A radar signal processing method and device based on joint ESPRIT

Abstract

The application provides a radar signal processing method and device based on joint ESPRIT, acquires non-uniform slow time sampling data of a radar, and constructs a corresponding signal model; calculates a time difference according to the sampling data, and constructs a corresponding equivalent shift relationship in combination with the signal model; extracts a subspace for the sampling data to obtain a corresponding signal subspace; stacks the equivalent shift relationship and the signal subspace to obtain a joint shift relationship; and calculates a target Doppler frequency according to the joint shift relationship. Without changing the existing radar hardware structure, the method realizes coherent fusion of cross-frame data by unified modeling and subspace processing on segmented sampling data, thereby significantly improving the Doppler estimation accuracy and resolution.

Description

Technical Field

[0001] This invention relates to the field of radar signal processing, and in particular to a radar signal processing method and apparatus based on the joint ESPRIT. Background Technology

[0002] With the widespread application of vehicle-mounted millimeter-wave radar in intelligent driving, target velocity estimation typically relies on slow-time Doppler processing. Current mainstream methods often employ equal-interval sampling over a slow time dimension, constructing a coherent processing interval (CPI) of a certain length, and then performing a Fast Fourier Transform (FFT) on the slow-time data to obtain the Doppler spectrum. While these methods are simple to implement and have mature engineering practices, their Doppler resolution is inversely proportional to the slow-time observation duration. Therefore, it is usually necessary to extend the continuous uniform sampling time as much as possible to improve velocity resolution and estimation accuracy.

[0003] However, in practical vehicle-mounted radar systems, it is difficult to achieve long-term, continuous, uniform slow-time sampling due to system architecture and hardware limitations. One major reason is that vehicle-mounted radar typically operates on a "frame" basis. After transmitting and receiving a set of chirs within a certain time period, a certain interval needs to be maintained between frames. This interval is mainly used for on-chip data processing (such as FFT, CFAR, etc.), data transfer and buffer management, multi-transmit channel (TDM-MIMO) scheduling and switching, and meeting system real-time and power consumption control requirements. Therefore, slow-time sampling often manifests as approximately uniform sampling within each frame, while there are non-negligible time intervals between different frames, resulting in a segmented and non-uniform distribution of overall sampling over time. This inter-frame interval directly undermines the uniform sampling assumption upon which traditional FFT methods rely.

[0004] Under the aforementioned hardware limitations, existing technologies typically process data only within a single frame, treating consecutive chirps within a frame as a single CPI for Doppler estimation, making it difficult to coherently process data from multiple frames. The limited number of chirps within a single frame restricts the duration of slow-time observations, hindering further improvements in Doppler resolution and estimation accuracy. Furthermore, directly concatenating data from different frames for FFT processing introduces phase discontinuities and spectral distortion due to time intervals between frames, further degrading detection performance.

[0005] Furthermore, while some high-resolution methods based on subspaces or sparse reconstruction can theoretically overcome the resolution limitations of the FFT, they are typically built upon uniform sampling or regular structures. Under piecewise non-uniform sampling conditions, their core model assumptions (such as shift invariance) are difficult to satisfy, making the methods difficult to apply directly or significantly degrading in performance.

[0006] Therefore, there is an urgent need for a radar signal processing method and device based on the joint ESPRIT to improve the above problems. Summary of the Invention

[0007] The purpose of this invention is to provide a radar signal processing method and apparatus based on the joint ESPRIT, which can achieve coherent fusion of cross-frame data, thereby significantly improving the accuracy and resolution of Doppler estimation.

[0008] In a first aspect, the present invention provides a radar signal processing method based on joint ESPRIT, comprising the steps of: acquiring non-uniform slow-time sampling data of the radar and constructing a corresponding signal model; calculating the time difference based on the sampling data and constructing a corresponding equivalent shift relationship in combination with the signal model; performing subspace extraction on the sampling data to obtain a corresponding signal subspace; stacking the equivalent shift relationship and the signal subspace to obtain a joint shift relationship; and calculating the target Doppler frequency based on the joint shift relationship.

[0009] Optionally, acquiring non-uniform slow-time sampling data from the radar and constructing a corresponding signal model includes: acquiring multi-frame segmented non-uniform slow-time sampling data from the radar, wherein the sampling data includes sampling timestamps and complex echo data after range-dimensional fast Fourier transform; constructing a corresponding non-uniform sampling signal model based on the sampling timestamps and echo data for the same range cell or detection target; and / or the signal model is:

[0010]

[0011] in, The target Doppler frequency; This is the timestamp of the actual sampling. For the first The amplitude of the echo of each target; Estimated target quantity; It is noise.

[0012] Optionally, calculating the time difference based on the sampled data and constructing the corresponding equivalent shift relationship in conjunction with the signal model includes: calculating pairwise time differences for all sampling times based on the sampled data, and grouping samples with the same time difference to obtain multiple time difference sets; constructing two sets of data vectors based on the sample pairs corresponding to each time difference, and constructing the corresponding equivalent shift relationship in conjunction with the signal model.

[0013] Optionally, subspace extraction of the sampled data to obtain the corresponding signal subspace includes: constructing a data matrix and a covariance matrix based on the sampled data, and performing eigenvalue decomposition on the covariance matrix to obtain the signal subspace.

[0014] Optionally, obtaining the joint shift relationship by stacking the equivalent shift relationship and the signal subspace includes: constructing a selection matrix for each time difference, and stacking the subspaces corresponding to all time differences according to the signal subspace to obtain the joint shift relationship.

[0015] Optionally, calculating the target Doppler frequency based on the joint shift relationship includes: constructing an augmented matrix based on the joint shift relationship, performing singular value decomposition on the augmented matrix to obtain a rotation matrix, and performing eigenvalue decomposition on the rotation matrix to obtain the target Doppler frequency.

[0016] Secondly, the present invention provides a radar signal processing apparatus based on the Joint ESPRIT, which includes modules / units for performing any of the possible design methods described in the first aspect above. These modules / units can be implemented in hardware or by hardware executing corresponding software.

[0017] Thirdly, the present invention provides an electronic device including a memory and a processor, wherein the memory stores a program executable on the processor, and when the program is executed by the processor, the electronic device implements a method for performing any of the possible designs described above.

[0018] Fourthly, the present invention provides a readable storage medium storing a program, which, when executed, implements a method of any possible design of any of the above aspects.

[0019] Fifthly, the present invention provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0020] The beneficial effects of the method of this invention are as follows: acquiring non-uniform slow-time sampling data from the radar and constructing a corresponding signal model; calculating the time difference based on the sampling data and constructing a corresponding equivalent shift relationship in conjunction with the signal model; performing subspace extraction on the sampling data to obtain a corresponding signal subspace; stacking the equivalent shift relationship and the signal subspace to obtain a joint shift relationship; and calculating the target Doppler frequency based on the joint shift relationship. Without changing the existing radar hardware structure, by uniformly modeling and processing the segmented sampling data, coherent fusion of cross-frame data is achieved, thereby significantly improving the accuracy and resolution of Doppler estimation. Attached Figure Description

[0021] Figure 1 A flowchart illustrating a radar signal processing method based on the joint ESPRIT provided in an embodiment of the present invention;

[0022] Figure 2A schematic diagram of a radar signal processing device based on the Joint ESPRIT provided in an embodiment of the present invention;

[0023] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions in the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention. Unless otherwise defined, the technical or scientific terms used herein should have the ordinary meaning understood by those skilled in the art. The terms "comprising" and similar expressions used herein mean that the element or object preceding the word covers the element or object listed following the word and its equivalents, but do not exclude other elements or objects.

[0025] The technical solutions of the embodiments of the present invention will be described below with reference to the accompanying drawings. In the description of the embodiments of the present invention, the terminology used in the following embodiments is for the purpose of describing specific embodiments only and is not intended to limit the present invention. The singular expressions “a,” “the,” “the,” and “this” are intended to also include expressions such as “one or more,” unless the context clearly indicates otherwise. It should also be understood that in the following embodiments of the present invention, “at least one” and “one or more” refer to one or more (including two). The term “and / or” is used to describe the relationship between related objects, indicating that three relationships can exist; for example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character “ / ” generally indicates that the preceding and following related objects are in an “or” relationship.

[0026] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in connection with that embodiment is included in one or more embodiments of the invention. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," and "in still other embodiments" appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including, but not limited to," unless otherwise specifically emphasized. The term "connection" includes both direct and indirect connections, unless otherwise stated. "First" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated.

[0027] In embodiments of the present invention, "exemplarily" or "for example" are used to indicate that they are examples, illustrations, or descriptions. Any embodiment or design described as "exemplarily" or "for example" in embodiments of the present invention should not be construed as being more preferred or advantageous than other embodiments or design solutions. Rather, the use of "exemplarily" or "for example" is intended to present the relevant concepts in a specific manner.

[0028] like Figure 1 As shown, this invention provides a radar signal processing method based on the joint ESPRIT, including the following steps:

[0029] S101: Acquire non-uniform slow-time sampling data from the radar and construct the corresponding signal model.

[0030] In some embodiments, acquiring non-uniform slow-time sampling data of the radar and constructing a corresponding signal model includes: acquiring multi-frame segmented non-uniform slow-time sampling data of the radar, wherein the sampling data includes sampling timestamps and complex echo data after range dimension fast Fourier transform; and constructing a corresponding non-uniform sampling signal model based on the sampling timestamps and echo data for the same range cell or detection target.

[0031] In other embodiments, the signal model is:

[0032]

[0033] in, The target Doppler frequency; This is the timestamp of the actual sampling. For the first The amplitude of the echo of each target; Estimated target quantity; It is noise.

[0034] S102, calculate the time difference based on the sampled data, and construct the corresponding equivalent shift relationship in conjunction with the signal model.

[0035] In some embodiments, calculating the time difference based on the sampled data and constructing the corresponding equivalent shift relationship in conjunction with the signal model includes: calculating pairwise time differences for all sampling times based on the sampled data, and grouping samples with the same time difference to obtain multiple time difference sets; constructing two sets of data vectors based on the sample pairs corresponding to each time difference, and constructing the corresponding equivalent shift relationship in conjunction with the signal model.

[0036] S103, perform subspace extraction on the sampled data to obtain the corresponding signal subspace.

[0037] In some embodiments, subspace extraction of the sampled data to obtain the corresponding signal subspace includes: constructing a data matrix and a covariance matrix based on the sampled data, and performing eigenvalue decomposition on the covariance matrix to obtain the signal subspace.

[0038] S104, stacking the equivalent shift relationship and signal subspace to obtain the joint shift relationship.

[0039] In some embodiments, stacking the equivalent shift relationship and the signal subspace to obtain the joint shift relationship includes: constructing a selection matrix for each time difference, and stacking the subspaces corresponding to all time differences according to the signal subspace to obtain the joint shift relationship.

[0040] S105, the target Doppler frequency is calculated based on the joint shift relationship.

[0041] In some embodiments, calculating the target Doppler frequency based on the joint shift relationship includes: constructing an augmented matrix based on the joint shift relationship, performing singular value decomposition on the augmented matrix to obtain a rotation matrix, and performing eigenvalue decomposition on the rotation matrix to obtain the target Doppler frequency.

[0042] The advantage of this invention is that by constructing a time difference set and introducing the joint ESPRIT subspace processing method, it achieves effective utilization of non-uniform sampling data. Without changing the existing radar hardware structure, it achieves coherent fusion of cross-frame data by uniformly modeling and subspace processing the segmented sampling data, thereby significantly improving the accuracy and resolution of Doppler estimation.

[0043] To facilitate understanding, this embodiment further elaborates on the specific implementation process of the above method in conjunction with a specific application scenario. Taking vehicle-mounted radar as an example, the specific steps include:

[0044] 1. System Overall Architecture and Data Acquisition

[0045] This invention is applied to an automotive millimeter-wave radar system, which includes a radio frequency front-end (transmit / receive channel), an analog-to-digital converter (ADC), a storage and cache module (on-chip SRAM / external DDR), and a signal processing unit (DSP / GPU / CPU).

[0046] The vehicle-mounted radar system operates on a frame-by-frame basis, with each frame containing several consecutively transmitted chirp signals. The ADC samples each chirp echo and performs a Fast Fourier Transform (FFT) in the range dimension before writing the data to the storage module. Due to the processing and scheduling intervals between frames, the acquired slow-time sampling points are segmented and non-uniformly distributed on the time axis; that is, the sampled data consists of non-uniform slow-time sampling data from multiple frames.

[0047] Non-equidistant

[0048] This invention directly utilizes the non-uniform timestamp data for subsequent processing without modifying the original sampling process.

[0049] 2. Unified modeling of segmented data

[0050] For a selected range cell or detection target, extract its complex echo data across multiple frames to construct a slow time series:

[0051]

[0052] in, The target Doppler frequency; The timestamp of the actual sampling (non-uniform across frames); For the first The amplitude of the echo of each target; Estimated target quantity; It is noise.

[0053] This model transforms the piecewise sampling problem into a non-uniform sampling frequency estimation problem, providing a unified mathematical foundation for subsequent subspace processing.

[0054] 3. Time difference construction and equivalent displacement structure recovery

[0055] Under non-uniform slow-time sampling conditions, due to the sampling time series Since the evenly spaced distribution is not satisfied, the fixed shift invariance structure upon which the traditional ESPRIT method relies no longer holds. To recover this structure, an equivalent time-shift relationship needs to be constructed in the existing sampling points.

[0056] First, calculate the pairwise time difference for all sampling time points based on the sampled data:

[0057]

[0058] Based on this, sample pairs with the same time difference are grouped to obtain multiple time difference sets:

[0059]

[0060] For each time difference You can select all sample pairs that meet the following conditions:

[0061]

[0062] Let the number of sample pairs corresponding to this time difference be . Then two sets of data vectors can be constructed:

[0063]

[0064]

[0065] in, The set of sampling times that satisfy the above time difference condition.

[0066] In an ideal situation (ignoring noise), according to the signal model:

[0067]

[0068] We can obtain:

[0069]

[0070] in, This indicates element-wise multiplication, and:

[0071]

[0072] From the perspective of subspaces, this relationship is equivalent to:

[0073]

[0074] That is, in each time difference Under these conditions, an equivalent shift relationship can be established.

[0075] By constructing multiple different time differences And using all corresponding sample pairs (scale of This allows for the recovery of multiple sets of shift-invariant constraints from non-uniformly sampled data, providing a foundation for subsequent joint ESPRIT subspace solutions. It should be noted that:

[0076] different Corresponding sample size They can be different;

[0077] Time differences with a larger sample size and stable distribution can be prioritized to improve estimation accuracy;

[0078] Multiple time differences together provide redundant constraints, which helps to improve the robustness and noise resistance of the algorithm.

[0079] 4. Data Matrix Construction and Signal Subspace Extraction

[0080] Obtaining non-uniform slow-time sampling data Then, for a specific range cell or detection target, its corresponding complex echo data can be collected over multiple frames, and a slow-time data vector can be constructed:

[0081]

[0082] in, This indicates the total number of sampling points accumulated across frames.

[0083] In practical systems, to improve the robustness of estimation, a data matrix can be constructed using multiple observations or multi-channel data:

[0084]

[0085] in, The number of observations (snapshots) can come from different frames, different antennas, or adjacent distance cells, and each column represents an independent observation.

[0086] Further construct the covariance matrix:

[0087]

[0088] Perform eigenvalue decomposition on the covariance matrix:

[0089]

[0090] in, For signal subspace; For noise subspace; The target number (signal dimension).

[0091] This step enables the mapping from raw observation data to a low-dimensional signal subspace, providing a foundation for subsequent parameter estimation.

[0092] 5. Construction of joint shift relationships

[0093] Based on the time difference set constructed in Part 3 For each time difference The corresponding data vector has been obtained:

[0094]

[0095] in, The number of sample pairs that satisfy this time difference condition.

[0096] Define the selection matrix To satisfy:

[0097]

[0098] Apply the same operation to the signal subspace. ,get:

[0099]

[0100]

[0101] For all time differences The corresponding subspaces are stacked:

[0102]

[0103]

[0104] Establish the following relationship:

[0105]

[0106] in, , It is a rotation matrix.

[0107] 6. Solving rotation matrices using TLS

[0108] Construct the augmented matrix:

[0109]

[0110] Perform singular value decomposition on it:

[0111]

[0112] Will The blocks are as follows:

[0113]

[0114] get:

[0115]

[0116] 7. Doppler frequency estimation

[0117] right Perform eigenvalue decomposition:

[0118]

[0119] From this, the Doppler frequency is obtained:

[0120]

[0121] This invention addresses the problem of high-resolution Doppler estimation for vehicle-mounted radar under segmented slow-time sampling conditions, and proposes a radar signal processing method based on joint ESPRIT. Its key technical aspects are as follows:

[0122] (1) Unified modeling of piecewise non-uniform slow-time sampling

[0123] To address the issues of inter-frame gaps and discontinuous slow-time sampling in vehicle-mounted radar, cross-frame sampling data is uniformly modeled as a non-uniform time series:

[0124]

[0125] An exponential signal model based on the actual sampling time is established, enabling the algorithm to be applied directly to non-uniformly sampled data without resampling or interpolation.

[0126] (2) Construction of equivalent shift structure based on time difference

[0127] Construct a time difference set from all sampling time points. And select sample pairs that meet the conditions to establish multiple data subsets:

[0128]

[0129] This allows for the recovery of equivalent shift relationships in non-uniformly sampled data, providing a structural foundation for subsequent subspace methods. This method does not rely on the uniformity of the original sampling, which is crucial for realizing ESPRIT under non-uniform conditions.

[0130] (3) Subspace Consistent Extraction Mechanism Based on Selection Matrix

[0131] By constructing the selection matrix The data extraction process in step (2) is mapped onto the signal subspace to achieve the following:

[0132]

[0133] as well as:

[0134]

[0135] Ensure consistency between the data layer and the subspace layer processing, so that the non-uniform sampling structure can be effectively utilized in the subspace.

[0136] (4) ESPRIT modeling method with multiple time difference joint constraints

[0137] Utilizing multiple time differences The corresponding subspace relationships are jointly modeled and constructed through stacking:

[0138]

[0139] This method achieves unified constraints on multiple sets of shift relationships. It can fully utilize redundant information in segmented sampling data, improving the stability and accuracy of parameter estimation.

[0140] (5) Robust solution method for rotation matrix based on TLS

[0141] Solving the rotation matrix using the total least squares (TLS) method By performing singular value decomposition on the augmented matrix, the two-sided error is modeled. Compared to the traditional least squares method, TLS exhibits higher estimation accuracy and numerical stability under non-uniform sampling and low signal-to-noise ratio conditions.

[0142] (6) Coherent fusion mechanism for cross-frame data

[0143] By using the aforementioned joint ESPRIT modeling method, coherent processing of segmented sampled data between multiple frames is achieved. Without changing the hardware sampling method, the equivalent slow-time observation length is effectively extended, thereby breaking through the limitation of single-frame processing.

[0144] (7) Feasibility under existing hardware architecture

[0145] This method requires no modification to the radar hardware and is implemented solely based on existing sampling data and processing units.

[0146] No need to increase the ADC sampling rate;

[0147] No need to increase the number of chirps per frame;

[0148] No need to change the TDM-MIMO scheduling structure;

[0149] It can be directly deployed on existing DSP / GPU platforms to achieve performance improvements.

[0150] (8) Improvement in Doppler resolution and estimation accuracy

[0151] Achieve the following under finite sampling conditions through non-uniform sampling modeling and joint subspace processing:

[0152] Higher Doppler resolution;

[0153] More accurate frequency estimation;

[0154] Stronger weak target detection capability;

[0155] It effectively overcomes the problem of traditional FFT methods being limited by slow time length.

[0156] The advantages of this invention are that, addressing the problem of limited Doppler estimation accuracy in vehicle-mounted radar under segmented slow-time sampling conditions, it constructs a time difference set and introduces a joint ESPRIT subspace processing method, thereby achieving effective utilization of non-uniform sampling data. This significantly improves system performance without altering the existing hardware structure, specifically in the following aspects:

[0157] (1) Overcoming the limitation of slow time sampling length and improving Doppler resolution

[0158] Traditional methods typically utilize only continuously sampled data within a single frame, and their Doppler resolution is limited by the number of chirps and the duration of each frame. This invention achieves cross-frame coherent fusion by jointly modeling and processing segmented sampled data from multiple frames, effectively extending the slow-time observation length and thus significantly improving Doppler resolution.

[0159] (2) Achieving high-precision frequency estimation under non-uniform sampling conditions

[0160] In real-world systems, due to the presence of inter-frame intervals, slow-time sampling exhibits a non-uniform distribution, making it difficult to directly apply traditional FFT methods. This invention constructs a model based on the actual sampling time and restores the equivalent shift relationship through time difference construction, enabling the subspace method to function normally under non-uniform sampling conditions, thereby achieving high-precision Doppler estimation.

[0161] (3) No need to modify the hardware structure, improving the overall system performance

[0162] This invention is based entirely on processing existing sampling data, without requiring any changes to the radar hardware architecture, and includes:

[0163] No need to increase the ADC sampling rate;

[0164] No need to increase the number of chirps per frame;

[0165] No need to change the frame structure;

[0166] System performance can be improved through algorithm optimization, which has good engineering applicability.

[0167] (4) Improve weak target detection capability and noise resistance performance

[0168] The joint ESPRIT method, based on subspace decomposition, is more robust to noise than the traditional FFT method. Furthermore, by employing joint constraints across multiple time differences, it effectively utilizes redundant information in the data, improving the stability of frequency estimation and enabling the effective detection of weak targets even under low signal-to-noise ratio conditions.

[0169] (5) Avoid performance degradation caused by directly splicing segmented data.

[0170] In existing methods, directly concatenating different frames of data for FFT processing leads to phase discontinuities due to the inter-frame interval, resulting in spectral leakage and increased sidelobes. This invention, through time difference modeling and subspace processing, avoids dependence on data continuity, thereby eliminating the aforementioned problems.

[0171] (6) Improve the stability and numerical robustness of parameter estimation

[0172] This invention uses the total least squares (TLS) method to solve the rotation matrix. It considers the errors on both sides during the modeling process. Compared with the traditional least squares method, it has higher numerical stability and estimation accuracy under non-uniform sampling and noise conditions.

[0173] (7) The computational complexity is controllable and it is suitable for real-time systems.

[0174] The main calculations in this invention focus on:

[0175] Subspace decomposition (existing processing results can be reused);

[0176] Small-scale matrix operations (dimension: );

[0177] The overall computational complexity is low, making it suitable for real-time implementation on automotive radar DSP or GPU platforms.

[0178] (8) Applicable to various vehicle radar application scenarios

[0179] The method of this invention is applicable to a variety of typical applications, including but not limited to:

[0180] High-precision velocity estimation;

[0181] Multi-target separation;

[0182] Low-speed, weak target detection;

[0183] Perception of complex traffic environments;

[0184] It has strong versatility and expandability.

[0185] like Figure 2As shown, based on the above method, the present invention provides a radar signal processing device based on joint ESPRIT, comprising: an acquisition unit 201, used to acquire non-uniform slow-time sampling data of the radar and construct a corresponding signal model; a construction unit 202, used to calculate the time difference based on the sampling data and construct a corresponding equivalent shift relationship in combination with the signal model; an extraction unit 203, used to extract subspaces from the sampling data to obtain a corresponding signal subspace; a stacking unit 204, used to stack the equivalent shift relationship and the signal subspace to obtain a joint shift relationship; and a calculation unit 205, used to calculate the target Doppler frequency based on the joint shift relationship.

[0186] It should be understood that all relevant content of each step involved in the above method embodiments can be referenced to the functional description of the corresponding functional module, and will not be repeated here. Furthermore, the use of suffixes such as "module," "component," or "unit" to represent elements is merely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "component," or "unit" can be used interchangeably. Terminals can be implemented in various forms. For example, the terminals described in this invention may include mobile terminals such as mobile phones, tablets, laptops, handheld computers, personal digital assistants (PDAs), portable media players (PMPs), navigation devices, wearable devices, smart bracelets, pedometers, etc., as well as fixed terminals such as digital TVs and desktop computers. The following description will use mobile terminals as examples; those skilled in the art will understand that, in addition to elements specifically designed for mobile purposes, the construction according to embodiments of the present invention can also be applied to fixed-type terminals.

[0187] In other embodiments of the present invention, an electronic device 300 is disclosed, such as... Figure 3 As shown, the device may include: one or more processors 301; memory 302; display 303; one or more application programs (not shown); and one or more computer programs 304. These devices can be connected via one or more communication buses 305. The one or more computer programs 304 are stored in the memory 302 and configured to be executed by the one or more processors 301. The one or more computer programs 304 include instructions that can be used to perform actions such as... Figure 1 Each step in the corresponding embodiment.

[0188] Processor 301 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0189] The memory 302 can be an internal storage unit of the electronic device 300, such as a hard disk or RAM of the electronic device 300. The memory 302 can also be an external storage device of the electronic device 300, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or FlashCard equipped on the electronic device 300. Furthermore, the memory 302 can include both internal and external storage units of the electronic device 300. The memory 302 is used to store computer programs and other programs and data required by the electronic device. The memory 302 can also be used to temporarily store data that has been output or will be output.

[0190] The computer program 304 can be divided into one or more modules / units. The one or more modules / units can be a series of computer program instruction segments that can perform a specific function. The instruction segments are used to describe the execution process of the computer program 304 in the electronic device 300.

[0191] In addition to the above-described structure, those skilled in the art will understand that Figure 3 This is merely an example of electronic device 300 and does not constitute a limitation on electronic device 300. Electronic device 300 may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device may also include input / output devices, network access devices, buses, etc.

[0192] Those skilled in the art will understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the functions described above can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this invention. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0193] Based on the above embodiments, the present invention also discloses a computer-readable storage medium having at least one computer program stored thereon, wherein the computer program, when executed by a processor, implements the methods described in the foregoing embodiments.

[0194] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing a processor. The program can be stored in a computer-readable storage medium, which is a non-transitory medium, such as random access memory, read-only memory, flash memory, hard disk, solid-state drive, magnetic tape, floppy disk, optical disk, and any combination thereof. The storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. This available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital video disc (DVD)), or a semiconductor medium (e.g., solid-state drive (SSD)).

[0195] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0196] Although the embodiments of the present invention have been described in detail above, it will be apparent to those skilled in the art that various modifications and variations can be made to these embodiments. The above descriptions are merely embodiments of the present invention and do not limit the patent scope of the present invention. However, it should be understood that such modifications and variations fall within the scope and spirit of the present invention. Moreover, the present invention described herein may have other embodiments and can be implemented or realized in various ways. All equivalent transformations made based on the description and drawings of the present invention, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A radar signal processing method based on joint ESPRIT, characterized in that, Including the following steps: Acquire non-uniform slow-time sampling data from the radar and construct the corresponding signal model; The time difference is calculated based on the sampled data, and the corresponding equivalent shift relationship is constructed in conjunction with the signal model; The sampled data is subjected to subspace extraction to obtain the corresponding signal subspace; The joint shift relationship is obtained by stacking the equivalent shift relationship and the signal subspace. The target Doppler frequency is calculated based on the joint shift relationship.

2. The method according to claim 1, characterized in that, Acquiring non-uniform slow-time sampling data from the radar and constructing the corresponding signal model includes: Acquire multi-frame segmented non-uniform slow-time sampling data from the radar, wherein the sampling data includes sampling timestamps and complex echo data after range-dimensional fast Fourier transform; For the same distance unit or detection target, a corresponding non-uniform sampling signal model is constructed based on the sampling timestamp and echo data; And / or the signal model is: in, The target Doppler frequency; This is the timestamp of the actual sampling. For the first The amplitude of the echo of each target; Estimated target quantity; It is noise.

3. The method according to claim 1 or 2, characterized in that, Calculating the time difference based on the sampled data and constructing the corresponding equivalent shift relationship in conjunction with the signal model includes: Based on the sampling data, calculate the pairwise time difference for all sampling times, and group the samples with the same time difference to obtain multiple time difference sets; Two sets of data vectors are constructed based on the sample pairs corresponding to each time difference, and the corresponding equivalent shift relationship is constructed in conjunction with the signal model.

4. The method according to claim 1, characterized in that, The sampled data is subjected to subspace extraction to obtain the corresponding signal subspace, which includes: Based on the sampled data, a data matrix and a covariance matrix are constructed, and the covariance matrix is ​​decomposed into eigenvalues ​​to obtain the signal subspace.

5. The method according to claim 1, characterized in that, Based on the equivalent shift relationship and the signal subspace, the joint shift relationship is obtained by stacking the data, including: A selection matrix is ​​constructed for each time difference. The subspaces corresponding to all time differences are stacked according to the signal subspace to obtain the joint shift relationship.

6. The method according to claim 1, characterized in that, The target Doppler frequencies calculated based on the joint shift relationship include: An augmented matrix is ​​constructed based on the joint shift relationship, and a singular value decomposition is performed on the augmented matrix to obtain a rotation matrix; The target Doppler frequency is obtained by performing eigenvalue decomposition on the rotation matrix.

7. A radar signal processing apparatus based on Joint ESPRIT, used in the method according to any one of claims 1-6, characterized in that, include: The acquisition unit is used to acquire non-uniform slow-time sampling data from the radar and construct the corresponding signal model. A construction unit is used to calculate the time difference based on the sampled data and construct the corresponding equivalent shift relationship in conjunction with the signal model; The extraction unit is used to extract subspace from the sampled data to obtain the corresponding signal subspace; Stacking units are used to stack according to the equivalent shift relationship and the signal subspace to obtain a joint shift relationship; The calculation unit is used to calculate the target Doppler frequency based on the joint shift relationship.

8. An electronic device, characterized in that, The device includes a memory and a processor, wherein the memory stores a program that can run on the processor, and when the program is executed by the processor, causes the electronic device to perform the method of any one of claims 1-6.

9. A readable storage medium storing a program, characterized in that, When the program is executed, it implements the method of any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1-6.

Images