo signals using the global compensation amount to obtain the calibrated radar echo signals.

[0051] Specifically, 50 sets of processed data are extracted sequentially from the cell array stored in the preprocessing stage, and calculated according to the above process to obtain 50 independent compensation amounts.

[0052] In this embodiment, two methods for fitting the compensation amount are given. One method is to perform mean fitting on 50 independent compensation amounts to improve the compensation accuracy. The expression is as follows: ; Another approach is to perform median fitting based on 50 independent compensation values ​​to reduce the impact of outliers and achieve robust estimation. The expression is as follows: ; In the formula, Indicates the global compensation amount; This indicates the number of radar echo signal groups; in this embodiment, the value is 50 groups. Indicates the first The first of the array elements Group compensation amount.

[0053] In this embodiment, as Figures 3 to 5 As shown, based on standard deviation Statistical error analysis of the difference between the global compensation value and the independent compensation values ​​of each group shows that, except for channel 4, the compensation values ​​obtained by both fitting schemes are relatively concentrated. This indicates that the method proposed in this invention is robust and can reliably obtain accurate phase compensation values. It is worth noting that the relatively large fluctuation range of channel 4 is due to inherent instability in the hardware of the radar system used, and is unrelated to the processing method. Nevertheless, the method proposed in this invention can still accurately estimate the target position, further demonstrating its superior anti-interference performance.

[0054] Then, based on the global compensation amount Construct the compensation matrix, represented as: ; Error correction of the radar echo signal is performed using a compensation matrix to obtain the calibrated radar echo signal, which is represented as follows: ; In the formula, Represents the compensation matrix; Indicates the first Signal compensation value for each channel, Values ; Represents the imaginary unit; Indicates the first The global phase compensation amount of the radar echo signal received by each array element, where ; This represents the input signal vector for the correction process, i.e., the received radar data; Represents the actual signal steering vector; Represents the carrier model vector; This represents an additive noise vector.

[0055] The phase inconsistency between channels was eliminated by calibrating the radar echo signal. ; Ideally: ; In the formula, This represents the original phase of the m-th channel of the signal to be processed. This formula indicates that the phase response after calibration is completely consistent with the theoretical model.

[0056] In the specific implementation process of steps 204 to 206, the covariance matrix of the calibrated radar echo signal is calculated, and the expression is: ; In the formula, Denotes the conjugate transpose, where, Indicates a variable that can be replaced; Indicates the number of snapshots; In discrete processing In the The value that can be taken at a point.

[0057] It is understandable that by calculating the covariance matrix, all the information needed for spatial spectrum estimation can be obtained, thereby capturing the statistical properties of the signal.

[0058] The subspace decomposition of the covariance matrix is ​​expressed as follows: ; In the formula, Represents the eigenvector matrix; Represents an eigenvalue diagonal matrix; This represents a diagonal matrix composed of the signal's eigenvalues; This represents a diagonal matrix composed of noise eigenvalues; Represents the signal subspace; This represents the noise subspace.

[0059] It can be seen that the eigenvalues ​​are arranged in descending order, i.e. This divides the subspace into signal subspaces. and noise subspace .

[0060] It is understandable that the purpose of subspace decomposition of the covariance matrix is ​​to separate the signal subspace and the noise subspace, and to achieve super-resolution estimation by utilizing the orthogonality of the subspaces.

[0061] Estimate the number of sources based on the MDL criterion. The MDL criterion is expressed as:

[0062] The first term in the formula is the likelihood function, which measures the distribution of eigenvalues; the second term is the model complexity penalty.

[0063] Then, the number of information sources The estimation criteria are:

[0064] By estimating the number of information sources using the MDL criterion, the number of targets can be automatically determined without needing prior knowledge of the number of target sources or the failure of positioning results due to subjective parameter settings.

[0065] Two-dimensional MUSIC spectrum calculation is a crucial step in joint distance-angle estimation. Analysis shows that the FDA exhibits a two-dimensional dependence on distance and orientation, making joint estimation difficult with traditional MUSIC methods. Applying MUSIC separately in each of the two dimensions increases computational load and processing time. In this embodiment, the two-dimensional MUSIC spectrum estimation method is derived. First, the expression for the two-dimensional MUSIC spectrum function is determined as follows: ; Let the actual signal steering vector be... When orthogonal to the noise subspace, we have

[0066] It can be seen that the complexity of the derived two-dimensional MUSIC spectrum estimation method is related to the range of the distance-angle two-dimensional scan, the number of information sources, and the number of channels.

[0067] Therefore, based on the estimated number of information sources, the two-dimensional MUSIC spectrum function can be derived through the noise subspace, and the two-dimensional MUSIC spectrum of the reference target can be calculated.

[0068] Based on the calculated MUSIC spectrum, target parameters can be extracted through peak detection, and the final localization result can be output. The expression for peak detection of the two-dimensional MUSIC spectrum is: ; Theoretical reference target azimuth resolution for: ; Theoretical reference target range resolution for: ; In the formula, This indicates the estimated location of the reference target. Indicates the array aperture; Indicates bandwidth.

[0069] In one embodiment, the feasibility of the method described in the invention was demonstrated through experiments, which can accurately estimate the distance-angle of an unknown target and achieve search and positioning. Figure 6 The original two-dimensional music spectrum without compensation is displayed, and the effective location information of the target cannot be identified. Figure 7 and Figure 8 The corrected two-dimensional MUSIC spectrum of the target is shown. The compensation phase matrix is ​​fitted with the mean and the median respectively. It can be seen that both schemes can restore the uniqueness of the target's spatial location well, which proves the effectiveness of the present invention.

[0070] As can be seen, this invention achieves the target localization task of terahertz automatic electronically scanned radar, restores the uniqueness of FDA radar spatial targets, and has high accuracy and computational efficiency. Furthermore, this invention can perform error correction based on a small amount of data, avoiding long-term, large-scale pre-training and improving offline calibration efficiency. By jointly estimating the target distance-angle two-dimensional parameters, it overcomes the problem that traditional subspace algorithms are susceptible to noise and array errors, which can cause a rapid decline in estimation performance.

[0071] It is worth noting that the specific embodiments proposed in this invention are mainly for correcting target echo data obtained by sparse terahertz FDA radar systems. They are still applicable to other schemes that change the position design and frequency offset design of FDA radar array elements, and are also applicable to other array form radar modifications.

[0072] It should be understood that, although this embodiment Figure 1 The steps are shown sequentially as indicated by the arrows, but they are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order in which these steps are performed; they can be executed in other orders. Figure 1At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0073] Example 2 Based on the terahertz frequency diversity array radar error correction and target localization method in Embodiment 1, this embodiment discloses a terahertz frequency diversity array radar error correction and target localization device, such as... Figure 9 As shown, the terahertz frequency diversity array radar error correction and target localization device includes: an analytical signal acquisition module 401, a compensation calculation module 402, a calibration module 403, a subspace decomposition module 404, a two-dimensional MUSIC spectrum calculation module 405, and a target localization module 406, wherein: The analytical signal acquisition module 401 is used to acquire several sets of radar echo signals with known range and azimuth through a terahertz frequency diversity array, and to preprocess each set of radar echo signals to obtain analytical signals.

[0074] The compensation calculation module 402 is used to calculate the ideal steering vector based on the target prior information and extract the fixed phase error of the analytical signal; and to calculate the compensation amount based on the ideal steering vector and the fixed phase error.

[0075] The calibration module 403 is used to correct the error of the radar echo signal by means of compensation to obtain the calibrated radar echo signal.

[0076] The subspace decomposition module 404 is used to calculate the covariance matrix of the calibrated radar echo signal, and to perform subspace decomposition on the covariance matrix to obtain the noise subspace.

[0077] The two-dimensional MUSIC spectrum calculation module 405 is used to estimate the number of information sources according to the MDL criterion, derive the two-dimensional MUSIC spectrum function through the noise subspace based on the estimated number of information sources, and calculate the two-dimensional MUSIC spectrum of the target.

[0078] The target localization module 406 is used to perform peak detection on the two-dimensional MUSIC spectrum to obtain the final target location information.

[0079] In this embodiment, the specific working process and working principle of the analytical signal acquisition module 401, compensation calculation module 402, calibration module 403, subspace decomposition module 404, two-dimensional MUSIC spectrum calculation module 405, and target positioning module 406 are the same as those in Embodiment 1, and therefore will not be described again in this embodiment. Each unit module can be implemented entirely or partially through software, hardware, or a combination thereof. Each unit module can be embedded in or independent of the processor in the computer device in hardware form, or it can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above unit modules.

[0080] Example 3 like Figure 10 The diagram illustrates a terminal device disclosed in this embodiment, comprising a transmitter, a receiver, a memory, and a processor. The transmitter transmits instructions and data, the receiver receives instructions and data, the memory stores computer-executed instructions, and the processor executes the computer-executed instructions stored in the memory to implement the method described in Embodiment 1 above.

[0081] It is important to note that the aforementioned memory can be either standalone or integrated with the processor. When the memory is set up independently, the terminal device also includes a bus for connecting the memory and the processor.

[0082] Example 4 This embodiment discloses a computer-readable storage medium storing computer-executable instructions. When a processor executes the computer-executable instructions, it implements the method in Embodiment 1 above.

[0083] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0084] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0085] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method for error correction and target localization of a terahertz frequency diversity array radar, characterized in that, The method includes: Several sets of radar echo signals with known range and azimuth are obtained by using a terahertz frequency diversity array. Each set of radar echo signals is preprocessed to obtain an analytical signal. Calculate the ideal steering vector based on the target prior information, and extract the fixed phase error of the analytical signal; calculate the compensation amount based on the ideal steering vector and the fixed phase error; The radar echo signal is corrected for errors by the compensation amount to obtain a calibrated radar echo signal. Calculate the covariance matrix of the calibrated radar echo signal, and perform subspace decomposition on the covariance matrix to obtain the noise subspace; The number of information sources is estimated according to the MDL criterion. Based on the estimated number of information sources, the two-dimensional MUSIC spectrum function is derived through the noise subspace, and the two-dimensional MUSIC spectrum of the target is calculated. Peak detection is performed on the two-dimensional MUSIC spectrum to obtain the final target location information.

2. The terahertz frequency diversity array radar error correction and target localization method according to claim 1, characterized in that, Preprocessing of each group of radar echo signals yields analytical signals, including: Extract the effective data segments of each group of radar echo signals, and then separate and reassemble the effective data segments into a channel matrix according to the channel. The signals of each channel in the channel matrix are filtered to obtain a filtered signal; the filtered signal is then subjected to a Hilbert transform to obtain an analytical signal.

3. The terahertz frequency diversity array radar error correction and target localization method according to claim 2, characterized in that, The expression for the analyzed signal is: ; In the formula, Indicates an analytical signal; Indicates radar echo signal; Represents the imaginary unit; This represents the Hilbert transform operator.

4. The terahertz frequency diversity array radar error correction and target localization method according to claim 1, characterized in that, Calculate the ideal guidance vector based on prior target information, including: Pre-set target prior information The quantity is known; Ignoring propagation attenuation, we define the target reflection coefficient, and then construct the first... Model of radar echo signal received by each array element; Based on the radar echo signal model, the measured signal is obtained, and an ideal model signal is constructed. An optimization function is constructed with the objective of minimizing the difference between the ideal model signal and the measured signal. Based on the optimization function, the azimuth steering vector and range steering vector under ideal conditions are calculated according to the target prior information; The ideal guidance vector is synthesized by combining the azimuth guidance vector and the distance guidance vector.

5. The terahertz frequency diversity array radar error correction and target localization method according to claim 4, characterized in that, The ideal guidance vector is synthesized from the azimuth guidance vector and the distance guidance vector, and its expression is: ; ; In the formula, Represents the set of ideal guidance vectors; Indicates the orientation guidance vector; Indicates the distance steering vector; Indicates the first The ideal guiding vector of each array element; Represents the imaginary unit; Indicates the base frequency transmitted by the array radar; Indicates the spacing between array elements; Represents the speed of light; Indicates the first The position index value of each array element; Indicates the first Frequency increment of each array element; Indicates the reference target azimuth angle; Indicates the distance to the reference target; This indicates element-wise multiplication.

6. The terahertz frequency diversity array radar error correction and target localization method according to claim 3, characterized in that, Extracting the fixed phase error of the analyzed signal includes: Instantaneous phase extraction is performed on the analyzed signal; Using the first channel in the channel matrix as a reference, the relative phase difference is calculated based on the extracted instantaneous phase. Phase unwrapping is performed based on the relative phase difference to obtain the unfolded phase difference; then, the unfolded phase difference is averaged in the time domain to obtain a fixed phase error.

7. The terahertz frequency diversity array radar error correction and target localization method according to any one of claims 1 to 6, characterized in that, The compensation amount is calculated based on the ideal steering vector and the fixed phase error, and the expression is as follows: ; In the formula, Indicates the amount of compensation; Indicates a fixed phase error; Indicates the first The ideal guiding vector for each array element.

8. The terahertz frequency diversity array radar error correction and target localization method according to claim 7, characterized in that, After calculating the compensation amount based on the ideal steering vector and the fixed phase error, the method further includes: Calculate the compensation amount for each group of radar echo signals; The global compensation amount is obtained by fitting the mean of all compensation amounts or by fitting the median of all compensation amounts. The radar echo signal is corrected for errors by using the global compensation amount to obtain a calibrated radar echo signal.

9. The terahertz frequency diversity array radar error correction and target localization method according to claim 8, characterized in that, The radar echo signal is corrected for errors using the global compensation amount to obtain a calibrated radar echo signal, including: Based on the global compensation amount, a compensation matrix is ​​constructed, which is expressed as follows: ; Error correction of the radar echo signal is performed using the compensation matrix to obtain the calibrated radar echo signal, which is expressed as: ; In the formula, Represents the compensation matrix; Represents the imaginary unit; Indicates the first The global phase compensation amount of the radar echo signal received by each array element, where... ; This represents the input signal vector for the correction process, i.e., the received radar data; Represents the actual signal steering vector; Represents the carrier model vector; This represents an additive noise vector.

10. A terahertz frequency diversity array radar error correction and target positioning device, characterized in that, The device includes: The analytical signal acquisition module is used to acquire several sets of radar echo signals with known range and azimuth through a terahertz frequency diversity array, and to preprocess each set of radar echo signals to obtain the analytical signal. The compensation calculation module is used to calculate the ideal steering vector based on the target prior information and extract the fixed phase error of the analytical signal; and to calculate the compensation amount based on the ideal steering vector and the fixed phase error. The calibration module is used to perform error correction on the radar echo signal using the compensation amount to obtain a calibrated radar echo signal. The subspace decomposition module is used to calculate the covariance matrix of the calibrated radar echo signal, and to perform subspace decomposition on the covariance matrix to obtain the noise subspace. The two-dimensional MUSIC spectrum calculation module is used to estimate the number of information sources according to the MDL criterion, derive the two-dimensional MUSIC spectrum function through the noise subspace based on the estimated number of information sources, and calculate the two-dimensional MUSIC spectrum of the target. The target localization module is used to perform peak detection on the two-dimensional MUSIC spectrum to obtain the final target location information.