Micro-doppler curve extraction method and device, electronic equipment and storage medium

Abstract

The application provides a micro-Doppler curve extraction method and device, electronic equipment and a storage medium, and belongs to the technical field of radar target recognition. The method comprises the following steps: performing window fractional Fourier transform on a target radar echo signal to obtain a three-dimensional time-frequency order information tensor; performing Viterbi calculation processing on a time-frequency order energy information set corresponding to a target moment according to the three-dimensional time-frequency order information tensor to obtain a target curve deviation degree corresponding to each time-frequency order energy modulus value in the time-frequency order energy information set and an optimal previous energy modulus value; performing energy curve backtracking processing according to the target curve deviation degree corresponding to each time-frequency order energy modulus value in the multiple time-frequency order energy information sets and the optimal previous energy modulus value to obtain a target time-frequency energy representation curve; and performing approximate inversion processing based on the target time-frequency energy representation curve to obtain a target micro-Doppler curve. The application can further improve the accuracy of micro-Doppler curve extraction in a strong sea clutter interference environment.

Description

Technical Field

[0001] This application relates to the field of radar target recognition technology, and in particular to a micro-Doppler curve extraction method and apparatus, electronic equipment and storage medium. Background Technology

[0002] In the field of radar target identification technology, the micro-Doppler curve of a sea surface target is obtained by using methods such as time-frequency transformation and peak extraction based on the radar echo signal reflected from the sea surface, thus providing data support for target identification. However, when the sea surface target is in a cluttered environment, the clutter signal can interfere with the peak extraction process. This means that the micro-Doppler curve extraction method cannot accurately extract the micro-Doppler curve of the sea surface target, thus limiting its accuracy.

[0003] Currently, some methods in the industry have improved upon this type of micro-Doppler curve extraction. These methods utilize the fractional Fourier transform (FFT) to suppress sea clutter interference by leveraging the energy concentration difference between the echo components of sea surface targets and sea clutter in the FFT domain. The micro-Doppler curve of the sea surface target is then obtained through a peak search method. However, in environments with strong sea clutter interference, this improved micro-Doppler curve extraction method suffers from problems. Strong sea clutter signals generate spurious peaks in the FFT domain, interfering with the peak extraction process and reducing the accuracy of micro-Doppler curve extraction. Therefore, how to further improve the accuracy of micro-Doppler curve extraction in environments with strong sea clutter interference is a pressing technical problem that needs to be solved in the industry. Summary of the Invention

[0004] The main objective of this application is to provide a micro-Doppler curve extraction method, apparatus, electronic device, and storage medium to further improve the accuracy of micro-Doppler curve extraction in environments with strong sea clutter interference.

[0005] To achieve the above objectives, a first aspect of this application proposes a method for micro-Doppler curve extraction, the method comprising: Transmit target radar detection signals; In response to the signal reflection formed by the target radar detection signal illuminating the sea surface target, the corresponding target radar echo signal is acquired; A window fractional Fourier transform is performed on the target radar echo signal to obtain a three-dimensional time-frequency order information tensor. The three-dimensional time-frequency order information tensor includes multiple time-frequency order energy information sets corresponding to multiple target times. The time-frequency order energy information sets include multiple time-frequency order energy magnitudes and corresponding frequency domain coordinate information and order coordinate information. For each target time, Viterbi calculation is performed on the time-frequency order energy information set corresponding to the target time according to the three-dimensional time-frequency order information tensor to obtain the target curve deviation and optimal preceding energy magnitude corresponding to each time-frequency order energy magnitude in the time-frequency order energy information set; Based on the target curve deviation and the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value in the multiple time-frequency order energy information sets, energy curve backtracking is performed to obtain the target time-frequency energy characterization curve; Based on the target time-frequency energy characterization curve, an approximate inversion process is performed to obtain the target micro-Doppler curve of the sea surface target.

[0006] In some embodiments, the step of performing Viterbi calculation processing on the time-frequency order energy information set corresponding to the target time according to the three-dimensional time-frequency order information tensor for each target time, to obtain the target curve deviation and optimal preceding energy magnitude corresponding to each time-frequency order energy magnitude in the time-frequency order energy information set, includes: For each target time, the time-frequency order energy information set corresponding to the adjacent time before the target time is determined in the three-dimensional time-frequency order information tensor; For each time-frequency order energy magnitude value corresponding to each target time, curve deviation analysis is performed based on the time-frequency order energy magnitude value, the corresponding frequency domain coordinate information and order coordinate information, and the time-frequency order energy information set corresponding to the adjacent time, to obtain multiple candidate curve deviations corresponding to the time-frequency order energy magnitude value. The candidate curve deviation with the smallest deviation among the multiple candidate curve deviations is determined as the target curve deviation corresponding to the time-frequency order energy modulus. In the time-frequency order energy information set corresponding to the adjacent time, the time-frequency order energy modulus associated with the target curve deviation is determined as the optimal preceding energy modulus.

[0007] In some embodiments, for each time-frequency order energy magnitude value corresponding to each target time, curve deviation analysis is performed based on the time-frequency order energy magnitude value, the corresponding frequency domain coordinate information and order coordinate information, and the time-frequency order energy information set corresponding to the adjacent time moments to obtain multiple candidate curve deviations corresponding to the time-frequency order energy magnitude value, including: For each time-frequency order energy magnitude value corresponding to each target time, an energy deviation analysis is performed based on the time-frequency order energy magnitude value and the corresponding frequency domain coordinate information and order coordinate information to obtain the energy deviation value corresponding to the time-frequency order energy magnitude value. Based on the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy modulus and the time-frequency order energy information set corresponding to the adjacent time moments, a curve continuity deviation analysis is performed to obtain multiple curve continuity deviations corresponding to the time-frequency order energy modulus. Based on the energy deviation corresponding to the time-frequency order energy modulus, energy correction is performed on each of the multiple curve continuous deviations to obtain multiple candidate curve deviations corresponding to the time-frequency order energy modulus.

[0008] In some embodiments, the step of performing energy deviation analysis based on the time-frequency order energy magnitude and the corresponding frequency domain coordinate information and order coordinate information to obtain the energy deviation corresponding to the time-frequency order energy magnitude includes: Based on the time-frequency order energy modulus and multiple time-frequency order energy modulus values ​​in the time-frequency order energy information set corresponding to the target time, an energy deviation amplitude analysis is performed to obtain the energy deviation amplitude of the time-frequency order energy modulus. Based on the time-frequency order energy modulus value, the corresponding frequency domain coordinate information and order coordinate information, and multiple time-frequency order energy modulus values ​​in the time-frequency order energy information set corresponding to the target time, an energy concentration deviation analysis is performed to obtain the energy concentration deviation value corresponding to the time-frequency order energy modulus value. The energy deviation degree corresponding to the time-frequency order energy modulus is determined based on the energy deviation amplitude and the energy concentration deviation degree.

[0009] In some embodiments, the step of performing energy correction on each of the multiple continuous curve deviations based on the energy deviation corresponding to the time-frequency order energy magnitude to obtain multiple candidate curve deviations corresponding to the time-frequency order energy magnitude includes: Determine the optimal preceding energy information set corresponding to the adjacent time points. The optimal preceding energy information set includes the optimal preceding energy modulus sequence associated with each time-frequency order energy modulus in the time-frequency order energy information set corresponding to the adjacent time points. Based on the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy modulus, the time-frequency order energy information set corresponding to the adjacent time moments and the optimal preceding energy information set, a smoothness deviation analysis is performed to obtain multiple smoothness deviations corresponding to the time-frequency order energy modulus. Based on the energy deviation corresponding to the time-frequency order energy modulus, a comprehensive correction is made to each continuous deviation of the multiple curves and each smooth deviation of the multiple smooth deviations to obtain multiple candidate curve deviations corresponding to the time-frequency order energy modulus.

[0010] In some embodiments, the step of performing energy curve backtracking processing based on the target curve deviation and the optimal preceding energy modulus corresponding to each of the multiple time-frequency order energy information sets to obtain the target time-frequency energy characterization curve includes: Determine the last time among multiple target times, and based on the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value in multiple time-frequency order energy information sets, start from each time-frequency order energy modulus value in the time-frequency order energy information set corresponding to the last time, perform energy curve backtracking to obtain multiple candidate time-frequency energy characterization curves; The total deviation of the candidate time-frequency energy characterization curve is determined based on the deviation of the target curve corresponding to the multiple time-frequency order energy modulus values ​​included in each candidate time-frequency energy characterization curve. The candidate time-frequency energy characterization curve corresponding to the curve with the smallest total deviation among the multiple curves is determined as the target time-frequency energy characterization curve.

[0011] In some embodiments, after performing an approximate inversion process based on the target time-frequency energy characterization curve to obtain the target micro-Doppler curve of the sea surface target, the process includes: The three-dimensional time-frequency order information tensor is filtered based on the multiple time-frequency order energy magnitudes included in the target time-frequency energy characterization curve, as well as the corresponding frequency domain coordinate information and order coordinate information, to obtain the filtered three-dimensional time-frequency order information tensor. Based on the filtered three-dimensional time-frequency order information tensor, the micro-Doppler curve is extracted again to obtain the filtered micro-Doppler curve.

[0012] To achieve the above objectives, a second aspect of this application provides a micro-Doppler curve extraction device, the device comprising: Signal transmitting unit, used to transmit target radar detection signals; The signal acquisition unit is used to acquire the corresponding target radar echo signal in response to the signal reflection formed by the target radar detection signal illuminating the sea surface target. The Fourier transform unit is used to perform a window fractional Fourier transform on the target radar echo signal to obtain a three-dimensional time-frequency order information tensor. The three-dimensional time-frequency order information tensor includes multiple time-frequency order energy information sets corresponding to multiple target times. The time-frequency order energy information sets include multiple time-frequency order energy magnitudes and corresponding frequency domain coordinate information and order coordinate information. The Viterbi calculation unit is used to perform Viterbi calculation processing on the time-frequency order energy information set corresponding to the target time according to the three-dimensional time-frequency order information tensor for each target time, so as to obtain the target curve deviation and the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value in the time-frequency order energy information set. The curve backtracking unit is used to perform energy curve backtracking processing based on the target curve deviation and the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value in the multiple time-frequency order energy information sets, so as to obtain the target time-frequency energy characterization curve. The curve extraction unit is used to perform approximate inversion processing based on the target time-frequency energy characterization curve to obtain the target micro-Doppler curve of the sea surface target.

[0013] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0014] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0015] This application embodiment obtains the corresponding target radar echo signal by transmitting a target radar detection signal and responding to the signal reflection formed by the target radar detection signal illuminating the sea surface target. Then, a windowed fractional Fourier transform is performed on the target radar echo signal to obtain a three-dimensional time-frequency order information tensor. The three-dimensional time-frequency order information tensor includes multiple time-frequency order energy information sets corresponding to multiple target times. The time-frequency order energy information sets include multiple time-frequency order energy magnitudes and corresponding frequency domain coordinate information and order coordinate information. Next, for each target time, Viterbi calculation processing is performed on the time-frequency order energy information set corresponding to the target time according to the three-dimensional time-frequency order information tensor to obtain the target curve deviation and optimal preceding energy magnitude corresponding to each time-frequency order energy magnitude in the time-frequency order energy information set. Then, energy curve backtracking processing is performed based on the target curve deviation and optimal preceding energy magnitude corresponding to each time-frequency order energy magnitude in multiple time-frequency order energy information sets to obtain the target time-frequency energy characterization curve. Finally, an approximate inversion processing is performed based on the target time-frequency energy characterization curve to obtain the target micro-Doppler curve of the sea surface target. Thus, this embodiment of the application can use windowed fractional Fourier transform to make the time-frequency order energy modulus of the corresponding sea surface target and the time-frequency order energy modulus of the corresponding sea clutter in the radar echo signal exhibit different distribution states in the order domain, thereby reducing the interference of spurious peaks generated by sea clutter in the time-frequency domain on the extraction of time-frequency order energy modulus. At the same time, through Viterbi calculation processing, the spurious peaks generated by sea clutter correspond to a higher target curve deviation, and the time-frequency order energy modulus of the sea surface target corresponds to a lower target curve deviation. In this way, during the energy curve backtracking process, spurious peaks with high target curve deviation are avoided, and the target time-frequency energy characterization curve is obtained. Finally, the target micro-Doppler curve is obtained through approximate inversion processing. That is, this embodiment of the application can further improve the accuracy of micro-Doppler curve extraction in a strong sea clutter interference environment. Attached Figure Description

[0016] Figure 1 This is a flowchart of the micro-Doppler curve extraction method provided in the embodiments of this application; Figure 2 yes Figure 1 The flowchart of step S104 in the process; Figure 3 yes Figure 2 The flowchart of step S202 in the document; Figure 4 yes Figure 3 The flowchart of step S301 in the process; Figure 5 yes Figure 3 The flowchart of step S303 in the process; Figure 6 yes Figure 1 The flowchart of step S105 in the process; Figure 7 yes Figure 1 The flowchart of the steps following step S106; Figure 8 This is a schematic diagram of the micro-Doppler curve extraction device provided in the embodiments of this application; Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0018] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0020] First, let's analyze some of the terms used in this application: Micro-Doppler (mD) curves are time-varying frequency curves formed by the additional frequency modulation of radar echo signals caused by the micro-motions of sea surface targets during radar detection, such as vibration, rotation, tumbling, etc. Micro-Doppler curves reflect the micro-motion characteristics of sea surface targets. Different tonnages and structures of sea surface targets produce different micro-motion characteristics, resulting in different micro-Doppler curve shapes. Therefore, micro-Doppler curves can serve as key features for radar target identification. For example, large cargo ships and small fishing boats exhibit different vibration amplitudes and frequencies under wave action, reflected in different frequency variation ranges and periodic characteristics on the micro-Doppler curve. Similarly, patrol boats equipped with propellers and sailboats with sails differ in radar echo modulation characteristics, reflected in the presence of periodic spike components corresponding to the blade rotation frequency on the micro-Doppler curve.

[0021] In the field of radar target identification technology, the micro-Doppler curve of a sea surface target is obtained by using methods such as time-frequency transformation and peak extraction based on the radar echo signal reflected from the sea surface, thus providing data support for target identification. However, when the sea surface target is in a cluttered environment, the clutter signal can interfere with the peak extraction process. This means that the micro-Doppler curve extraction method cannot accurately extract the micro-Doppler curve of the sea surface target, thus limiting its accuracy.

[0022] Currently, some methods in the industry have improved upon this type of micro-Doppler curve extraction. One approach uses the fractional Fourier transform (FRFT) to suppress sea clutter interference by leveraging the energy concentration difference between the echo component of the sea surface target and sea clutter in the FRFT domain. Then, a peak search method is used to obtain the micro-Doppler curve of the sea surface target. However, in environments with strong sea clutter interference, the strong sea clutter signal can generate spurious peaks in the FRFT domain, interfering with the peak extraction process and reducing the accuracy of the micro-Doppler curve extraction. Alternatively, some methods use the short-time fractional Fourier transform (STFRFT), introducing a short-time observation window to perform a FRFT on the signal, obtaining the frequency domain representation of the signal at an instant, and then using peak search to extract the instantaneous frequency of complex moving targets. However, in environments with strong sea clutter interference, this improved micro-Doppler curve extraction method suffers from limitations. Strong sea clutter signals generate spurious peaks in the short-time fractional Fourier transform domain, which overlap with the target energy peak in the time-frequency domain. This results in the peak extraction process failing to accurately identify the target micro-Doppler curve, thus reducing the accuracy of micro-Doppler curve extraction. Therefore, this application provides a micro-Doppler curve extraction method, apparatus, electronic device, and storage medium, aiming to further improve the accuracy of micro-Doppler curve extraction in environments with strong sea clutter interference.

[0023] The micro-Doppler curve extraction method provided in this application relates to the field of radar target recognition technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the micro-Doppler curve extraction method, but is not limited to the above forms.

[0024] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0025] Figure 1 This is an optional flowchart of the micro-Doppler curve extraction method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S101 to S106: Step S101: Transmit the target radar detection signal; Step S102: In response to the signal reflection formed by the target radar detection signal illuminating the sea surface target, the corresponding target radar echo signal is acquired; Step S103: Perform a window fractional Fourier transform on the target radar echo signal to obtain a three-dimensional time-frequency order information tensor. The three-dimensional time-frequency order information tensor includes multiple time-frequency order energy information sets corresponding to multiple target times. The time-frequency order energy information sets include multiple time-frequency order energy magnitudes and corresponding frequency domain coordinate information and order coordinate information. Step S104: For each target time, Viterbi calculation is performed on the time-frequency order energy information set corresponding to the target time according to the three-dimensional time-frequency order information tensor to obtain the target curve deviation and the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value in the time-frequency order energy information set. Step S105: Based on the deviation of the target curve corresponding to the energy modulus value of each time-frequency order in the multiple time-frequency order energy information sets and the optimal preceding energy modulus value, perform energy curve backtracking processing to obtain the target time-frequency energy characterization curve; Step S106: Based on the target time-frequency energy characterization curve, perform approximate inversion processing to obtain the target micro-Doppler curve of the sea surface target.

[0026] Steps S101 to S106 as shown in this embodiment involve transmitting a target radar detection signal and obtaining the corresponding target radar echo signal in response to the signal reflection formed by the target radar detection signal illuminating the sea surface target; then, performing a windowed fractional Fourier transform on the target radar echo signal to obtain a three-dimensional time-frequency order information tensor. This three-dimensional time-frequency order information tensor includes multiple time-frequency order energy information sets corresponding to multiple target times, each time-frequency order energy modulus and corresponding frequency domain coordinate information and order coordinate information; then, for each target… At the target time, Viterbi calculation is performed on the time-frequency order energy information set corresponding to the target time based on the three-dimensional time-frequency order information tensor to obtain the target curve deviation and optimal preceding energy magnitude corresponding to each time-frequency order energy magnitude in the time-frequency order energy information set; then, energy curve backtracking is performed based on the target curve deviation and optimal preceding energy magnitude corresponding to each time-frequency order energy magnitude in multiple time-frequency order energy information sets to obtain the target time-frequency energy characterization curve; finally, approximate inversion processing is performed based on the target time-frequency energy characterization curve to obtain the target micro-Doppler curve of the sea surface target. Thus, this embodiment of the application can use windowed fractional Fourier transform to make the time-frequency order energy modulus of the corresponding sea surface target and the time-frequency order energy modulus of the corresponding sea clutter in the radar echo signal exhibit different distribution states in the order domain, thereby reducing the interference of spurious peaks generated by sea clutter in the time-frequency domain on the extraction of time-frequency order energy modulus. At the same time, through Viterbi calculation processing, the spurious peaks generated by sea clutter correspond to a higher target curve deviation, and the time-frequency order energy modulus of the sea surface target corresponds to a lower target curve deviation. In this way, during the energy curve backtracking process, spurious peaks with high target curve deviation are avoided, and the target time-frequency energy characterization curve is obtained. Finally, the target micro-Doppler curve is obtained through approximate inversion processing. That is, this embodiment of the application can further improve the accuracy of micro-Doppler curve extraction in a strong sea clutter interference environment.

[0027] In step S101 of some embodiments, the target radar detection signal can refer to an electromagnetic wave signal emitted during radar detection. For example, the target radar detection signal can be an electromagnetic wave signal with a fixed pulse repetition frequency, used to be emitted towards a sea surface target during radar detection; or, the target radar detection signal can also be an electromagnetic wave signal with linear frequency modulation characteristics. It is understood that the specific waveform parameters of the target radar detection signal can be adjusted according to actual detection requirements.

[0028] In step S102 of some embodiments, the sea surface target can refer to an object located on the sea surface that can be illuminated by electromagnetic waves. For example, the sea surface target can refer to a ship sailing on the sea surface; or, the sea surface target can refer to a buoy floating on the sea surface. It is understood that the specific type of sea surface target can be adjusted according to the actual detection scenario. The signal reflection formed in response to the target radar detection signal illuminating the sea surface target can refer to the process in which the target radar detection signal is radiated to the sea surface target in the form of electromagnetic waves, and after electromagnetic scattering occurs on the surface of the sea surface target, part of the scattered energy returns along the original propagation path, thereby forming a target radar echo signal. The target radar echo signal can refer to the electromagnetic wave signal formed after the target radar detection signal is reflected by the sea surface target. For example, the target radar echo signal can be the electromagnetic wave signal formed after the target radar detection signal with a fixed pulse repetition frequency is reflected by the sea surface target; or, the target radar echo signal can also be the electromagnetic wave signal formed after the target radar detection signal with linear frequency modulation characteristics is reflected by the sea surface target.

[0029] It should be noted that target radar echo signals are typically a mixture of target signal components and sea clutter signal components. The target signal component can refer to the electromagnetic wave signal reflecting from a sea surface target, carrying information about subtle movements; the sea clutter signal component can refer to the interfering electromagnetic wave signal reflected from scattering bodies such as sea waves. The target radar echo signal can be expressed by the following formula:

[0030] in, Refers to the target radar echo signal. Refers to the target signal components, This refers to the sea clutter signal component. It can be understood that the target signal component in the target radar echo signal can be considered a linear frequency modulated (LFM) signal; a LFM signal can refer to an electromagnetic wave signal whose instantaneous frequency changes linearly with time.

[0031] In step S103 of some embodiments, the windowed fractional Fourier transform can refer to performing a fractional Fourier transform on the target radar echo signal within a short-time observation window, traversing multiple orders to obtain the frequency domain representation of the target radar echo signal at an instant, thereby obtaining the energy distribution information of the target radar echo signal in the time domain, frequency domain, and order domain. The three-dimensional time-frequency order information tensor can refer to a three-dimensional data structure obtained after the windowed fractional Fourier transform, including a set of time-frequency order energy information corresponding to multiple target times. The three-dimensional time-frequency order information tensor is a data structure used to characterize the energy distribution state of the target radar echo signal at different times, different frequency domains, and different orders. The time-frequency order energy information set can refer to the set of energy information corresponding to the target time in the three-dimensional time-frequency order information tensor, including multiple time-frequency order energy magnitudes and corresponding frequency domain coordinate information and order coordinate information.

[0032] The target time can refer to each instantaneous observation moment corresponding to the sliding of the short-time observation window on the time axis of the target radar echo signal during the fractional Fourier transform of the window. The short-time observation window can refer to the time window used to extract local time segments of the target radar echo signal for fractional Fourier transform in the window. For example, the short-time observation window can be a time window centered at 0.2 seconds, covering 0.1 seconds before and after it, and the target time can refer to multiple discrete moments such as 0.2 seconds, 0.3 seconds, and 0.4 seconds corresponding to the short-time observation window sliding on the time axis of the target radar echo signal with a step size of 0.1 seconds; or, the short-time observation window can be a time window centered at 0.5 seconds, covering 0.2 seconds before and after it, and the target time can refer to multiple discrete moments such as 0.5 seconds, 0.55 seconds, and 0.6 seconds corresponding to the short-time observation window sliding on the time axis of the target radar echo signal with a step size of 0.05 seconds. It is understood that the specific duration and sliding step size of the short-time observation window can be adjusted according to actual signal processing requirements.

[0033] The time-frequency order energy magnitude refers to the energy intensity value obtained by taking the modulus of the target radar echo signal after a windowed fractional Fourier transform at the target time. Frequency domain coordinate information refers to the frequency coordinates of the time-frequency order energy magnitude in the fractional Fourier transform frequency domain. Order coordinate information refers to the order coordinates of the time-frequency order energy magnitude in the fractional Fourier transform order domain. For example, since the target radar echo signal includes a mixed signal of target signal components and sea clutter signal components, the time-frequency order energy magnitude can be the energy intensity value obtained by taking the modulus of the target signal component after a windowed fractional Fourier transform at the target time. The frequency domain coordinate information can be the frequency coordinate of this time-frequency order energy magnitude in the fractional Fourier transform frequency domain, and the order coordinate information can be the order coordinate of this time-frequency order energy magnitude in the fractional Fourier transform order domain. Alternatively, the time-frequency order energy magnitude can also be the energy intensity value obtained by taking the modulus of the sea clutter signal component after a windowed fractional Fourier transform at the target time. The frequency domain coordinate information can be another frequency coordinate of this time-frequency order energy magnitude in the fractional Fourier transform frequency domain, and the order coordinate information can be another order coordinate of this time-frequency order energy magnitude in the fractional Fourier transform order domain. Here, the fractional Fourier transform frequency domain can refer to the frequency distribution space of the target radar echo signal after the fractional Fourier transform. The fractional Fourier transform order domain can refer to the distribution space of the target radar echo signal in the rotational angle dimension after undergoing a fractional Fourier transform.

[0034] It should be noted that the specific implementation process of the window fractional Fourier transform can be shown by the following formula:

[0035] in, It can refer to a three-dimensional time-frequency order information tensor; It can refer to the time variable on the time axis of the target radar echo signal, used to represent the value of the signal at different times; It can refer to the target time; it should be noted that, Used for integrating signals. Used to identify the center position of the short-term observation window; It can refer to frequency domain coordinate information; It can refer to the order coordinate information; It can refer to the target's radar echo signal; It can refer to a short-term observation window; It can refer to the fractional Fourier transform kernel; It can refer to time variables Integral infinitesimal elements are used to represent time variables. Perform integration; It can be used for modulo operations.

[0036] It should be noted that for each target time... , In the frequency domain of fractional Fourier transform and order coordinates The values ​​on the plane are used to form a time-frequency order energy information set after modulo operation, which includes multiple time-frequency order energy moduli and corresponding frequency domain coordinate information and order coordinate information.

[0037] It should be noted that the specific calculation principle of the fractional Fourier transform kernel can be shown by the following formula:

[0038]

[0039] in, It can refer to the imaginary unit; It can refer to the angle of rotation; It can refer to an exponential function with the natural constant as its base; It can refer to any integer, and its value can be 0, 1, or -1, etc. It can refer to the Dirac function, which is used to degenerate a fractional Fourier transform kernel into a Fourier transform when the rotation angle is an integer multiple of π.

[0040] In step S104 of some embodiments, the Viterbi calculation process can refer to evaluating the degree of curve deviation between each time-frequency order energy magnitude and each time-frequency order energy magnitude in the time-frequency order energy information set corresponding to the target time in the three-dimensional time-frequency order information tensor, obtaining multiple curve deviation degrees corresponding to each time-frequency order energy magnitude, determining the minimum curve deviation degree among the multiple curve deviation degrees as the final curve deviation degree corresponding to each time-frequency order energy magnitude, and determining the time-frequency order energy magnitude value corresponding to the target time that generated the minimum curve deviation degree. The target curve deviation degree can refer to the value obtained after the Viterbi calculation process, used to quantitatively evaluate whether the time-frequency order energy magnitude value is suitable as the basis for extracting the micro-Doppler curve. It should be noted that the smaller the value of the target curve deviation degree, the more likely the time-frequency order energy magnitude value is to be used as the basis for extracting the micro-Doppler curve; the larger the value of the target curve deviation degree, the less likely the time-frequency order energy magnitude value is to be used as the basis for extracting the micro-Doppler curve. The optimal preceding energy modulus can refer to the time-frequency order energy modulus of the previous target moment corresponding to the deviation from the target curve during the Viterbi calculation process.For example, the time-frequency order energy information set corresponding to target time 1 includes time-frequency order energy magnitude 1 (frequency domain coordinate information is 10 Hz, order coordinate information is 0.5) and time-frequency order energy magnitude 2 (frequency domain coordinate information is 20 Hz, order coordinate information is 0.8); the time-frequency order energy information set corresponding to target time 2 includes time-frequency order energy magnitude 3 (frequency domain coordinate information is 12 Hz, order coordinate information is 0.6) and time-frequency order energy magnitude 4 (frequency domain coordinate information is 22 Hz, order coordinate information is 0.8). The time-frequency order energy information set corresponding to target time 3 includes time-frequency order energy magnitude 5 (frequency domain coordinate information is 14 Hz, order coordinate information is 0.7) and time-frequency order energy magnitude 6 (frequency domain coordinate information is 24 Hz, order coordinate information is 1.0). During the Viterbi calculation process, for the time-frequency order energy magnitude 3 of target time 2, the degree of curve deviation between time-frequency order energy magnitude 3 and each time-frequency order energy magnitude in the time-frequency order energy information set corresponding to target time 1 is analyzed. Without evaluation, if it is determined that the deviation between the time-frequency order energy modulus 3 and the time-frequency order energy modulus 1 is minimal, then the quantized value of the deviation between the time-frequency order energy modulus 3 and the time-frequency order energy modulus 1 in the time-frequency order energy information set corresponding to the target time 1 is determined as the target curve deviation, and the time-frequency order energy modulus 1 is determined as the optimal preceding energy modulus corresponding to the time-frequency order energy modulus 3; or, for the time-frequency order energy modulus 5 at the target time 3, the deviation between the time-frequency order energy modulus 5 and the target time... The deviation of the curves between each time-frequency order energy modulus value in the time-frequency order energy information set corresponding to time 2 is evaluated. If the deviation between the time-frequency order energy modulus value 5 and the time-frequency order energy modulus value 3 is determined to be the smallest, then the quantized value of the deviation between the time-frequency order energy modulus value 5 and the time-frequency order energy modulus value 3 in the time-frequency order energy information set corresponding to the target time 2 is determined as the target curve deviation, and the time-frequency order energy modulus value 3 is determined as the optimal preceding energy modulus value corresponding to the time-frequency order energy modulus value 5.

[0041] It should be noted that among multiple target times, the initial target time does not have a time-frequency order energy information set corresponding to the previous target time. Therefore, the deviation of the target curve for each time-frequency order energy modulus value corresponding to the initial target time, as well as the corresponding optimal preceding energy modulus value, can be predetermined. For example, the deviation of the target curve for each time-frequency order energy modulus value in the time-frequency order energy information set corresponding to the initial target time can be assigned a value based on the energy intensity of each time-frequency order energy modulus value in the time-frequency order energy information set; the higher the energy intensity, the smaller the deviation of the target curve. Furthermore, the optimal preceding energy modulus value for each time-frequency order energy modulus value corresponding to the initial target time can be a preset value, such as 0.

[0042] In step S105 of some embodiments, the energy curve backtracking process can refer to the process of concatenating the energy moduli corresponding to multiple target times along the optimal preceding energy moduli, starting from the energy moduli corresponding to the last target time, based on the target curve deviation and the optimal preceding energy moduli corresponding to each time-frequency order energy moduli in multiple time-frequency order energy information sets, to obtain multiple curves formed by concatenating the energy moduli corresponding to multiple time-frequency order energy moduli corresponding to multiple target times. The curve with the smallest cumulative deviation value of the target curves corresponding to the multiple time-frequency order energy moduli among these curves is then determined as the final curve formed by concatenating the energy moduli corresponding to multiple time-frequency order energy moduli corresponding to multiple target times. The target time-frequency energy characterization curve can refer to the curve obtained after the energy curve backtracking process, which is used as the basis for extracting the micro-Doppler curve and is formed by concatenating the energy moduli corresponding to multiple target times. For example, starting from the time-frequency order energy modulus 5 in the time-frequency order energy information set corresponding to target time 3, and tracing back along the optimal preceding energy modulus chain corresponding to time-frequency order energy modulus 5, we obtain time-frequency order energy modulus 3 and time-frequency order energy modulus 1. We then concatenate time-frequency order energy modulus 1, time-frequency order energy modulus 3, and time-frequency order energy modulus 5 to form a curve formed by concatenating time-frequency order energy modulus values ​​corresponding to multiple target times. Starting from the time-frequency order energy modulus 6 in the time-frequency order energy information set corresponding to target time 3, and tracing back along the optimal preceding energy modulus chain corresponding to time-frequency order energy modulus 6, we obtain time-frequency order energy modulus 4 and time-frequency order energy modulus 2. We then concatenate time-frequency order energy modulus 2, time-frequency order energy modulus 4, and time-frequency order energy modulus 6 to form another curve formed by concatenating time-frequency order energy modulus values ​​corresponding to multiple target times. Based on the cumulative deviation values ​​of the target curves corresponding to time-frequency energy modulus 1, time-frequency energy modulus 3, and time-frequency energy modulus 5, and the cumulative deviation values ​​of the target curves corresponding to time-frequency energy modulus 2, time-frequency energy modulus 4, and time-frequency energy modulus 6, if the cumulative deviation value of the target curves corresponding to time-frequency energy modulus 1, time-frequency energy modulus 3, and time-frequency energy modulus 5 is determined to be the smallest, then the curve formed by concatenating time-frequency energy modulus 1, time-frequency energy modulus 3, and time-frequency energy modulus 5 is determined as the target time-frequency energy characterization curve.

[0043] In step S106 of some embodiments, the approximate inversion processing can refer to the process of performing frequency domain coordinate information and order coordinate information corresponding to each time-frequency order energy magnitude value based on the target time-frequency energy characterization curve, and then concatenating multiple instantaneous Doppler frequencies obtained after frequency conversion to form a time-varying frequency curve. The target micro-Doppler curve of the sea surface target can refer to the time-varying frequency curve obtained after the approximate inversion processing, which is formed by concatenating multiple instantaneous Doppler frequencies corresponding to multiple target times in chronological order and reflects the micro-motion characteristics of the sea surface target. For example, in the approximate inversion process, for the time-frequency order energy magnitude 1 of target time 1 included in the target time-frequency energy characterization curve, the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy magnitude 1 are frequency-converted to obtain the instantaneous Doppler frequency corresponding to target time 1; for the time-frequency order energy magnitude 3 of target time 2 included in the target time-frequency energy characterization curve, the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy magnitude 3 are frequency-converted to obtain the instantaneous Doppler frequency corresponding to target time 2; for the time-frequency order energy magnitude 5 of target time 3 included in the target time-frequency energy characterization curve, the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy magnitude 5 are frequency-converted to obtain the instantaneous Doppler frequency corresponding to target time 3; the instantaneous Doppler frequencies corresponding to target time 1, target time 2, and target time 3 are concatenated in chronological order to form the target micro-Doppler curve of the sea surface target.

[0044] It should be noted that the instantaneous Doppler frequency can refer to the Doppler frequency shift value corresponding to the target time, obtained by frequency transformation of the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy magnitude included in the target time-frequency energy characterization curve during the approximate inversion process. The specific calculation process of the instantaneous Doppler frequency can be shown in the following formula:

[0045] in, It can refer to the instantaneous Doppler frequency corresponding to the target moment; It can refer to the frequency domain coordinate information of the time-frequency order energy modulus value corresponding to the target time in the target time-frequency energy characterization curve; It can refer to the order coordinate information of the time-frequency order energy modulus value corresponding to the target time in the target time-frequency energy characterization curve.

[0046] Please see Figure 2 In some embodiments, step S104 may include, but is not limited to, steps S201 to S203: Step S201: For each target time, determine the set of time-frequency order energy information corresponding to the adjacent time before the target time in the three-dimensional time-frequency order information tensor. Step S202: For each time-frequency order energy modulus value corresponding to each target time, perform curve deviation analysis based on the time-frequency order energy modulus value, the corresponding frequency domain coordinate information and order coordinate information, and the time-frequency order energy information set corresponding to adjacent times to obtain multiple candidate curve deviations corresponding to the time-frequency order energy modulus value. Step S203: The candidate curve deviation with the smallest deviation among multiple candidate curves is determined as the target curve deviation corresponding to the time-frequency order energy modulus value. In the time-frequency order energy information set corresponding to adjacent time moments, the time-frequency order energy modulus value associated with the target curve deviation is determined as the optimal preceding energy modulus value.

[0047] In step S201 of some embodiments, the adjacent time point can refer to the preceding target time point that is temporally adjacent to the target time point in the three-dimensional time-frequency order information tensor. For example, for target time point 2, the adjacent time point can refer to target time point 1, and the time-frequency order energy information set corresponding to target time point 1 includes time-frequency order energy magnitude 1 (frequency domain coordinate information is 10 Hz, order coordinate information is 0.5) and time-frequency order energy magnitude 2 (frequency domain coordinate information is 20 Hz, order coordinate information is 0.8); or, for target time point 3, the adjacent time point can refer to target time point 2, and the time-frequency order energy information set corresponding to target time point 2 includes time-frequency order energy magnitude 3 (frequency domain coordinate information is 12 Hz, order coordinate information is 0.6) and time-frequency order energy magnitude 4 (frequency domain coordinate information is 22 Hz, order coordinate information is 0.9). It is understood that the specific identifier of the adjacent time point can be determined according to the temporal order of the target time point among multiple target time points.

[0048] In step S202 of some embodiments, curve deviation analysis can refer to the process of evaluating the degree of deviation between the time-frequency order energy magnitude value and each time-frequency order energy magnitude value corresponding to the target time, based on the time-frequency order energy magnitude value, the corresponding frequency domain coordinate information, and the order coordinate information, combined with each time-frequency order energy magnitude value in the time-frequency order energy information set corresponding to the adjacent time. Candidate curve deviation can refer to the quantified value of the degree of deviation between the time-frequency order energy magnitude value and each time-frequency order energy magnitude value in the time-frequency order energy information set corresponding to the adjacent time during the curve deviation analysis process. For example, for the time-frequency order energy modulus 3 at target time 2, the deviation between the time-frequency order energy modulus 3 and the time-frequency order energy modulus 1 in the time-frequency order energy information set corresponding to the adjacent time (target time 1) is evaluated to obtain the candidate curve deviation between the time-frequency order energy modulus 3 and the time-frequency order energy modulus 1; the deviation between the time-frequency order energy modulus 3 and the time-frequency order energy modulus 2 in the time-frequency order energy information set corresponding to the adjacent time (target time 1) is evaluated to obtain the candidate curve deviation between the time-frequency order energy modulus 3 and the time-frequency order energy modulus 2. Alternatively, for the time-frequency order energy modulus 5 at target time 3, the deviation between the time-frequency order energy modulus 5 and the time-frequency order energy modulus 3 in the time-frequency order energy information set corresponding to the adjacent time (target time 2) is evaluated to obtain the candidate curve deviation between the time-frequency order energy modulus 5 and the time-frequency order energy modulus 3; the deviation between the time-frequency order energy modulus 5 and the time-frequency order energy modulus 4 in the time-frequency order energy information set corresponding to the adjacent time (target time 2) is evaluated to obtain the candidate curve deviation between the time-frequency order energy modulus 5 and the time-frequency order energy modulus 4.

[0049] In step S203 of some embodiments, the target curve deviation can refer to the smallest curve deviation value determined from multiple candidate curve deviations after the curve deviation analysis is completed, used to indicate the degree of deviation of the time-frequency order energy modulus. The optimal preceding energy modulus can refer to the time-frequency order energy modulus value at the adjacent time to which the target curve deviation is generated during the curve deviation analysis. For example, for the time-frequency order energy modulus value 3 at target time 2, among the candidate curve deviations between time-frequency order energy modulus value 3 and the time-frequency order energy modulus value 1 corresponding to the adjacent time (target time 1), and the candidate curve deviations between time-frequency order energy modulus value 3 and the time-frequency order energy modulus value 2 corresponding to the adjacent time (target time 1), if the candidate curve deviation between time-frequency order energy modulus value 3 and time-frequency order energy modulus value 1 is the smallest, then the candidate curve deviation between time-frequency order energy modulus value 3 and time-frequency order energy modulus value 1 is determined as the target curve deviation, and time-frequency order energy modulus value 1 is determined as the optimal preceding energy modulus value corresponding to time-frequency order energy modulus value 3; Alternatively, for the time-frequency order energy magnitude 5 at target time 3, among the candidate curve deviations between the time-frequency order energy magnitude 5 and the time-frequency order energy magnitude 3 corresponding to the adjacent time (target time 2), and the candidate curve deviations between the time-frequency order energy magnitude 5 and the time-frequency order energy magnitude 4 corresponding to the adjacent time (target time 2), if the candidate curve deviation between the time-frequency order energy magnitude 5 and the time-frequency order energy magnitude 3 is the smallest, then the candidate curve deviation between the time-frequency order energy magnitude 5 and the time-frequency order energy magnitude 3 is determined as the target curve deviation, and the time-frequency order energy magnitude 3 is determined as the optimal preceding energy magnitude corresponding to the time-frequency order energy magnitude 5.

[0050] It is understood that, for each target time, this application embodiment determines the time-frequency order energy information set corresponding to the adjacent time before the target time in the three-dimensional time-frequency order information tensor; then, for each time-frequency order energy magnitude corresponding to the target time, curve deviation analysis is performed based on the time-frequency order energy magnitude, the corresponding frequency domain coordinate information and order coordinate information, and the time-frequency order energy information set corresponding to the adjacent time, to obtain multiple candidate curve deviations corresponding to the time-frequency order energy magnitude; then, the candidate curve deviation with the smallest deviation among the multiple candidate curve deviations is determined as the target curve deviation corresponding to the time-frequency order energy magnitude, and in the time-frequency order energy information set corresponding to the adjacent time, the time-frequency order energy magnitude associated with the target curve deviation is determined as the optimal preceding energy magnitude. Thus, this embodiment of the application determines the degree of curve deviation between each time-frequency order energy modulus and the energy modulus of each time-frequency order at adjacent times through curve deviation analysis, and establishes a correlation between each time-frequency order energy modulus and the time-frequency order energy modulus with the smallest curve deviation at adjacent times through minimum value screening. The target signal component presents similar frequency domain coordinate information and order coordinate information between adjacent times due to continuous motion, and the corresponding target curve deviation is small. The sea clutter signal component presents significantly different frequency domain coordinate information and order coordinate information between adjacent times due to random distribution, and the corresponding target curve deviation is large. The difference between the target signal component and the sea clutter signal component in the target curve deviation is thus amplified, thereby obtaining a more accurate target curve deviation.

[0051] Please see Figure 3 In some embodiments, step S202 may include, but is not limited to, steps S301 to S303: Step S301: For each time-frequency order energy magnitude value corresponding to each target time, perform energy deviation analysis based on the time-frequency order energy magnitude value and the corresponding frequency domain coordinate information and order coordinate information to obtain the energy deviation value corresponding to the time-frequency order energy magnitude value. Step S302: Based on the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy modulus and the time-frequency order energy information set corresponding to adjacent times, perform curve continuity deviation analysis to obtain multiple curve continuity deviations corresponding to the time-frequency order energy modulus. Step S303: Based on the energy deviation corresponding to the time-frequency order energy modulus, perform energy correction on each continuous deviation of multiple curves to obtain multiple candidate curve deviations corresponding to the time-frequency order energy modulus.

[0052] In step S301 of some embodiments, energy deviation analysis can refer to the process of evaluating the degree of energy intensity deviation of each time-frequency order energy magnitude value corresponding to the target time, based on the time-frequency order energy magnitude value and the corresponding frequency domain coordinate information and order coordinate information, within the time-frequency order energy information set corresponding to the target time. Energy deviation can refer to the value reflecting the degree of energy intensity deviation of the time-frequency order energy magnitude value within the time-frequency order energy information set corresponding to the target time during the energy deviation analysis process. It should be noted that the higher the energy intensity of the time-frequency order energy magnitude value, the less deviation it exhibits within the time-frequency order energy information set corresponding to the target time, and the smaller the corresponding energy deviation; conversely, the lower the energy intensity of the time-frequency order energy magnitude value, the more deviation it exhibits within the time-frequency order energy information set corresponding to the target time, and the larger the corresponding energy deviation.

[0053] In step S302 of some embodiments, the curve continuity deviation analysis can refer to the process of evaluating the degree of positional deviation between the time-frequency order energy modulus value and each time-frequency order energy modulus value corresponding to the target time, based on the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy modulus value, combined with each time-frequency order energy modulus value in the time-frequency order energy information set corresponding to adjacent time times. Curve continuity deviation can refer to the value reflecting the degree of positional deviation between the time-frequency order energy modulus value and the time-frequency order energy modulus value corresponding to adjacent time times during the curve continuity deviation analysis process. It should be noted that a time-frequency order energy modulus value with a smaller curve continuity deviation indicates a smaller degree of positional deviation from the time-frequency order energy modulus value corresponding to adjacent time times, and a smaller degree of curve deviation; a time-frequency order energy modulus value with a larger curve continuity deviation indicates a larger degree of positional deviation from the time-frequency order energy modulus value corresponding to adjacent time times, and a larger degree of curve deviation. For example, the specific calculation process of curve continuity deviation can be shown in the following formula:

[0054]

[0055] in, It can refer to the degree of continuous deviation of the curve; It can refer to the target time; It can refer to adjacent moments; It can refer to the time-frequency order energy modulus corresponding to the target time. It can refer to the energy modulus of the time-frequency order corresponding to adjacent moments; It can refer to the frequency domain coordinate information of the energy modulus of the time-frequency order corresponding to the target time. It can refer to the frequency domain coordinate information of the energy modulus of the time-frequency order corresponding to adjacent time points; It can refer to the order coordinate information of the time-frequency order energy modulus corresponding to the target time. It can refer to the order coordinate information of the time-frequency order energy modulus corresponding to adjacent time points; It can refer to the total number of points of the frequency domain coordinate information in the three-dimensional time-frequency order information tensor; It can refer to the total number of points of order coordinate information in the three-dimensional time-frequency order information tensor; d can refer to the normalized distance between the time-frequency order energy magnitude value corresponding to the target time and the time-frequency order energy magnitude value corresponding to the adjacent time. It can refer to the threshold distance in the analysis of continuous deviation of curves. It is a preset value, and the specific value can be adjusted according to actual needs.

[0056] In step S303 of some embodiments, energy correction can refer to the process of correcting the continuous deviation of the curve based on the energy deviation corresponding to the time-frequency order energy magnitude, to obtain a comprehensive curve reflecting the deviation degree of the time-frequency order energy magnitude. Candidate curve deviation can refer to the quantified value of the deviation between the time-frequency order energy magnitude and each time-frequency order energy magnitude in the time-frequency order energy information set corresponding to adjacent times, obtained after correcting the continuous deviation of the curve based on the energy deviation during the energy correction process. For example, regarding the time-frequency order energy magnitude 3 at target time 2, if the curve deviation between the time-frequency order energy magnitude 3 and the time-frequency order energy magnitude 1 at the adjacent time (target time 1) is small, then the curve deviation of the time-frequency order energy magnitude 3 is first determined to be small. If the energy deviation corresponding to the time-frequency order energy magnitude 3 is also small, then the curve deviation of the time-frequency order energy magnitude 3 can be further determined to be even smaller, thereby determining the final candidate curve deviation between the time-frequency order energy magnitude 3 and the time-frequency order energy magnitude 1. If the curve deviation between the time-frequency order energy magnitude 3 and the time-frequency order energy magnitude 2 at the adjacent time (target time 1) is large, then the curve deviation of the time-frequency order energy magnitude 3 is first determined to be large. If the energy deviation corresponding to the time-frequency order energy magnitude 3 is also large, then the curve deviation of the time-frequency order energy magnitude 3 can be further determined to be even larger, thereby determining the final candidate curve deviation between the time-frequency order energy magnitude 3 and the time-frequency order energy magnitude 2. Alternatively, for the time-frequency order energy magnitude 5 at target time 3, if the curve deviation between the time-frequency order energy magnitude 5 and the time-frequency order energy magnitude 3 corresponding to the adjacent time (target time 2) is small, then the curve deviation of the time-frequency order energy magnitude 5 is first determined to be small. If the energy deviation corresponding to the time-frequency order energy magnitude 5 is large, then the curve deviation of the time-frequency order energy magnitude 5 is further increased to a moderate level, thereby determining the final candidate curve deviation between the time-frequency order energy magnitude 5 and the time-frequency order energy magnitude 3. If the curve deviation between the time-frequency order energy magnitude 5 and the time-frequency order energy magnitude 4 corresponding to the adjacent time (target time 2) is large, then the curve deviation of the time-frequency order energy magnitude 5 is first determined to be large. If the energy deviation corresponding to the time-frequency order energy magnitude 5 is small, then the curve deviation of the time-frequency order energy magnitude 5 is further reduced to a moderate level, thereby determining the final candidate curve deviation between the time-frequency order energy magnitude 5 and the time-frequency order energy magnitude 4. It should be noted that in energy correction, time-frequency order energy modulus values ​​with smaller energy deviations can further reduce the deviation of candidate curves, while time-frequency order energy modulus values ​​with larger energy deviations can further increase the deviation of candidate curves.

[0057] It is understood that, in this embodiment of the application, for each time-frequency order energy magnitude value corresponding to each target time, energy deviation analysis is performed based on the time-frequency order energy magnitude value and the corresponding frequency domain coordinate information and order coordinate information to obtain the energy deviation value corresponding to the time-frequency order energy magnitude value; then, based on the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy magnitude value and the time-frequency order energy information set corresponding to adjacent time times, curve continuity deviation analysis is performed to obtain multiple curve continuity deviations; then, based on the energy deviation, energy correction is performed on each of the multiple curve continuity deviations to obtain multiple candidate curve deviations. Thus, in this embodiment, energy deviation is obtained through energy deviation analysis, and curve continuity deviation is obtained through curve continuity deviation analysis. Then, energy correction is used to merge the energy deviation and curve continuity deviation into a candidate curve deviation. This results in a smaller time-frequency order energy magnitude corresponding to a smaller candidate curve deviation, and a larger time-frequency order energy magnitude corresponding to a larger candidate curve deviation. The difference between the target signal component and the sea clutter signal component in the candidate curve deviation is thus amplified, thereby obtaining a more accurate candidate curve deviation.

[0058] Please see Figure 4 In some embodiments, step S301 may include, but is not limited to, steps S401 to S403: Step S401: Analyze the energy deviation amplitude based on the time-frequency order energy modulus and multiple time-frequency order energy modulus values ​​in the time-frequency order energy information set corresponding to the target time, and obtain the energy deviation amplitude of the time-frequency order energy modulus. Step S402: Based on the time-frequency order energy modulus value, the corresponding frequency domain coordinate information and order coordinate information, and multiple time-frequency order energy modulus values ​​in the time-frequency order energy information set corresponding to the target time, perform energy concentration deviation analysis to obtain the energy concentration deviation value corresponding to the time-frequency order energy modulus value. Step S403: Determine the energy deviation corresponding to the time-frequency order energy modulus based on the energy deviation amplitude and energy concentration deviation.

[0059] In step S401 of some embodiments, the energy deviation magnitude analysis can refer to the process of evaluating the relative position of the energy intensity of each time-frequency order energy modulus value in the time-frequency order energy information set corresponding to the target time, based on the time-frequency order energy modulus value and multiple time-frequency order energy modulus values ​​in the time-frequency order energy information set corresponding to the target time, for each time-frequency order energy modulus value corresponding to the target time. The energy deviation magnitude can refer to the value reflecting the ranking position of the energy intensity in the time-frequency order energy information set corresponding to the target time during the energy deviation magnitude analysis process. For example, the specific calculation process of the energy deviation magnitude can be shown in the following formula:

[0060] in, It can refer to the target time; It can refer to frequency domain coordinate information; It can refer to the order coordinate information. "J" can refer to the energy deviation magnitude; "J" can refer to the sequence number of the time-frequency order energy modulus value in the time-frequency order energy information set corresponding to the target time, sorted by energy intensity from high to low. For example, for the time-frequency order energy information set corresponding to target time 1, if the time-frequency order energy modulus values ​​1 and 2 are sorted by energy intensity from high to low, and the energy intensity of time-frequency order energy modulus value 1 is higher than that of time-frequency order energy modulus value 2, then the energy deviation magnitude corresponding to time-frequency order energy modulus value 1 is 1, and the energy deviation magnitude corresponding to time-frequency order energy modulus value 2 is 2. Or, for the time-frequency order energy information set corresponding to target time 2, if the time-frequency order energy modulus values ​​3 and 4 are sorted by energy intensity from high to low, and the energy intensity of time-frequency order energy modulus value 3 is higher than that of time-frequency order energy modulus value 4, then the energy deviation magnitude corresponding to time-frequency order energy modulus value 3 is 1, and the energy deviation magnitude corresponding to time-frequency order energy modulus value 4 is 2. It should be noted that the smaller the energy deviation, the higher the energy intensity of the time-frequency order energy information concentration at the target time, and the larger the energy deviation, the lower the energy intensity of the time-frequency order energy information concentration at the target time.

[0061] In step S402 of some embodiments, the energy concentration deviation analysis can refer to the process of evaluating the energy concentration state of each time-frequency order energy modulus corresponding to the target time, based on the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy modulus, and combining multiple time-frequency order energy modulus values ​​in the time-frequency order energy information set corresponding to the target time, within a local frequency domain and order range. The energy concentration deviation can refer to the value reflecting the degree of deviation of the energy concentration state of the time-frequency order energy modulus within a local frequency domain and order range during the energy concentration deviation analysis process. For example, the specific calculation process of the energy concentration deviation can be shown in the following formula:

[0062]

[0063] in, It can refer to the degree of deviation in energy concentration; It can refer to the degree of energy concentration; It can refer to the time-frequency order energy modulus corresponding to the target time. It can refer to the neighborhood index within the frequency domain energy observation window; It can refer to the neighborhood index within the energy observation window of the order field; It can refer to the half-window width of the frequency domain energy observation window; It can refer to the half-width of the energy observation window in the order domain; It can refer to multiple time-frequency order energy moduli surrounding the current time-frequency order energy moduli at the target time, as well as multiple time-frequency order energy moduli within the order neighborhood. It can refer to the minimum energy concentration at the target time. It can refer to the maximum energy concentration at the target time.

[0064] It should be noted that the smaller the energy concentration deviation, the higher the degree of energy concentration of the energy modulus of the time-frequency order within the local frequency domain and order range corresponding to the target time; the larger the energy concentration deviation, the lower the degree of energy concentration of the energy modulus of the time-frequency order within the local frequency domain and order range corresponding to the target time.

[0065] In step S403 of some embodiments, the energy deviation can refer to a value that reflects the degree of deviation of the energy intensity of the time-frequency order energy modulus, determined comprehensively based on the energy deviation amplitude and energy concentration deviation corresponding to the time-frequency order energy modulus. For example, the energy deviation can be the value obtained by directly summing the energy deviation amplitude and energy concentration deviation corresponding to the time-frequency order energy modulus; or, the energy deviation can also be the value obtained by multiplying the energy deviation amplitude and energy concentration deviation corresponding to the time-frequency order energy modulus by preset weights and then summing them. It should be noted that the specific method for determining the energy deviation can be adjusted according to actual needs.

[0066] It is understood that, in this embodiment of the application, for each time-frequency order energy magnitude value corresponding to each target time, energy deviation amplitude analysis is performed based on the time-frequency order energy magnitude value and multiple time-frequency order energy magnitude values ​​in the time-frequency order energy information set corresponding to the target time to obtain the energy deviation amplitude of the time-frequency order energy magnitude value; then, based on the time-frequency order energy magnitude value, the corresponding frequency domain coordinate information and order coordinate information, and multiple time-frequency order energy magnitude values ​​in the time-frequency order energy information set corresponding to the target time to obtain the energy concentration deviation corresponding to the time-frequency order energy magnitude value; and finally, the energy deviation corresponding to the time-frequency order energy magnitude value is determined based on the energy deviation amplitude and the energy concentration deviation. Thus, this embodiment of the application, through energy deviation amplitude analysis and energy concentration deviation analysis, integrates global energy intensity ranking information and local energy concentration state information to determine the energy deviation, reducing the limitations of single-dimensional energy assessment in distinguishing target signal components from sea clutter signal components, improving the comprehensiveness and accuracy of energy deviation assessment, and thereby obtaining a more accurate energy deviation.

[0067] Please see Figure 5In some embodiments, step S303 may also include, but is not limited to, steps S501 to S503: Step S501: Determine the optimal preceding energy information set corresponding to adjacent time points. The optimal preceding energy information set includes the optimal preceding energy modulus sequence associated with each time-frequency order energy modulus in the time-frequency order energy information set corresponding to adjacent time points. Step S502: Based on the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy modulus, the time-frequency order energy information set corresponding to adjacent time moments and the optimal preceding energy information set, perform smoothness deviation analysis to obtain multiple smoothness deviations corresponding to the time-frequency order energy modulus. Step S503: Based on the energy deviation corresponding to the time-frequency order energy modulus, perform comprehensive correction on each continuous deviation of multiple curves and each smooth deviation of multiple smooth deviations to obtain multiple candidate curve deviations corresponding to the time-frequency order energy modulus.

[0068] In step S501 of some embodiments, the optimal preceding energy information set may refer to the optimal preceding energy modulus sequence associated with each time-frequency order energy modulus in the time-frequency order energy information set corresponding to adjacent time moments, and the set of frequency domain coordinate information and order coordinate information corresponding to each time-frequency order energy modulus included in the optimal preceding energy modulus sequence. The optimal preceding energy modulus sequence may refer to a sequence formed by concatenating multiple optimal preceding energy modulus values ​​in chronological order, starting from the time-frequency order energy modulus corresponding to adjacent time moments and tracing back step-by-step according to the optimal preceding energy modulus associated with each time-frequency order energy modulus until the initial time moment. For example, if the target time is 3 and the adjacent time is 2, the optimal preceding energy modulus sequence associated with the time-frequency order energy modulus 3 in the time-frequency order energy information set corresponding to the adjacent time is a sequence formed by concatenating the time-frequency order energy modulus 1 and the time-frequency order energy modulus 3 in chronological order (because the optimal preceding energy modulus associated with the time-frequency order energy modulus 3 is the time-frequency order energy modulus 1); or, if the target time is 4 and the adjacent time is 3, the optimal preceding energy modulus sequence associated with the time-frequency order energy modulus 5 in the time-frequency order energy information set corresponding to the adjacent time is a sequence formed by concatenating the time-frequency order energy modulus 1, the time-frequency order energy modulus 3, and the time-frequency order energy modulus 5 in chronological order (because the optimal preceding energy modulus associated with the time-frequency order energy modulus 5 is the time-frequency order energy modulus 3, and the optimal preceding energy modulus associated with the time-frequency order energy modulus 3 is the time-frequency order energy modulus 1).

[0069] In step S502 of some embodiments, smoothness deviation analysis can refer to the process of evaluating the smoothness of the trajectory formed by the time-frequency order energy magnitude and the optimal preceding energy magnitude sequence for each time-frequency order energy magnitude corresponding to the target time, based on the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy magnitude, the time-frequency order energy information set corresponding to adjacent times, and the optimal preceding energy information set. Smoothness deviation can refer to the value reflecting the smoothness of the trajectory formed by the time-frequency order energy magnitude and the optimal preceding energy magnitude sequence during the smoothness deviation analysis process. For example, for the time-frequency order energy magnitude value 5 at target time 3, with an adjacent time of 2, the optimal preceding energy magnitude value sequence associated with the time-frequency order energy magnitude value 3 at the adjacent time is a sequence formed by concatenating the time-frequency order energy magnitude values ​​1 and 3 in chronological order. Adding the time-frequency order energy magnitude value 5 to this sequence constitutes a trajectory including the time-frequency order energy magnitude values ​​1, 3, and 5. Second-order difference calculations are performed on the frequency domain coordinate information and order coordinate information of the time-frequency order energy magnitude values ​​5, 3, and 1 in this trajectory to obtain the smoothness deviation corresponding to the time-frequency order energy magnitude value 5. Alternatively, for the time-frequency order energy magnitude value 7 at target time 4, with an adjacent time of 3, the time-frequency order energy magnitude value 5 at the adjacent time is... The optimal preceding energy modulus sequence is a sequence formed by concatenating time-frequency order energy modulus 1, 3, and 5 in chronological order. Adding time-frequency order energy modulus 7 to this sequence creates a trajectory comprising time-frequency order energy modulus 1, 3, 5, and 7. Second-order difference calculations are performed on the frequency domain coordinates and order coordinates of time-frequency order energy modulus 5, 3, and 1 in this trajectory, as well as on the frequency domain coordinates and order coordinates of time-frequency order energy modulus 7, 5, and 3. The results of the two second-order difference calculations are then summed to obtain the smoothness deviation corresponding to time-frequency order energy modulus 7.

[0070] It should be noted that the specific calculation process for the smoothing deviation can be shown in the following formula:

[0071] in, It can refer to the smoothness deviation; It can refer to adjacent moments; It can refer to the target time; It can refer to the time-frequency order energy modulus corresponding to the target time. It can refer to the energy modulus of the time-frequency order corresponding to adjacent moments; It can refer to the frequency domain coordinate information of the energy modulus of the time-frequency order corresponding to the target time. It can refer to the frequency domain coordinate information of the energy modulus of the time-frequency order corresponding to adjacent time points; It can refer to the order coordinate information of the time-frequency order energy modulus corresponding to the target time. It can refer to the order coordinate information of the time-frequency order energy modulus corresponding to adjacent time points; It can refer to the length of the smoothness observation window, which is an integer greater than or equal to three; It can refer to the trajectory point index; It can refer to the first The order coordinate information of each trajectory point; It can refer to the first The order coordinate information of each trajectory point; It can refer to the first The order coordinate information of each trajectory point; It can refer to the first Frequency domain coordinate information of each trajectory point; It can refer to the first Frequency domain coordinate information of each trajectory point; It can refer to the first Frequency domain coordinate information of each trajectory point; It can refer to absolute value operations.

[0072] It should be noted that time-frequency order energy modulus values ​​with smaller smoothness deviations indicate a higher degree of smoothness in the trajectory formed by the optimal preceding energy modulus value sequence, and the corresponding curve deviation is smaller; time-frequency order energy modulus values ​​with larger smoothness deviations indicate a lower degree of smoothness in the trajectory formed by the optimal preceding energy modulus value sequence, and the corresponding curve deviation is larger.

[0073] It should be noted that the smoothing deviation analysis is based on the smoothness of the trajectory formed by the optimal preceding energy modulus sequence. Therefore, the smoothing deviation analysis starts from the third target time among multiple target times, and the time-frequency order energy modulus corresponding to the first two target times does not participate in the smoothing deviation analysis.

[0074] In step S503 of some embodiments, comprehensive correction may refer to the process of comprehensively adjusting the continuous deviation and smooth deviation of the curve based on the energy deviation corresponding to the time-frequency order energy magnitude value, to obtain a comprehensive curve reflecting the deviation degree of the time-frequency order energy magnitude value. The candidate curve deviation may refer to the quantified value of the deviation between the time-frequency order energy magnitude value and each time-frequency order energy magnitude value in the time-frequency order energy information set corresponding to adjacent times, obtained after comprehensively adjusting the continuous deviation and smooth deviation of the curve based on the energy deviation during the comprehensive correction process. For example, the candidate curve deviation may be the value obtained by directly summing the energy deviation corresponding to the time-frequency order energy magnitude value, the continuous deviation of the curve, and the smooth deviation; or, the candidate curve deviation may also be the value obtained by multiplying the energy deviation corresponding to the time-frequency order energy magnitude value, the continuous deviation of the curve, and the smooth deviation by preset weights and then summing them.

[0075] It should be noted that the energy deviation is determined based on the energy deviation amplitude and the energy concentration deviation. Therefore, the candidate curve deviation can also be considered as being jointly determined by the energy deviation amplitude, the energy concentration deviation, the curve continuity deviation, and the smoothness deviation. For example, the specific formula for calculating the candidate curve deviation can be shown below:

[0076] in, It can refer to the target time; It can refer to the frequency domain coordinate information of the energy modulus of the time-frequency order corresponding to the target time. It can refer to the order coordinate information of the time-frequency order energy modulus corresponding to the target time. It can refer to adjacent moments; It can refer to the frequency domain coordinate information of the energy modulus of the time-frequency order corresponding to adjacent time points; It can refer to the order coordinate information of the time-frequency order energy modulus corresponding to adjacent time points; It can refer to the magnitude of the energy deviation of the time-frequency order energy modulus corresponding to the target time in the time-frequency order energy information set corresponding to the target time; It can refer to the degree of continuous deviation of the curve between the energy modulus of the time-frequency order corresponding to adjacent time points and the energy modulus of the time-frequency order corresponding to the target time point; It can refer to the degree of energy concentration deviation of the time-frequency order energy modulus corresponding to the target time within the local frequency domain and order range corresponding to the target time; It can refer to the smoothness deviation of the trajectory formed by adding the optimal preceding energy modulus sequence associated with the time-frequency order energy modulus of adjacent time points to the time-frequency order energy modulus corresponding to the target time point; It can refer to the deviation of the candidate curve from the time-frequency order energy modulus value corresponding to the adjacent time time to the time-frequency order energy modulus value corresponding to the target time. This can refer to the preset weighting coefficient corresponding to the magnitude of energy deviation; It can refer to the preset weighting coefficient corresponding to the continuous deviation of the curve; This can refer to the preset weighting coefficient corresponding to the deviation of energy concentration; It can refer to the preset weighting coefficient corresponding to the smoothness deviation; It is understood that, for each time-frequency order energy magnitude corresponding to each target time, this application embodiment determines the optimal preceding energy information set corresponding to adjacent time moments; then, based on the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy magnitude, the time-frequency order energy information set corresponding to adjacent time moments, and the optimal preceding energy information set, a smoothness deviation analysis is performed to obtain multiple smoothness deviations corresponding to the time-frequency order energy magnitude; furthermore, based on the energy deviations corresponding to the time-frequency order energy magnitude, each continuous deviation of multiple curves and each smoothness deviation of multiple smoothness deviations are comprehensively corrected to obtain multiple candidate curve deviations corresponding to the time-frequency order energy magnitude. Thus, by introducing historical trajectory information composed of the optimal preceding energy information set, this embodiment of the application evaluates the smoothness between the time-frequency order energy magnitude and the historical trajectory, making the random irregular pseudo-trajectories generated by sea clutter correspond to a higher smoothness deviation, and making the continuous smooth motion trajectory corresponding to the sea surface target correspond to a lower smoothness deviation. Consequently, the random irregular pseudo-trajectories of sea clutter correspond to a higher candidate curve deviation, and the continuous smooth motion trajectory of the sea surface target correspond to a lower candidate curve deviation. The difference between the target signal component and the sea clutter signal component in the candidate curve deviation is thus amplified, resulting in a more accurate candidate curve deviation.

[0077] Please see Figure 6 In some embodiments, step S105 may include, but is not limited to, steps S601 to S603: Step S601: Determine the last time among multiple target times; based on the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value in multiple time-frequency order energy information sets, start from each time-frequency order energy modulus value in the time-frequency order energy information set corresponding to the last time, perform energy curve backtracking, and obtain multiple candidate time-frequency energy characterization curves. Step S602: Determine the total deviation of the candidate time-frequency energy characterization curves based on the deviation of the target curves corresponding to the multiple time-frequency order energy moduli included in each candidate time-frequency energy characterization curve. Step S603: The candidate time-frequency energy characterization curve corresponding to the curve with the smallest total deviation among multiple curves is determined as the target time-frequency energy characterization curve.

[0078] In step S601 of some embodiments, the last moment can refer to the last target moment among multiple target moments arranged in chronological order. For example, if the multiple target moments include target moment 1, target moment 2, and target moment 3, then the last moment can refer to target moment 3. Energy curve backtracking can refer to the process of starting from each time-frequency order energy modulus value in the time-frequency order energy information set corresponding to the last moment, and backtracking step by step according to the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value until the initial moment, and connecting multiple time-frequency order energy modulus values ​​passed during the backtracking process in chronological order to form a trajectory. Candidate time-frequency energy characterization curve can refer to the curve formed by backtracking and connecting multiple time-frequency order energy modulus values ​​corresponding to multiple target moments in chronological order during the energy curve backtracking process, starting from a time-frequency order energy modulus value in the time-frequency order energy information set corresponding to the last moment. For example, for the time-frequency order energy modulus 5 in the time-frequency order energy information set corresponding to the last moment, backtracking is performed based on the optimal preceding energy modulus (time-frequency order energy modulus 3) corresponding to time-frequency order energy modulus 5, and then backtracking is performed again based on the optimal preceding energy modulus (time-frequency order energy modulus 1) corresponding to time-frequency order energy modulus 3 to the initial moment. Time-frequency order energy modulus 1, time-frequency order energy modulus 3 and time-frequency order energy modulus 5 are connected in time order to form a candidate time-frequency energy characterization curve; Alternatively, for the time-frequency order energy modulus 6 in the time-frequency order energy information set corresponding to the last moment, backtracking is performed based on the optimal preceding energy modulus (time-frequency order energy modulus 4) corresponding to time-frequency order energy modulus 6, and then backtracking is performed again based on the optimal preceding energy modulus (time-frequency order energy modulus 2) corresponding to time-frequency order energy modulus 4 to the initial moment. Time-frequency order energy modulus 2, time-frequency order energy modulus 4 and time-frequency order energy modulus 6 are concatenated in time order to form another candidate time-frequency energy characterization curve.

[0079] In step S602 of some embodiments, the total curve deviation can refer to the value obtained by summing the deviations of the target curves corresponding to the multiple time-frequency order energy moduli included in the candidate time-frequency energy characterization curve. For example, for the time-frequency order energy moduli 1, time-frequency order energy moduli 3, and time-frequency order energy moduli 5 included in the candidate time-frequency energy characterization curve, the deviations of the target curves corresponding to time-frequency order energy moduli 1, time-frequency order energy moduli 3, and time-frequency order energy moduli 5 are summed to obtain the total curve deviation corresponding to the candidate time-frequency energy characterization curve (the curve formed by connecting time-frequency order energy moduli 1, time-frequency order energy moduli 3, and time-frequency order energy moduli 5). Alternatively, for the time-frequency energy characterization curves including time-frequency order energy magnitude 2, time-frequency order energy magnitude 4, and time-frequency order energy magnitude 6, the deviation of the target curve corresponding to time-frequency order energy magnitude 2, the deviation of the target curve corresponding to time-frequency order energy magnitude 4, and the deviation of the target curve corresponding to time-frequency order energy magnitude 6 are summed to obtain the total deviation of the candidate time-frequency energy characterization curve (the curve formed by the cascaded connection of time-frequency order energy magnitude 2, time-frequency order energy magnitude 4, and time-frequency order energy magnitude 6).

[0080] In step S603 of some embodiments, the target time-frequency energy characterization curve may refer to the candidate time-frequency energy characterization curve with the smallest total deviation among multiple candidate time-frequency energy characterization curves. For example, among the total deviations of the candidate time-frequency energy characterization curves (the curve of time-frequency order energy magnitude 1, time-frequency order energy magnitude 3, and time-frequency order energy magnitude 5 connected in series) and the total deviations of the candidate time-frequency energy characterization curves (the curve of time-frequency order energy magnitude 2, time-frequency order energy magnitude 4, and time-frequency order energy magnitude 6 connected in series), if the total deviation of the candidate time-frequency energy characterization curves (the curve of time-frequency order energy magnitude 1, time-frequency order energy magnitude 3, and time-frequency order energy magnitude 5 connected in series) is smaller, then the candidate time-frequency energy characterization curves (the curve of time-frequency order energy magnitude 1, time-frequency order energy magnitude 3, and time-frequency order energy magnitude 5 connected in series) are determined as the target time-frequency energy characterization curve.

[0081] It is understood that, in this embodiment of the application, the final time is determined among multiple target times. Based on the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value in multiple time-frequency order energy information sets, energy curve backtracking is performed starting from each time-frequency order energy modulus value in the time-frequency order energy information set corresponding to the final time to obtain multiple candidate time-frequency energy characterization curves. Then, based on the target curve deviation degree corresponding to the multiple time-frequency order energy modulus values ​​included in each candidate time-frequency energy characterization curve, the total curve deviation degree corresponding to the candidate time-frequency energy characterization curve is determined. Finally, the candidate time-frequency energy characterization curve corresponding to the smallest total curve deviation degree among the multiple curve deviation degrees is determined as the target time-frequency energy characterization curve. Thus, this embodiment of the application obtains multiple candidate time-frequency energy characterization curves by backtracking from the energy modulus values ​​of multiple time-frequency orders corresponding to the last moment. This reduces the risk of interference from local noise or sea clutter pseudo-peaks at a single backtracking starting point. On this basis, the deviation of the target curve at each moment is accumulated into the total deviation of the curve. This results in a relatively low total deviation of the curve for the continuous and smooth motion trajectory of the sea surface target due to the relatively small deviation of the target curve at each moment, while a relatively high total deviation of the curve for the random and irregular pseudo-trajectory of the sea clutter due to the relatively large deviation of the target curve at some moments. In this way, the candidate time-frequency energy characterization curve with a relatively small total deviation is determined as the target time-frequency energy characterization curve, reducing the risk of the sea clutter pseudo-trajectory being determined as the target time-frequency energy characterization curve and improving the accuracy of micro-Doppler curve extraction in strong sea clutter scenarios.

[0082] Please see Figure 7 In some embodiments, after step S106, steps S701 to S702 may also be included, but are not limited to: Step S701: Based on the multiple time-frequency order energy magnitudes included in the target time-frequency energy characterization curve, as well as the corresponding frequency domain coordinate information and order coordinate information, the three-dimensional time-frequency order information tensor is filtered to obtain the filtered three-dimensional time-frequency order information tensor. Step S702: Extract the micro-Doppler curve again based on the filtered three-dimensional time-frequency order information tensor to obtain the filtered micro-Doppler curve.

[0083] In step S701 of some embodiments, the filtering process can refer to the process of covering a certain range in the frequency domain and order domain of a three-dimensional time-frequency order information tensor, centered on the frequency domain coordinate information corresponding to each time-frequency order energy magnitude value included in the target time-frequency energy characterization curve, and covering a certain range in the order domain, centered on the order domain coordinate information corresponding to each time-frequency order energy magnitude value, based on the multiple time-frequency order energy magnitude values ​​included in the target time-frequency energy characterization curve. The process then sets the time-frequency order energy magnitude values ​​within the covered range to zero, thereby eliminating the energy components corresponding to the target time-frequency energy characterization curve. The filtered three-dimensional time-frequency order information tensor can refer to the three-dimensional time-frequency order information tensor in which the energy components corresponding to the target time-frequency energy characterization curve have been eliminated after the filtering process. For example, considering the target time-frequency energy characterization curve, including the time-frequency order energy magnitude 1 at target time 1 (frequency domain coordinate information is 10 Hz, order coordinate information is 0.5), the time-frequency order energy magnitude 3 at target time 2 (frequency domain coordinate information is 12 Hz, order coordinate information is 0.6), and the time-frequency order energy magnitude 5 at target time 3 (frequency domain coordinate information is 14 Hz, order coordinate information is 0.7), in the three-dimensional time-frequency order information tensor, a certain range is covered in the frequency domain centered on the frequency domain coordinate information of 10 Hz corresponding to the time-frequency order energy magnitude 1, and a certain range is covered in the order domain centered on the order coordinate information of 0.5 corresponding to the time-frequency order energy magnitude 1. The following steps are taken: First, a certain range is covered in the frequency domain, centered at the frequency domain coordinate information of time-frequency order energy magnitude 3 (12 Hz). Then, a certain range is covered in the order domain, centered at the order coordinate information of time-frequency order energy magnitude 3 (0.6 Hz). The time-frequency order energy magnitudes within this covered range are then set to zero. Similarly, a certain range is covered in the frequency domain, centered at the frequency domain coordinate information of time-frequency order energy magnitude 5 (14 Hz). Finally, a certain range is covered in the order domain, centered at the order coordinate information of time-frequency order energy magnitude 5 (0.7 Hz). The time-frequency order energy magnitudes within these covered ranges are then set to zero. This process yields the filtered three-dimensional time-frequency order information tensor. It should be noted that the size of the coverage range in the frequency domain and the size of the coverage range in the order domain can be adjusted according to actual needs.

[0084] In step S702 of some embodiments, micro-Doppler curve extraction can refer to the process of re-performing Viterbi calculation, energy curve backtracking, and approximate inversion processing on the filtered three-dimensional time-frequency order information tensor. The filtered micro-Doppler curve can refer to the micro-Doppler curve obtained after micro-Doppler curve extraction from the filtered three-dimensional time-frequency order information tensor. It should be noted that sea surface targets typically include multiple parts capable of reflecting target radar detection signals. Each part capable of reflecting target radar detection signals can be called a scattering point. The target radar echo signal includes signal components reflected from multiple scattering points, and each scattering point corresponds to a micro-Doppler curve. In the three-dimensional time-frequency order information tensor, the energy concentration positions of different scattering points are different. By filtering the three-dimensional time-frequency order information tensor and eliminating the energy component of a scattering point corresponding to the extracted target time-frequency energy characterization curve, the filtered three-dimensional time-frequency order information tensor retains the energy concentration information of other scattering points. By re-extracting the micro-Doppler curves from the filtered three-dimensional time-frequency order information tensor, the filtered micro-Doppler curves corresponding to other scattering points can be extracted from the filtered three-dimensional time-frequency order information tensor. By repeatedly performing the filtering process and micro-Doppler curve extraction, the energy components of the scattering points corresponding to the extracted micro-Doppler curves can be eliminated one by one, and the micro-Doppler curves corresponding to the remaining scattering points can be extracted one by one, thereby achieving the separation and extraction of micro-Doppler curves of multiple scattering points.

[0085] It is understood that, in this embodiment, the three-dimensional time-frequency order information tensor is filtered based on the multiple time-frequency order energy magnitudes included in the target time-frequency energy characterization curve, as well as the corresponding frequency domain coordinate information and order coordinate information, to obtain a filtered three-dimensional time-frequency order information tensor. Then, the micro-Doppler curve is extracted again based on the filtered three-dimensional time-frequency order information tensor to obtain a filtered micro-Doppler curve. Thus, by filtering the three-dimensional time-frequency order information tensor, this embodiment reduces the interference of the energy components of the scattering points corresponding to the extracted target time-frequency energy characterization curve on the energy aggregation information of the remaining scattering points in the three-dimensional time-frequency order information tensor. This reduces the adverse effects of the overlapping of multiple scattering point energy components on the accuracy of micro-Doppler curve extraction, thereby improving the accuracy of separating and extracting micro-Doppler curves from multiple scattering points.

[0086] Please see Figure 8 This application also provides a micro-Doppler curve extraction device, which can realize the above-mentioned micro-Doppler curve extraction method. The device includes: Signal transmitting unit 801 is used to transmit target radar detection signals; The signal acquisition unit 802 is used to acquire the corresponding target radar echo signal in response to the signal reflection formed by the target radar detection signal illuminating the sea surface target. Fourier transform unit 803 is used to perform window fractional Fourier transform on the target radar echo signal to obtain a three-dimensional time-frequency order information tensor. The three-dimensional time-frequency order information tensor includes a set of time-frequency order energy information corresponding to multiple target times. The set of time-frequency order energy information includes multiple time-frequency order energy magnitudes and corresponding frequency domain coordinate information and order coordinate information. The Viterbi calculation unit 804 is used to perform Viterbi calculation processing on the time-frequency order energy information set corresponding to the target time according to the three-dimensional time-frequency order information tensor for each target time, so as to obtain the target curve deviation and the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value in the time-frequency order energy information set. The curve backtracking unit 805 is used to perform energy curve backtracking processing based on the target curve deviation and the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value in multiple time-frequency order energy information sets, so as to obtain the target time-frequency energy characterization curve. The curve extraction unit 806 is used to perform approximate inversion processing based on the target time-frequency energy characterization curve to obtain the target micro-Doppler curve of the sea surface target.

[0087] The specific implementation of this micro-Doppler curve extraction device is basically the same as the specific embodiment of the micro-Doppler curve extraction method described above, and will not be repeated here.

[0088] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the micro-Doppler curve extraction method described above. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0089] Please see Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 901 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 902 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 902 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901 using the micro-Doppler curve extraction method of the embodiments of this application. The 903 input / output interface is used to implement information input and output. The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904); The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0090] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described micro-Doppler curve extraction method.

[0091] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0092] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0093] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0094] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0095] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0096] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0097] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "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 three cases: only A exists, only B exists, and both A and B exist simultaneously, 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. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0098] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0099] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0100] Furthermore, the functional units in the various embodiments of this application 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.

[0101] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0102] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for extracting micro-Doppler curves, characterized in that, The method includes: Transmit target radar detection signals; In response to the signal reflection formed by the target radar detection signal illuminating the sea surface target, the corresponding target radar echo signal is acquired; A window fractional Fourier transform is performed on the target radar echo signal to obtain a three-dimensional time-frequency order information tensor. The three-dimensional time-frequency order information tensor includes multiple time-frequency order energy information sets corresponding to multiple target times. The time-frequency order energy information sets include multiple time-frequency order energy magnitudes and corresponding frequency domain coordinate information and order coordinate information. For each target time, Viterbi calculation is performed on the time-frequency order energy information set corresponding to the target time according to the three-dimensional time-frequency order information tensor to obtain the target curve deviation and optimal preceding energy magnitude corresponding to each time-frequency order energy magnitude in the time-frequency order energy information set; Based on the target curve deviation and the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value in the multiple time-frequency order energy information sets, energy curve backtracking is performed to obtain the target time-frequency energy characterization curve; Based on the target time-frequency energy characterization curve, an approximate inversion process is performed to obtain the target micro-Doppler curve of the sea surface target.

2. The method according to claim 1, characterized in that, For each target time, the Viterbi calculation is performed on the time-frequency order energy information set corresponding to the target time based on the three-dimensional time-frequency order information tensor to obtain the target curve deviation and optimal preceding energy magnitude corresponding to each time-frequency order energy magnitude in the time-frequency order energy information set, including: For each target time, the time-frequency order energy information set corresponding to the adjacent time before the target time is determined in the three-dimensional time-frequency order information tensor; For each time-frequency order energy magnitude value corresponding to each target time, curve deviation analysis is performed based on the time-frequency order energy magnitude value, the corresponding frequency domain coordinate information and order coordinate information, and the time-frequency order energy information set corresponding to the adjacent time, to obtain multiple candidate curve deviations corresponding to the time-frequency order energy magnitude value. The candidate curve deviation with the smallest deviation among the multiple candidate curve deviations is determined as the target curve deviation corresponding to the time-frequency order energy modulus. In the time-frequency order energy information set corresponding to the adjacent time, the time-frequency order energy modulus associated with the target curve deviation is determined as the optimal preceding energy modulus.

3. The method according to claim 2, characterized in that, For each time-frequency order energy magnitude value corresponding to each target time, a curve deviation analysis is performed based on the time-frequency order energy magnitude value, the corresponding frequency domain coordinate information and order coordinate information, and the time-frequency order energy information set corresponding to the adjacent time moments, to obtain multiple candidate curve deviations corresponding to the time-frequency order energy magnitude value, including: For each time-frequency order energy magnitude value corresponding to each target time, an energy deviation analysis is performed based on the time-frequency order energy magnitude value and the corresponding frequency domain coordinate information and order coordinate information to obtain the energy deviation value corresponding to the time-frequency order energy magnitude value. Based on the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy modulus and the time-frequency order energy information set corresponding to the adjacent time moments, a curve continuity deviation analysis is performed to obtain multiple curve continuity deviations corresponding to the time-frequency order energy modulus. Based on the energy deviation corresponding to the time-frequency order energy modulus, energy correction is performed on each of the multiple curve continuous deviations to obtain multiple candidate curve deviations corresponding to the time-frequency order energy modulus.

4. The method according to claim 3, characterized in that, The step of performing energy deviation analysis based on the time-frequency order energy magnitude value and the corresponding frequency domain coordinate information and order coordinate information to obtain the energy deviation value corresponding to the time-frequency order energy magnitude value includes: Based on the time-frequency order energy modulus and multiple time-frequency order energy modulus values ​​in the time-frequency order energy information set corresponding to the target time, an energy deviation amplitude analysis is performed to obtain the energy deviation amplitude of the time-frequency order energy modulus. Based on the time-frequency order energy modulus value, the corresponding frequency domain coordinate information and order coordinate information, and multiple time-frequency order energy modulus values ​​in the time-frequency order energy information set corresponding to the target time, an energy concentration deviation analysis is performed to obtain the energy concentration deviation value corresponding to the time-frequency order energy modulus value. The energy deviation degree corresponding to the time-frequency order energy modulus is determined based on the energy deviation amplitude and the energy concentration deviation degree.

5. The method according to claim 4, characterized in that, The step of performing energy correction on each of the multiple continuous curve deviations based on the energy deviation corresponding to the time-frequency order energy magnitude value to obtain multiple candidate curve deviations corresponding to the time-frequency order energy magnitude value includes: Determine the optimal preceding energy information set corresponding to the adjacent time points. The optimal preceding energy information set includes the optimal preceding energy modulus sequence associated with each time-frequency order energy modulus in the time-frequency order energy information set corresponding to the adjacent time points. Based on the frequency domain coordinate information and order coordinate information corresponding to the time-frequency order energy modulus, the time-frequency order energy information set corresponding to the adjacent time moments and the optimal preceding energy information set, a smoothness deviation analysis is performed to obtain multiple smoothness deviations corresponding to the time-frequency order energy modulus. Based on the energy deviation corresponding to the time-frequency order energy modulus, a comprehensive correction is made to each continuous deviation of the multiple curves and each smooth deviation of the multiple smooth deviations to obtain multiple candidate curve deviations corresponding to the time-frequency order energy modulus.

6. The method according to claim 1, characterized in that, The step of performing energy curve backtracking processing based on the target curve deviation and optimal preceding energy modulus corresponding to each time-frequency order energy modulus value in multiple time-frequency order energy information sets, to obtain the target time-frequency energy characterization curve, includes: Determine the last time among multiple target times, and based on the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value in multiple time-frequency order energy information sets, start from each time-frequency order energy modulus value in the time-frequency order energy information set corresponding to the last time, perform energy curve backtracking to obtain multiple candidate time-frequency energy characterization curves; The total deviation of the candidate time-frequency energy characterization curve is determined based on the deviation of the target curve corresponding to the multiple time-frequency order energy modulus values ​​included in each candidate time-frequency energy characterization curve. The candidate time-frequency energy characterization curve corresponding to the curve with the smallest total deviation among the multiple curves is determined as the target time-frequency energy characterization curve.

7. The method according to claim 1, characterized in that, After performing an approximate inversion process based on the target's time-frequency energy characterization curve to obtain the target micro-Doppler curve of the sea surface target, the process includes: The three-dimensional time-frequency order information tensor is filtered based on the multiple time-frequency order energy magnitudes included in the target time-frequency energy characterization curve, as well as the corresponding frequency domain coordinate information and order coordinate information, to obtain the filtered three-dimensional time-frequency order information tensor. Based on the filtered three-dimensional time-frequency order information tensor, the micro-Doppler curve is extracted again to obtain the filtered micro-Doppler curve.

8. A micro-Doppler curve extraction device, characterized in that, The device includes: Signal transmitting unit, used to transmit target radar detection signals; The signal acquisition unit is used to acquire the corresponding target radar echo signal in response to the signal reflection formed by the target radar detection signal illuminating the sea surface target. The Fourier transform unit is used to perform a window fractional Fourier transform on the target radar echo signal to obtain a three-dimensional time-frequency order information tensor. The three-dimensional time-frequency order information tensor includes multiple time-frequency order energy information sets corresponding to multiple target times. The time-frequency order energy information sets include multiple time-frequency order energy magnitudes and corresponding frequency domain coordinate information and order coordinate information. The Viterbi calculation unit is used to perform Viterbi calculation processing on the time-frequency order energy information set corresponding to the target time according to the three-dimensional time-frequency order information tensor for each target time, so as to obtain the target curve deviation and the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value in the time-frequency order energy information set. The curve backtracking unit is used to perform energy curve backtracking processing based on the target curve deviation and the optimal preceding energy modulus value corresponding to each time-frequency order energy modulus value in the multiple time-frequency order energy information sets, so as to obtain the target time-frequency energy characterization curve. The curve extraction unit is used to perform approximate inversion processing based on the target time-frequency energy characterization curve to obtain the target micro-Doppler curve of the sea surface target.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the micro-Doppler curve extraction method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the micro-Doppler curve extraction method according to any one of claims 1 to 7.

Images