Underwater target identification method based on single vector hydrophone

Abstract

The invention discloses an underwater target identification method based on a single-vector hydrophone, and belongs to the field of underwater target identification methods of single-vector hydrophones. An existing single-vector hydrophone can only measure the arrival angle of an underwater target and cannot measure other key parameters such as distance and speed, and application scenes are limited. And the performance of the single-vector hydrophone is greatly influenced by the environment. The method comprises the following steps: receiving a micro-motion Doppler signal through a hydrophone, and carrying out signal processing operations of filtering, feature extraction and target classification on the signal through a QT image processing library; micro-Doppler feature extraction is carried out based on empirical mode decomposition and a CLEAN algorithm; and combining the asynchronous transmission characteristics of the FIR filter pre-CUDA, designing an FIR filter parallel algorithm, and completing task execution of a serial program and a parallel program by adopting lake test data. The method not only has the combination of accurate analysis, real-time performance and high precision of micro-dynamic characteristics, but also has the characteristics of target classification and identification capability, anti-interference capability, robustness, expansibility, adaptability and the like.

Description

Technical Field

[0001] The invention relates to an underwater target recognition method, in particular to an underwater target recognition method based on a single vector hydrophone. Background Art

[0002] As the exploration of the ocean continues to deepen, sonar equipment has developed rapidly, and the topic of ocean target detection has received increasing attention in the scientific field. How to reduce the radiation noise of submarines and surface ships and enhance their concealment has become a top priority for countries around the world to research and develop submarines. In recent years, breakthrough progress has been made in submarine vibration reduction and noise reduction technology, which has greatly improved the concealment performance of submarines, causing great trouble for ocean target detection. Submarine and other underwater targets will contain rich low-frequency line spectrum information in their radiated noise spectrum due to the periodic motion of their own propellers. The sound-absorbing materials used by submarines in most countries in the world are anechoic tiles, which have certain limitations in the frequency range of sound absorption and can only absorb noise with higher frequency bands. However, for noise with lower frequency bands, especially for very low frequency noise below several hundred Hz, even if submarines use anechoic technology, their radiated noise spectrum still contains strong low-frequency line spectrum. The sound field vector information measured by vector hydrophone can be used to find these line spectra and calculate the azimuth of these target line spectra through detection algorithms. In addition, the traditional scalar hydrophone cannot collect complete underwater sound field data information, but the vector hydrophone can not only fill this gap, but also has the characteristic that the directivity is not limited by frequency. The vector hydrophone can not only use the combination of sound pressure and particle velocity, and select appropriate vector signal processing methods to achieve the purpose of target detection or parameter estimation, but also form an array to obtain the spatial gain of the array, which makes the vector sensor very popular in the target azimuth estimation of low-frequency targets.

[0003] The electromagnetic characteristic control technology for targets makes it increasingly difficult to capture the non-motion characteristics of space targets, which has a significant impact on the detection and identification of space targets by radars, and has produced many means of deceiving radars, such as controlling the shape, structure and material of space targets to achieve the "stealth" effect. However, due to the constraints of various practical factors, the motion characteristics of targets, especially the interactive motion characteristics, such as the vibration, tumbling and rotation of some parts of the target, are less controllable by the outside world. Therefore, the position of target motion and even micro-motion characteristics in the classification and identification of space targets is becoming increasingly prominent. How to accurately classify and identify different underwater targets has become the focus of current research. With the discovery of the micro-Doppler phenomenon of radar targets, the micro-motion characteristics of targets have gradually received widespread attention in radar automatic target recognition. Generally, small vibrations, rotations and other high-order movements of radar targets other than the center of mass translation are collectively referred to as micro-motion. The echo of the micro-motion target contains fine target micro-Doppler characteristic information, so the target-specific characteristic information can be extracted from it. The characteristics based on the target micro-motion echo time-frequency diagram have become an effective target classification feature because of its large amount of information. Experiments have proved that it is effective and feasible to classify and identify targets using micro-Doppler characteristics.

[0004] Underwater target recognition based on single vector hydrophone is a challenging and cutting-edge research. The micro-Doppler effect refers to the change in the frequency of sound waves when the sound source or receiver makes a small movement relative to the target in the water. This change is called micro-Doppler shift. Based on this effect, the sound signal received by the single vector hydrophone can be used to identify underwater targets.

[0005] Traditional underwater target detection cannot accurately determine the target position and speed: Micro-Doppler technology can determine the target's motion state and speed by analyzing the sound wave signals emitted by underwater targets, thereby achieving accurate detection and tracking. If this technology is not used, it may be necessary to rely on other detection methods, but their accuracy and effect may not be as good as micro-Doppler technology. Limited detection distance: In the absence of micro-Doppler technology, it may be necessary to rely on other sonar or detection equipment for underwater target detection. However, the detection range of these devices is usually shorter than that of micro-Doppler technology, so it may not be able to detect distant targets. Easy to be identified and interfered by the target: In addition to micro-Doppler technology, there are some other underwater target detection technologies, such as laser, electromagnetic, etc., but these technologies may be identified or interfered by the target. In contrast, micro-Doppler technology has high concealment and anti-interference, so it is widely used in underwater target detection. Micro-Doppler technology is an advanced underwater target detection technology with the advantages of high accuracy, long distance and anti-interference.

[0006] In summary, although the prior art has achieved many results in underwater target recognition using single vector hydrophone, the following problems still exist:

[0007] 1. A single vector hydrophone can only measure the arrival angle of underwater targets, but cannot measure other key parameters such as distance and speed. Therefore, its accuracy is greatly limited and its application scenarios are restricted.

[0008] 2. The performance of single-vector hydrophones is greatly affected by the environment. For example, multiple reflections and refractions in underwater channels can cause signal distortion and interference, which can affect the identification and positioning of targets. In addition, since the detection range of single-vector hydrophones is limited and they cannot accurately locate targets, they are easily interfered by other sound sources or noise, which can lead to problems such as misjudgment and missed judgment. Summary of the invention

[0009] The purpose of the present invention is to solve the problem that the existing single vector hydrophone can only measure the arrival angle of underwater targets but cannot measure other key parameters such as distance and speed, and the application scenarios are limited; and the performance of the single vector hydrophone is greatly affected by the environment, and propose an underwater target recognition method based on a single vector hydrophone.

[0010] The above purpose is achieved through the following technical solutions:

[0011] A method for underwater target recognition based on a single vector hydrophone is implemented by the following steps:

[0012] Step 1: Receive the micro-motion Doppler signal through a hydrophone, and perform signal processing operations such as filtering, feature extraction, and target classification on the signal through a QT image processing library;

[0013] Step 2: Extract micro-Doppler features based on empirical mode decomposition and CLEAN algorithm; the full name of empirical mode decomposition is Empirical Mode Decomposition, abbreviated as EMD;

[0014] Firstly, the EMD algorithm is used to decompose the underwater target echo. Then, based on the EMD decomposition results, the CLEAN method is used to analyze the underwater target echo, and the fuselage component and micro-motion component in the underwater target echo are completely separated. Then, based on the EMD and CLEAN decomposition results, the five-dimensional features reflecting the differences in micro-motion characteristics of different targets and the proportional features reflecting the fuselage component and micro-motion component of different targets are extracted to realize the classification of underwater propeller targets.

[0015] Step 3: Combine the FIR filter with CUDA asynchronous transmission characteristics, design the FIR filter parallel algorithm, and use the lake test data to complete the task execution of the serial program and the parallel program;

[0016] Step 4: Based on the micro-Doppler characteristics, an underwater target recognition algorithm is designed to display the target's motion trajectory, speed, direction and other information in a graphical way.

[0017] Furthermore, the signal processing operation of filtering, feature extraction, and target classification of the signal by the QT image processing library in step 1 is specifically:

[0018] For micro-Doppler signal data saved in the form of files or data streams directly transmitted through a network interface, and the data or data streams are in digital format, code is written in Qt to read these data or data streams, process and analyze them, and extract micro-Doppler features from the pre-processed echo data.

[0019] Furthermore, the step of extracting micro-Doppler features based on empirical mode decomposition and CLEAN algorithm described in step 2 is specifically:

[0020] In sonar target detection, the radial relative motion between the target and the sonar transmitter will cause the frequency of the echo signal to change, that is, the carrier frequency of the target echo signal will undergo a Doppler shift, and the Doppler shift satisfies the expression:

[0021]

[0022] In the formula, ν is the radial velocity of the target relative to the hydrophone, λ is the carrier wavelength, and the micro-Doppler effect obtains the frequency corresponding to each time point by taking the time derivative of the phase of the echo signal, thereby obtaining the micro-Doppler characteristics contained in the signal;

[0023] In the process of establishing the target model, when the physical size of the target is much larger than the sonar working wavelength, the target is regarded as composed of multiple scattering points, so the sonar echo of the target is formed by the addition of the electromagnetic scattering vectors of all scattering points; taking the scattering point target as the research object, the vibration and rotation micro-Doppler mathematical model is established, and the micro-Doppler characteristics of typical micro-motion are analyzed;

[0024] Assume that the narrowband coherent signal emitted by the sonar is:

[0025] x(t)=exp(j2πf c t)

[0026] Among them, f c is the hydrophone operating frequency;

[0027] The sonar echo signal of a single scattering point is:

[0028]

[0029] Where a(t) is the instantaneous amplitude, r(t) is the distance from the scattering point to the hydrophone, c is the speed of light, is the instantaneous phase. When the submarine target body component and various noise components are not considered, the complex envelope of the synthetic echo of K scattering centers is expressed as:

[0030]

[0031] Among them, a n (t) and denote the amplitude and phase of the nth scattering center, r n (t) represents the distance between the nth scattering center and the sonar, and λ represents the wavelength of the sonar emission signal;

[0032] The echo model of the propeller micro-motion component is derived, with the sonar line of sight coordinate system O as the origin of the coordinate system, and the rotation coordinate system (x, y, z) with the blade rotation center O as the origin of the coordinate system; the motion of any point on the rotor blade is decomposed into the synthesis of the translational motion with the submarine's navigation and the rotational motion with the blade rotation, that is, the synthesis of the movement of point O from translation to point O' and the rotation of point P to point P'; at the same time, let the radial navigation speed of the submarine be v, the angle between the navigation direction and the line of sight of the hydrophone be φ, the azimuth angle of the rotor be α, the pitch angle be β, the height of the rotor rotation center O be h, and the angular velocity of the rotor be ω r ;

[0033] Each blade of the propeller rotor is regarded as an equivalent scattering center; suppose there is a submarine target in the far field of the hydrophone, its propeller rotor contains N blades, and each blade of the rotor is a linear rigid antenna of the same type. The distance OP=l from any scattering point P on the blade to the rotation center O includes the effective blade length L=L 2 -L 1 , L 2 is the distance from the blade tip to the rotation center, L 1 is the distance from the blade root to the rotation center. For a propeller, the blade root does not coincide with the rotation center, that is, L 1 ≠0; at time t, the phase angle θ of the blade rotation t =θ 0 +ω r t, where θ 0 is the initial phase angle of the blade; the distance R from the blade rotation center O to the hydrophone t =R 0 +νtcosφ, where R 0 is the initial distance between point O and the hydrophone; the coordinate of point O at time t in the hydrophone line of sight coordinate system is (R t cosβcosα,R t cosβsinα,R tsinβ), the coordinates of point P in the rotating coordinate system are (lcosθ t ,lsinθ t ,0); the distance from any point P on the blade to the hydrophone at time t is:

[0034]

[0035] According to the above formula, the phase of the scattering point P at time t is:

[0036]

[0037] Then the normalized modulated echo of the scattering point P at time t is:

[0038]

[0039] Among them, f d =2νcosφ / λ is the Doppler frequency of the fuselage; the sonar echo of the entire blade is regarded as the linear integration of the echoes of all scattering points on the blade, thus obtaining the echo expression of a single blade:

[0040]

[0041] For a rotor with N blades, let the initial phase angle of each blade be θ k =θ 0 +2πk / N, k=0,1,...,N-1, then the scattered echo synthesized by all N blades can be expressed as follows:

[0042]

[0043] By estimating f from the echo d Compensate for the influence of the Doppler frequency of the fuselage; at the same time, similarly use exp(-j2πf c t) Compensation hydrophone carrier frequency f c After the influence of , the above formula is written as:

[0044]

[0045] The above formula is the baseband form of the propeller space micro-motion component modulated echo; where a k (t) represents the amplitude component, Represents the phase component; From the phase of the scattering point P at time t, it can be seen that the amplitude and phase components of the baseband part of the spatial micro-motion component modulated echo are affected by the rotation angular velocity ω r The amplitude of the periodic modulation is determined by a series of r The relevant sinc functions are superimposed modulo;

[0046] The above explicit expression of time domain echo indicates that the complex envelope of the rotor model echo should not include the rotor structure parameters N, L, blade rotation parameters ω r and the hydrophone parameter λ. The number of propeller blades and the speed of the rotor blades of modern submarines are fixed on the same type of submarines. At the same time, the speed of the rotor blades is also constant when the submarine is sailing normally and stably. Therefore, for a fixed type of submarine and hydrophone, when they are working in normal state, N, L, ω in the model r , λ are all constants, and the form of the modulated echo depends only on the pitch parameter β and the rotor parameters N, ω r related.

[0047] The beneficial effects of the present invention are:

[0048] The present invention is based on underwater target identification. By analyzing the characteristics of the hydrophone receiving signal and the characteristics of the micro-Doppler signal, the micro-Doppler technology is combined with the single vector hydrophone to achieve more accurate identification and tracking of underwater targets. The cross-platform and friendly graphical interface design of the Qt framework can also be used to implement the target azimuth detection algorithm based on the micro-Doppler principle. Through the powerful functional extensibility and efficient development efficiency of Qt, real-time processing of high-density data can be achieved with faster execution speed and lower computing delay.

[0049] The underwater target recognition method based on a single vector hydrophone of the present invention includes the analysis of micro-dynamic features, the application of signal processing technology in complex underwater environments, and the combination with deep learning. It not only has the accurate analysis of micro-dynamic features, the combination of real-time and high precision, but also has the characteristics of target classification and recognition ability, anti-interference ability and robustness, as well as scalability and adaptability, which brings important progress and innovation to the development of underwater target recognition and tracking technology. Specifically:

[0050] 1. Micro-Doppler technology can estimate the speed, direction and acceleration of underwater targets by analyzing the changes in the spectrum. However, single-vector hydrophones can only provide the location information of the target and cannot directly obtain the motion parameters of the target.

[0051] 2. Qt provides a powerful graphical interface library, including various predefined controls, layout managers and style sheets, and provides rich 2D and 3D graphics drawing functions, including drawing curves, scatter plots, image display and model rendering, etc. This is very useful for data visualization and result display of underwater targets, more intuitive understanding and analysis of recognition results, and can also be easily integrated with other commonly used libraries and tools. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in combination with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Specific implementation method one:

[0054] This embodiment provides an underwater target recognition method based on a single vector hydrophone, and the method is implemented by the following steps:

[0055] Step 1: Receive the micro-motion Doppler signal through a hydrophone, and perform signal processing operations such as filtering, feature extraction, and target classification on the signal through a QT image processing library;

[0056] Step 2: Extract micro-Doppler features based on empirical mode decomposition and CLEAN algorithm; the full name of empirical mode decomposition is Empirical Mode Decomposition, abbreviated as EMD;

[0057] Firstly, the EMD algorithm is used to decompose the underwater target echo. Then, based on the EMD decomposition results, the CLEAN method is used to analyze the underwater target echo, and the fuselage component and micro-motion component in the underwater target echo are completely separated. Then, based on the EMD and CLEAN decomposition results, five-dimensional features reflecting the differences in micro-motion characteristics of different targets and the proportional features reflecting the fuselage component and micro-motion component of different targets are extracted to achieve the classification of underwater propeller targets. The experimental results based on simulation and measured data show that the feature extraction method based on EMD-CLEAN can not only obtain better classification performance, but also has lower requirements on sonar system parameters.

[0058] Step 3: Combine the FIR filter with the CUDA asynchronous transmission characteristics to design a FIR filter parallel algorithm to analyze and compare the serial algorithm and parallel algorithm for target detection, and use the lake test data to complete the task execution of the serial program and the parallel program;

[0059] Step 4: Based on the micro-Doppler characteristics, an underwater target recognition algorithm is designed to display the recognition results to the user in a graphical way, showing the target's motion trajectory, speed, direction and other information. Specific implementation method 2:

[0061] Different from the first specific implementation mode, the underwater target recognition method based on a single vector hydrophone in this implementation mode is:

[0062] The signal processing operation steps of filtering, feature extraction, and target classification of the signal by the QT image processing library described in step 1 are specifically:

[0063] For micro-Doppler signal data saved in the form of files or data streams transmitted directly through a network interface, and the data or data streams are in digital format, there is no need to receive these signals through a hardware interface. In this case, code is written in Qt to read these data or data streams, process and analyze them, and extract micro-Doppler features from the pre-processed echo data. Specific implementation method three:

[0065] Different from the second specific implementation mode, in the underwater target recognition method based on a single vector hydrophone in this implementation mode, the step of extracting micro-Doppler features based on empirical mode decomposition and CLEAN algorithm described in step 2 is specifically:

[0066] In sonar target detection, the radial relative motion between the target and the sonar transmitter will cause the frequency of the echo signal to change, that is, the carrier frequency of the target echo signal will undergo a Doppler shift, and the Doppler shift satisfies the expression:

[0067]

[0068] In the formula, ν is the radial velocity of the target relative to the hydrophone, λ is the carrier wavelength, and the micro-Doppler effect obtains the frequency corresponding to each time point by taking the time derivative of the phase of the echo signal, thereby obtaining the micro-Doppler characteristics contained in the signal;

[0069] In the process of establishing the target model, when the physical size of the target is much larger than the sonar working wavelength, the target is regarded as composed of multiple scattering points, so the sonar echo of the target is formed by the addition of the electromagnetic scattering vectors of all scattering points; taking the scattering point target as the research object, the vibration and rotation micro-Doppler mathematical model is established, and the micro-Doppler characteristics of typical micro-motion are analyzed;

[0070] Assume that the narrowband coherent signal emitted by the sonar is:

[0071] x(t)=exp(j2πf c t)

[0072] Among them, f c is the hydrophone operating frequency;

[0073] The sonar echo signal of a single scattering point is:

[0074]

[0075] Where a(t) is the instantaneous amplitude, r(t) is the distance from the scattering point to the hydrophone, c is the speed of light, is the instantaneous phase. When the submarine target body component and various noise components are not considered, the complex envelope of the synthetic echo of K scattering centers is expressed as:

[0076]

[0077] Among them, a n (t) and denote the amplitude and phase of the nth scattering center, r n (t) represents the distance between the nth scattering center and the sonar, and λ represents the wavelength of the sonar emission signal;

[0078] In order to more intuitively derive the echo model of the propeller micro-motion components, a spatial coordinate system is established. The sonar line of sight coordinate system O is taken as the origin of the coordinates, and the rotation coordinate system (x, y, z) is taken as the origin of the coordinates of the blade rotation center O; the motion of any point on the rotor blade is decomposed into the synthesis of the two motions of translation with the submarine's navigation and rotation with the blade rotation, that is, the synthesis of the two motions of translation from point O to point O' and rotation from point P to point P'; at the same time, let the radial navigation speed of the submarine be v, the angle between the navigation direction and the line of sight of the hydrophone be φ, the azimuth angle of the rotor be α, the pitch angle be β, the height of the rotor rotation center O be h, and the angular velocity of the rotor be ω r .

[0079] Each blade of the propeller rotor is still in the sonar optical zone, and each blade is regarded as an equivalent scattering center. Suppose there is a submarine target in the far field of the hydrophone, and its propeller rotor contains N blades, and each blade of the rotor is a linear rigid antenna of the same type, and there is no blade shielding. The distance OP = l from any scattering point P on the blade to the rotation center O includes the effective blade length L = L 2 -L 1 , L 2 is the distance from the blade tip to the rotation center, L 1 is the distance from the blade root to the rotation center. For a propeller, the blade root does not coincide with the rotation center, that is, L 1 ≠0; at time t, the phase angle θ of the blade rotation t =θ 0 +ω r t, where θ 0 is the initial phase angle of the blade; the distance R from the blade rotation center O to the hydrophone t =R 0 +νtcosφ, where R 0 is the initial distance between point O and the hydrophone; the coordinate of point O at time t in the hydrophone line of sight coordinate system is (R t cosβcosα,R t cosβsinα,R tsinβ), the coordinates of point P in the rotating coordinate system are (lcosθ t ,lsinθ t ,0); the distance from any point P on the blade to the hydrophone at time t is:

[0080]

[0081] According to the above formula, the phase of the scattering point P at time t is:

[0082]

[0083] Then the normalized modulated echo of the scattering point P at time t is:

[0084]

[0085] Among them, f d =2νcosφ / λ is the Doppler frequency of the fuselage; the sonar echo of the entire blade is regarded as the linear integration of the echoes of all scattering points on the blade, thus obtaining the echo expression of a single blade:

[0086]

[0087] For a rotor with N blades, let the initial phase angle of each blade be θ k =θ 0 +2πk / N, k=0,1,...,N-1, then the scattered echo synthesized by all N blades can be expressed as follows:

[0088]

[0089] By estimating f from the echo d Compensate for the influence of the Doppler frequency of the fuselage; at the same time, similarly use exp(-j2πf c t) Compensation hydrophone carrier frequency f c After the influence of , the above formula is written as:

[0090]

[0091] The above formula is the baseband form of the propeller space micro-motion component modulated echo; where a k (t) represents the amplitude component, Represents the phase component; From the phase of the scattering point P at time t, it can be seen that the amplitude and phase components of the baseband part of the spatial micro-motion component modulated echo are affected by the rotation angular velocity ω r The amplitude of the periodic modulation is determined by a series of r The relevant sinc functions are superimposed modulo;

[0092] The above modeling derivation describes the explicit expression of the time domain echo generated by the modulation of the submarine target space micro-motion components, indicating that the complex envelope of the rotor model echo should not include the rotor structure parameters N, L, blade rotation parameters ω r and the hydrophone parameter λ, and has nothing to do with the submarine's speed, diving depth, distance from the hydrophone, sailing time and other parameters; At present, the propeller rotors of modern submarines use the best design based on the principle of hydroacoustic dynamics, and the number of blades and rotation speed are fixed on the same type of submarines. At the same time, the rotation speed of the rotor is also constant when the submarine is sailing normally and stably; Therefore, for a fixed type of submarine and hydrophone, when they are all working under normal conditions, N, L, ω in the model r , λ are all constants, and the form of the modulated echo depends only on the pitch parameter β and the rotor parameters N, ω r This indicates the relative invariance of the echo model, which provides the possibility of successful classification and identification.

[0093] The embodiments of the present invention disclose preferred embodiments, but are not limited thereto. A person skilled in the art can easily understand the spirit of the present invention based on the above embodiments and make different extensions and changes. However, as long as they do not deviate from the spirit of the present invention, they are all within the protection scope of the present invention.

Claims

1. An underwater target recognition method based on a single vector hydrophone, characterized in that: The method is implemented by the following steps: Step 1: Receive the micro-motion Doppler signal through a hydrophone, and perform signal processing operations such as filtering, feature extraction, and target classification on the signal through a QT image processing library; Step 2: Extract micro-Doppler features based on empirical mode decomposition and CLEAN algorithm; the full name of empirical mode decomposition is Empirical Mode Decomposition, abbreviated as EMD; Firstly, the EMD algorithm is used to decompose the underwater target echo. Then, based on the EMD decomposition results, the CLEAN method is used to analyze the underwater target echo, and the fuselage component and micro-motion component in the underwater target echo are completely separated. Then, based on the EMD and CLEAN decomposition results, the five-dimensional features reflecting the differences in micro-motion characteristics of different targets and the proportional features reflecting the fuselage component and micro-motion component of different targets are extracted to realize the classification of underwater propeller targets. Step 3: Combine the FIR filter with CUDA asynchronous transmission characteristics, design the FIR filter parallel algorithm, and use the lake test data to complete the task execution of the serial program and the parallel program; Step 4: Based on the micro-Doppler characteristics, an underwater target recognition algorithm is designed to display the target's motion trajectory, speed, direction and other information in a graphical way.

2. The underwater target recognition method based on a single vector hydrophone according to claim 1, characterized in that: The signal processing operation steps of filtering, feature extraction, and target classification of the signal by the QT image processing library described in step 1 are specifically: For micro-Doppler signal data saved in the form of files or data streams directly transmitted through a network interface, and the data or data streams are in digital format, code is written in Qt to read these data or data streams, process and analyze them, and extract micro-Doppler features from the pre-processed echo data.

3. The underwater target recognition method based on single vector hydrophone according to claim 2 is characterized in that: The steps of extracting micro-Doppler features based on empirical mode decomposition and CLEAN algorithm described in step 2 are specifically: In sonar target detection, the radial relative motion between the target and the sonar transmitter will cause the frequency of the echo signal to change, that is, the carrier frequency of the target echo signal will undergo a Doppler shift, and the Doppler shift satisfies the expression: In the formula, ν is the radial velocity of the target relative to the hydrophone, λ is the carrier wavelength, and the micro-Doppler effect obtains the frequency corresponding to each time point by taking the time derivative of the phase of the echo signal, thereby obtaining the micro-Doppler characteristics contained in the signal; In the process of establishing the target model, when the physical size of the target is much larger than the sonar working wavelength, the target is regarded as composed of multiple scattering points, so the sonar echo of the target is formed by the addition of the electromagnetic scattering vectors of all scattering points; taking the scattering point target as the research object, the vibration and rotation micro-Doppler mathematical model is established, and the micro-Doppler characteristics of typical micro-motion are analyzed; Assume that the narrowband coherent signal emitted by the sonar is: x(t)=exp(j2πf c t) Among them, f c is the hydrophone operating frequency; The sonar echo signal of a single scattering point is: Where a(t) is the instantaneous amplitude, r(t) is the distance from the scattering point to the hydrophone, c is the speed of light, is the instantaneous phase. When the submarine target body component and various noise components are not considered, the complex envelope of the synthetic echo of K scattering centers is expressed as: Among them, a n (t) and denote the amplitude and phase of the nth scattering center, r n (t) represents the distance between the nth scattering center and the sonar, and λ represents the wavelength of the sonar emission signal; The echo model of the propeller micro-motion component is derived, with the sonar line of sight coordinate system O as the origin of the coordinate system, and the rotation coordinate system (x, y, z) with the blade rotation center O as the origin of the coordinate system; the motion of any point on the rotor blade is decomposed into the synthesis of the translational motion with the submarine's navigation and the rotational motion with the blade rotation, that is, the synthesis of the movement of point O from translation to point O' and the rotation of point P to point P'; at the same time, let the radial navigation speed of the submarine be v, the angle between the navigation direction and the line of sight of the hydrophone be φ, the azimuth angle of the rotor be α, the pitch angle be β, the height of the rotor rotation center O be h, and the angular velocity of the rotor be ω r ; Each blade of the propeller rotor is regarded as an equivalent scattering center; suppose there is a submarine target in the far field of the hydrophone, and its propeller rotor contains N blades, and each blade of the rotor is a linear rigid antenna of the same type. The distance OP=l from any scattering point P on the blade to the rotation center O includes the effective blade length L=L2-L1, L2 is the distance from the blade end to the rotation center, and L1 is the distance from the blade root to the rotation center. For the propeller, the blade root does not coincide with the rotation center, that is, L1≠0; at time t, the phase angle θ of the blade rotation t =θ0+ω r t, where θ0 is the initial phase angle of the blade; the distance from the blade rotation center O to the hydrophone R t =R0+νtcosφ, where R0 is the initial distance between point O and the hydrophone; the coordinate of point O at time t in the hydrophone line of sight coordinate system is (R t cosβcosα,R t cosβsinα,R t sinβ), the coordinates of point P in the rotating coordinate system are (lcosθ t ,lsinθ t ,0); the distance from any point P on the blade to the hydrophone at time t is: According to the above formula, the phase of the scattering point P at time t is: Then the normalized modulated echo of the scattering point P at time t is: Among them, f d =2νcosφ / λ is the Doppler frequency of the fuselage; the sonar echo of the entire blade is regarded as the linear integration of the echoes of all scattering points on the blade, thus obtaining the echo expression of a single blade: For a rotor with N blades, let the initial phase angle of each blade be θ k =θ0+2πk / N, k=0,1,…,N-1, then the scattered echo synthesized by all N blades can be expressed as follows: By estimating f from the echo d Compensate for the influence of the fuselage Doppler frequency; at the same time, similarly use exp(-j2πfct) to compensate for the hydrophone carrier frequency f c After the influence of , the above formula is written as: The above formula is the baseband form of the propeller space micro-motion component modulated echo; where a k (t) represents the amplitude component, Represents the phase component; From the phase of the scattering point P at time t, it can be seen that the amplitude and phase components of the baseband part of the spatial micro-motion component modulated echo are affected by the rotation angular velocity ω r The amplitude of the periodic modulation is determined by a series of r The relevant sinc functions are superimposed modulo; The above explicit expression of time domain echo indicates that the complex envelope of the rotor model echo should not include the rotor structure parameters N, L, blade rotation parameters ω r and the hydrophone parameter λ. The number of propeller blades and the speed of the rotor blades of modern submarines are fixed on the same type of submarines. At the same time, the speed of the rotor blades is also constant when the submarine is sailing normally and stably. Therefore, for a fixed type of submarine and hydrophone, when they are working in normal state, N, L, ω in the model r , λ are all constants, and the form of the modulated echo depends only on the pitch parameter β and the rotor parameters N, ω r related.

Images