Clutter suppression method and device based on windowing mean filtering, equipment and medium

Abstract

The invention discloses a clutter suppression method, device and equipment based on windowing mean filtering and a medium, and relates to the technical field of radar signal processing, and the method comprises the steps: carrying out the pulse compression and coherent accumulation processing of an echo signal received by a radar, and obtaining an echo distance-Doppler matrix containing an actually measured clutter; the method comprises the following steps: establishing a distance-Doppler two-dimensional distribution model of theoretical clutters based on real-time inertial navigation data of an inertial navigation system and radar system parameters, and generating a theoretical clutter distribution curve; for each distance unit in the echo distance-Doppler matrix, determining a strip-shaped clutter suppression region by taking a theoretical value of the theoretical clutter distribution curve at the current distance unit as a center; and performing mean filtering replacement operation with a protection unit on each clutter unit in the clutter suppression region to obtain an echo distance-Doppler matrix after suppression processing, thereby solving the problems of complex engineering implementation, harsh application conditions, poor environmental adaptability and the like in the prior art.

Description

Technical Field

[0001] This invention relates to the field of radar signal processing technology, and specifically to a clutter suppression method, apparatus, device, and medium based on windowed mean filtering. Background Technology

[0002] During target detection using airborne radar, clutter has a significant impact on radar detection performance. High-speed radar platform movement causes severe broadening of the clutter spectrum, resulting in aliasing with the target's Doppler frequency shift. Simultaneously, the change in viewing angle introduced by the movement causes dramatic fluctuations in the spatial angular domain of clutter, forming a strong two-dimensional correlation in space and time. Clutter intensity from ground reflection is significantly higher than the target signal, and its scattering characteristics exhibit non-stationary statistical properties depending on terrain undulations, vegetation cover, and soil moisture; the radar cross-section differs greatly between mountainous and urban areas. Platform movement also induces range ambiguity and Doppler ambiguity, causing clutter from different range loops to alias in the time and frequency domains, exacerbating spectral pollution, especially when pulse repetition frequency design is limited.

[0003] Existing clutter suppression techniques mainly include Space-Time Adaptive Processing (STAP) and Offset Phase Center Antenna (DPCA). While STAP is theoretically optimal, it requires real-time calculation of the inverse of a high-dimensional matrix, resulting in extremely high computational complexity. Furthermore, its performance heavily relies on a large number of uniform statistical samples, leading to poor robustness in real-world non-uniform environments. DPCA's core principle requires the platform to maintain strictly uniform linear flight, making it extremely sensitive to aircraft maneuvers (such as turning, acceleration, and deceleration). Simultaneously, this technology places extremely stringent requirements on the amplitude and phase consistency across multiple receiving channels, making its engineering implementation difficult and costly.

[0004] In summary, existing mainstream airborne clutter suppression technologies, whether STAP or DPCA, all suffer from problems such as complex engineering implementation, stringent applicable conditions, and poor environmental adaptability. Ultimately, this leads to low target detection efficiency and high false alarm rate for airborne radar in strong clutter backgrounds. Summary of the Invention

[0005] The technical problem to be solved by the present invention is that existing airborne clutter suppression technologies have problems such as complex engineering implementation, strict applicable conditions, and poor environmental adaptability. The purpose is to provide a clutter suppression method, device, equipment and medium based on windowed mean filtering, which solves the above problems.

[0006] This invention is achieved through the following technical solution:

[0007] In a first aspect, the present invention provides a clutter suppression method based on windowed mean filtering, comprising:

[0008] The echo signal received by the radar is processed by pulse compression and coherent accumulation to obtain the echo range-Doppler matrix containing the measured clutter.

[0009] Based on real-time inertial navigation data and radar system parameters, a range-Doppler two-dimensional distribution model of theoretical clutter is established, and theoretical clutter distribution curves are generated.

[0010] For each range cell in the echo range-Doppler matrix, a strip-shaped clutter suppression region is determined centered on the theoretical value of the theoretical clutter distribution curve at the current range cell.

[0011] For each clutter unit within the clutter suppression region, a mean filter replacement operation with a guard unit is performed to obtain the echo distance-Doppler matrix after suppression processing.

[0012] Optionally, the method of establishing a range-Doppler two-dimensional distribution model of theoretical clutter based on real-time inertial navigation system data and radar system parameters, and generating theoretical clutter distribution curves, includes:

[0013] For each range cell in the echo range-Doppler matrix, based on the real-time inertial navigation data of the inertial navigation system and the radar system parameters, the theoretical clutter Doppler frequency of the current range cell is calculated using the radar Doppler principle; the inertial navigation data includes the aircraft speed and aircraft altitude; the radar system parameters include the radar wavelength, radar beam azimuth angle, and radar beam elevation angle.

[0014] The theoretical clutter Doppler frequency of each distance cell is mapped to the corresponding Doppler channel index in the echo distance-Doppler matrix, and all mapping points are connected to form the theoretical clutter distribution curve.

[0015] Optionally, for each range cell in the echo range-Doppler matrix, based on real-time inertial navigation data and radar system parameters, and using the radar Doppler principle, the theoretical clutter Doppler frequency of the current range cell is calculated, including:

[0016] For each range cell in the echo range-Doppler matrix, the slant range from the radar to the ground scattering point corresponding to the current range cell is calculated based on the aircraft altitude and the radar beam elevation angle.

[0017] Based on the slant range, the aircraft altitude, and the radar beam azimuth, calculate the angle between the vector of the aircraft velocity and the line-of-sight direction from the radar to the ground scattering point corresponding to the current range cell;

[0018] Based on the aircraft speed, the radar wavelength, and the included angle, the theoretical clutter Doppler frequency of the current range cell is calculated.

[0019] Optionally, the step of performing pulse compression and coherent accumulation processing on the radar-received echo signal to obtain the echo range-Doppler matrix containing measured clutter includes:

[0020] Using a reference signal designed with radar timing correlation parameters, pulse compression processing is performed on the echo signal received by the radar to obtain a pulse compressed signal;

[0021] Coherent accumulation processing is performed on multiple pulse compression signals within a coherent processing time to obtain an echo distance-Doppler matrix containing measured clutter.

[0022] Optionally, for each range cell in the echo range-Doppler matrix, determining a strip-shaped clutter suppression region centered on the theoretical value of the theoretical clutter distribution curve at the current range cell includes:

[0023] Divide the minimum relative velocity between the target and the carrier aircraft by the velocity resolution of a single Doppler channel to obtain a quotient.

[0024] The quotient is rounded up to obtain the number of Doppler channels that the clutter suppression region extends to each side of the theoretical clutter distribution curve, which is used as a preset number.

[0025] For each distance cell in the echo distance-Doppler matrix, taking the Doppler channel index corresponding to the theoretical clutter distribution curve where the current distance cell is located as the center, extend the preset number of Doppler channels to both sides, and determine the covered Doppler channel interval as the clutter suppression region of the current distance cell.

[0026] Optionally, the mean filtering replacement operation with protection unit includes:

[0027] Centered on each clutter cell, a local neighborhood window of size M×M is selected in the echo distance-Doppler matrix, where M is an odd number greater than 1;

[0028] The central region of the local neighborhood window is set as a protection unit; the protection unit contains the current clutter unit.

[0029] Calculate the average statistical data of the reference units outside the protection unit in the local neighborhood window;

[0030] Replace the original data value of the current clutter cell with the statistical mean.

[0031] Optionally, after performing a mean filtering replacement operation with guard cells on each clutter cell within the clutter suppression region to obtain the suppressed echo distance-Doppler matrix, the method further includes:

[0032] The echo distance-Doppler matrix after suppression processing is subjected to constant false alarm rate detection, and the cells that exceed the detection threshold are used as preliminary detection points;

[0033] The preliminary detection points are aggregated to obtain the target points.

[0034] Secondly, the present invention provides a clutter suppression device based on windowed mean filtering, comprising:

[0035] The processing module is used to perform pulse compression and coherent accumulation processing on the echo signal received by the radar to obtain the echo range-Doppler matrix containing the measured clutter.

[0036] The generation module is used to establish a range-Doppler two-dimensional distribution model of theoretical clutter based on real-time inertial navigation data and radar system parameters, and generate theoretical clutter distribution curves.

[0037] The determination module is used to determine a strip-shaped clutter suppression region for each range cell in the echo range-Doppler matrix, centered on the theoretical value of the theoretical clutter distribution curve at the current range cell.

[0038] The replacement module is used to perform a mean filter replacement operation with a guard unit on each clutter unit in the clutter suppression area to obtain the echo distance-Doppler matrix after suppression processing.

[0039] Thirdly, the present invention provides a computer device, the computer device including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the clutter suppression method based on windowed mean filtering as described in any one of the first aspects.

[0040] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, wherein a processor executes the computer program to implement the clutter suppression method based on windowed mean filtering as described in any one of the first aspects.

[0041] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0042] This application provides a clutter suppression method based on windowed mean filtering. The method includes: performing pulse compression and coherent accumulation processing on the echo signal received by the radar to obtain an echo range-Doppler matrix containing measured clutter; establishing a theoretical range-Doppler two-dimensional distribution model of clutter based on real-time inertial navigation data and radar system parameters, and generating a theoretical clutter distribution curve; for each range cell in the echo range-Doppler matrix, determining a strip-shaped clutter suppression region centered on the theoretical value of the theoretical clutter distribution curve at the current range cell; and performing a mean filtering replacement operation with guard cells on each clutter cell within the clutter suppression region to obtain the suppressed echo range-Doppler matrix.

[0043] As can be seen, this method acquires and utilizes inertial navigation data in real time for modeling. Regardless of the aircraft's maneuvering state, such as acceleration, deceleration, turning, or pitching, it can dynamically generate an accurate theoretical clutter curve for the current moment, thus overcoming the stringent constraints of DPCA technology on the aircraft's uniform linear motion. This method relies solely on precise inertial navigation information and a defined geometric model, making it insensitive to terrain undulations and scene non-uniformity, and eliminating dependence on uniform training samples, thus exhibiting excellent robustness. This method involves only logical judgments based on the theoretical curve and simple mean filtering, significantly reducing algorithm complexity and making it easy to implement in engineering. Through local mean filtering with guard units, this method effectively smooths and suppresses widely distributed clutter energy while avoiding the contamination of background estimation by strong clutter points. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0045] Figure 1 A flowchart illustrating the clutter suppression method based on windowed mean filtering provided in this application embodiment;

[0046] Figure 2 This is a schematic diagram of ground clutter velocity components provided in an embodiment of this application;

[0047] Figure 3 The distance-Doppler two-dimensional image corresponding to an azimuth angle of 0° provided in the embodiments of this application;

[0048] Figure 4 The theoretical clutter distribution curve corresponding to an azimuth angle of 0° is provided for the embodiments of this application;

[0049] Figure 5 The distance-Doppler two-dimensional image corresponding to an azimuth angle of 30° provided in this application embodiment;

[0050] Figure 6 The theoretical clutter distribution curve corresponding to an azimuth angle of 30° is provided for the embodiments of this application;

[0051] Figure 7 A schematic diagram showing the calculation results of the window opening positions of each distance unit provided in the embodiments of this application;

[0052] Figure 8 This is a schematic diagram of the clutter unit data value replacement method provided in the embodiments of this application;

[0053] Figure 9 A schematic diagram of the target point trace provided in the embodiments of this application. Figure 1 ;

[0054] Figure 10 A schematic diagram of the target point trace provided in the embodiments of this application. Figure 2 ;

[0055] Figure 11 This is a schematic diagram of the clutter suppression device based on windowed mean filtering provided in an embodiment of this application. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0057] Existing clutter suppression techniques mainly include Space-Time Adaptive Processing (STAP) and Offset Phase Center Antenna (DPCA). Although STAP technology has excellent performance in airborne radar clutter suppression, it still has the following drawbacks: (1) The computational complexity of this technology is too high. It requires real-time solution of the inverse operation of the covariance matrix with dimensions MN×MN (M is the number of pulses and N is the number of array elements), which has a computational load of up to For example, a 32-element × 32-pulse system needs to process a 1024-dimensional matrix. Even with a dimensionality reduction algorithm, its computational load is still far greater than that of traditional MTD processing. (2) Ideal clutter covariance matrix estimation requires a large number of uniform samples (usually 2MN≈2000 independent samples). In the actual environment, factors such as terrain undulation and target occlusion cause non-uniformity of samples, which leads to covariance matrix estimation error and output signal-to-noise ratio loss of more than 10dB.

[0058] The main disadvantages of DPCA technology include: (1) stringent requirements for the carrier platform: the core premise for the effective operation of DPCA is that the radar platform must maintain uniform straight flight. Any deviation in the platform's motion (such as acceleration, deceleration, turning, pitching, rolling, yaw) will disrupt the phase relationship between the echo signals of the two phase centers. (2) high requirements for amplitude and phase consistency: DPCA requires the two receiving channels (corresponding to two physically offset phase centers) to have extremely high amplitude and phase consistency (gain and phase response matching).

[0059] In summary, existing airborne clutter suppression technologies suffer from problems such as complex engineering implementation, stringent application conditions, and poor environmental adaptability. Therefore, this application provides a clutter suppression method based on windowed mean filtering. Please refer to... Figure 1 This is a flowchart illustrating the clutter suppression method based on windowed mean filtering provided in this application embodiment. The following is a further explanation. Figure 1The clutter suppression method based on windowed mean filtering is introduced.

[0060] S1. Perform pulse compression and coherent accumulation processing on the echo signal received by the radar to obtain the echo range-Doppler matrix containing the measured clutter.

[0061] In one possible embodiment, a reference signal designed using radar timing correlation parameters is used to perform pulse compression processing on the radar-received echo signal to obtain a pulse compressed signal; multiple pulse compressed signals within a coherent processing time are coherently accumulated to obtain an echo range-Doppler matrix containing measured clutter.

[0062] In practice, after receiving the relevant echo signal, the radar uses timing-related parameters such as the waveform, bandwidth, and duration of the radar's transmitted signal to design a reference signal. Through pulse compression processing (also known as matched filtering), this reference signal is correlated with the echo signal. This operation effectively compresses the signal duration, significantly improving the range dimension resolution and signal-to-noise ratio, focusing the echo energy onto the corresponding range cell, and outputting a series of pulse-compressed signals.

[0063] The signals transmitted by pulse-based radars are periodic pulse sequences. Therefore, it is necessary to process multiple pulse repetition periods within a coherent processing time. After acquiring multiple pulse compression signals, the radar performs moving target detection (MTD) processing (also known as coherent accumulation processing) on ​​these multiple pulse compression signals within a coherent processing time. The core of this processing is to perform a discrete Fourier transform on the slow time series, which transforms the signal from the range-time domain to the range-Doppler domain, thereby obtaining the echo range-Doppler matrix containing the measured clutter.

[0064] In the echo range-Doppler matrix, the range dimension (rows) represents different range cells, and its resolution is determined by the radar signal bandwidth. The Doppler dimension (columns) represents different Doppler frequency channels, and its resolution is determined by the coherent processing time. The matrix elements characterize the complex amplitude of the echo signal at a specific range and Doppler channel.

[0065] Because moving target echoes are coherent, their signal power can theoretically be increased by a factor of M² after coherent accumulation of M pulses. Noise, however, is not coherent, and its power increases only linearly by a factor of M with the number of accumulations. Therefore, MTD processing can theoretically improve the signal-to-noise ratio by a factor of M, significantly improving target detectability in noisy environments. This gives subsequent target detection based on the echo range-Doppler matrix obtained after MTD processing a greater advantage.

[0066] S2. Based on the real-time inertial navigation data and radar system parameters, establish a two-dimensional range-Doppler distribution model of theoretical clutter and generate theoretical clutter distribution curves.

[0067] In practice, the carrier aircraft is equipped with a radar system and an inertial navigation system. The inertial navigation system operates continuously, measuring the platform's motion state and spatial attitude parameters in real time, and transmitting these parameters as inertial navigation data to the radar system. The inertial navigation data includes the following parameters:

[0068] (1) Aircraft speed: The three-dimensional speed of the aircraft relative to the ground;

[0069] (2) Aircraft height: The vertical height of the aircraft relative to the ground;

[0070] (3) Attitude angles: These typically include pitch angle, roll angle, and heading angle. These angles define the relationship between the aircraft's body coordinate system and the geographic coordinate system.

[0071] Radar system parameters refer to a set of preset parameters that determine the radar detection mode, operating status, and signal characteristics, and include at least the following parameters:

[0072] (1) Radar beam azimuth angle: The angle between the projection of the radar antenna main lobe radiation direction onto the horizontal plane and the direction of the aircraft's motion;

[0073] (2) Radar beam angle: refers to the angle between the radiation direction of the main lobe of the radar antenna and the horizontal plane;

[0074] (3) Radar wavelength: refers to the center wavelength of radar signal propagating in free space.

[0075] In one possible embodiment, the specific steps of S2 include:

[0076] For each range cell in the echo range-Doppler matrix, based on the real-time inertial navigation data of the inertial navigation system and the radar system parameters, the theoretical clutter Doppler frequency of the current range cell is calculated using the radar Doppler principle. The theoretical clutter Doppler frequency of each range cell is mapped to the corresponding Doppler channel index in the echo range-Doppler matrix, and all mapping points are connected to form the theoretical clutter distribution curve.

[0077] Please refer to Figure 2 This is a schematic diagram of ground clutter velocity components provided in an embodiment of this application; where the black center dot represents the radar, and A represents the ground scattering point corresponding to the current range cell. The vector of the aircraft's velocity. H is the slant distance between the radar and the ground scattering point A, and H is the aircraft altitude. The radar beam elevation angle. This refers to the radar beam azimuth angle. The vector of the aircraft's velocity on R; Let be the angle between the vector of the aircraft's velocity and the line-of-sight direction from the radar to the ground scattering point A.

[0078] First, for each range cell in the echo range-Doppler matrix, the slant range from the radar to the ground scattering point corresponding to the current range cell is calculated based on the aircraft altitude and radar beam elevation angle.

[0079] The formula for calculating the slope distance R is as follows:

[0080]

[0081] Secondly, based on the slant range, aircraft altitude, and radar beam azimuth, the angle between the vector of the aircraft speed and the line-of-sight direction from the radar to the ground scattering point corresponding to the current range cell is calculated.

[0082] included angle The calculation formula is as follows:

[0083]

[0084] Finally, based on the aircraft speed, radar wavelength, and included angle, the theoretical clutter Doppler frequency of the current range cell is calculated.

[0085] Theoretical clutter Doppler frequency at ground scattering point A The calculation formula is as follows:

[0086]

[0087] Using the above formula, the theoretical clutter Doppler frequency can be calculated for each range cell (corresponding to a different slant range R) in the echo range-Doppler matrix, thus obtaining a complete range-Doppler two-dimensional distribution model. The theoretical clutter Doppler frequency of each range cell is mapped to the corresponding Doppler channel index in the echo range-Doppler matrix, and all mapping points are connected to form the theoretical clutter distribution curve.

[0088] For example, with an aircraft speed v = 20 m / s, an aircraft altitude H = 400 m, and a clutter ring coverage area of ​​8 km, this range-Doppler two-dimensional distribution model is used to model ground clutter at an azimuth angle of 0°, generating the corresponding range-Doppler two-dimensional image, as shown below. Figure 3 As shown, the corresponding theoretical clutter distribution curve is as follows: Figure 4 As shown, ground clutter with an azimuth angle of 30° is modeled, generating the corresponding range-Doppler 2D image, as shown. Figure 5 As shown, the corresponding theoretical clutter distribution curve is as follows: Figure 6 As shown.

[0089] S3. For each range cell in the echo range-Doppler matrix, determine a strip-shaped clutter suppression region centered on the theoretical value of the theoretical clutter distribution curve at the current range cell.

[0090] In practice, during actual flight of the aircraft, there will inevitably be error information between the measured clutter distribution curve and the theoretical clutter distribution curve. This causes the measured clutter distribution in the echo distance-Doppler matrix to fluctuate around the theoretical clutter curve, and they cannot be 100% identical. Therefore, in order to fully suppress the measured clutter, in this embodiment, multi-channel processing is performed near the theoretical clutter curve when windowing the measured clutter.

[0091] During airborne radar detection, the Doppler frequency of a target may only fall within the airborne clutter frequency range when the target is stationary or hovering. As long as the target has a certain speed, its velocity component relative to the airborne radar will differ from the clutter pattern. Therefore, statistically speaking, the probability of a target actually falling within the clutter frequency band is too small, and this application does not consider this situation. To ensure clutter suppression without affecting the detection of the target of interest, the window boundary issue needs to be considered when calculating the window position.

[0092] In one possible embodiment, the specific steps of S3 include:

[0093] Divide the minimum relative speed between the target and the carrier aircraft by the velocity resolution of a single Doppler channel to obtain a quotient. Round the quotient up to obtain the number of Doppler channels that the clutter suppression region extends to each side of the theoretical clutter distribution curve, which is used as a preset number. For each range cell in the echo range-Doppler matrix, take the Doppler channel index corresponding to the theoretical clutter distribution curve processed by the current range cell as the center and extend a preset number of Doppler channels to both sides. The covered Doppler channel interval is determined as the clutter suppression region of the current range cell.

[0094] In the specific implementation process, N= ;in, The minimum relative speed between the target and the carrier aircraft. For single Doppler channel velocity resolution, This indicates rounding up; N is the preset quantity.

[0095] Obtain the Doppler channel index for each range cell from the theoretical clutter distribution curve. ,by From the center, extend N channels to each side: This interval is the clutter suppression region (window) of the range cell (range gate).

[0096] Assuming a single Doppler channel velocity resolution of 0.4 m / s, and the target with a relative velocity of at least 2 m / s to the carrier aircraft needs to be detected, the calculated number of Doppler channels extending from the clutter suppression region to each side of the theoretical clutter distribution curve is 5. The calculation results for the windowing positions of each range cell are as follows: Figure 7 As shown, the dashed line formed by the black dots represents the theoretical clutter distribution curve. Taking the Doppler channel index corresponding to the theoretical clutter distribution curve processed by each distance cell as the center, five Doppler channels extend vertically and horizontally, and the Doppler channel interval covered is the clutter suppression region.

[0097] In this embodiment, considering that the theoretical clutter distribution curve is an ideal model, and the measured clutter fluctuates in its vicinity due to terrain undulations, systematic errors, etc., striped windowing (±N Doppler channels) can cover the actual clutter distribution range, avoiding residual suppression due to model errors. Furthermore, since the theoretical clutter distribution curve changes with distance, the windowing position also changes with distance, and this adaptive performance better matches the actual clutter distribution.

[0098] S4. For each clutter cell within the clutter suppression region, perform a mean filter replacement operation with guard cells to obtain the echo distance-Doppler matrix after suppression.

[0099] Clutter processing is performed on each clutter cell within the clutter suppression region to obtain the suppressed echo range-Doppler matrix. The specific steps of clutter processing are as follows:

[0100] Centered on each clutter cell, a local neighborhood window of size M×M is selected in the echo distance-Doppler matrix, where M is an odd number greater than 1; the central region of the local neighborhood window is set as the guard cell; the guard cell contains the current clutter cell; the statistical mean of the data of the reference cells in the local neighborhood window excluding the guard cell is calculated; the original data value of the current clutter cell is replaced with the statistical mean.

[0101] In the specific implementation process, to determine the local neighborhood window centered on the current clutter cell, a windowing method is adopted. Guard cells are set around each clutter cell, and each clutter cell is also assigned to the corresponding guard cell. The data of the guard cells are not included in the subsequent averaging calculation. After the guard cells are extracted, the average value of the remaining reference cell data in the local neighborhood window is calculated, and the original data value of the clutter cell is replaced with the average value.

[0102] Please refer to Figure 8This is a schematic diagram of the clutter cell data value replacement method provided in the embodiments of this application. The entire 9×9 rectangular window represents the local neighborhood window, the black rectangle in the center represents the current clutter cell, the gray rectangle around the current clutter cell represents the guard cell, and when performing the mean calculation, the current clutter cell is also assigned to the guard cell. All the white rectangles represent the reference cell.

[0103] Considering that replacing the entire clutter suppression region with the same global mean would result in a completely flat region, losing true background undulation information, this embodiment uses a local neighborhood window to calculate the mean. This ensures that the replacement values ​​for each clutter unit are unequal, making the replacement value for each unit closer to the actual data information near that unit, preserving the spatial variation characteristics of the echo distance-Doppler matrix, and better reflecting the spatial correlation of actual clutter. Considering that if a strong central clutter unit participates in the mean calculation, it would significantly increase the mean, resulting in a high residual value after replacement, this embodiment includes the central unit in the protection unit and excludes it from the calculation, ensuring that the mean reflects the true background level around it, thereby more thoroughly suppressing the clutter point.

[0104] In one possible embodiment, after performing S4, the method further includes:

[0105] Constant False Alarm Rate (CFAR) detection is performed on the suppressed echo distance-Doppler matrix to obtain cells that exceed the detection threshold, which are used as preliminary detection points. The preliminary detection points are then aggregated to obtain the target points, and the distance, velocity, and amplitude information of the target points can be output.

[0106] For a set of airborne real-world data, the target point traces obtained using conventional processing methods are as follows: Figure 9 As shown, the target point trace obtained by the clutter suppression method based on windowed mean filtering proposed in this application is as follows: Figure 10 As shown in the figure. The horizontal axis represents the signal frame number, also known as time, and the vertical axis represents the distance to the target point. The figure includes a target point moving towards the station from approximately 2.6 kilometers away. (Comparison) Figure 9 and Figure 10 It can be seen that the clutter suppression method based on windowed mean filtering provided in this application has a significant effect on suppressing clutter traces generated during airborne radar detection, and the target traces are clearly visible.

[0107] In this embodiment, the clutter suppression method based on windowed mean filtering provided in this application is used to obtain the range-Doppler matrix after suppression. The background is uniform and the signal-to-clutter ratio is significantly improved, which creates ideal conditions for constant false alarm rate detection, thereby greatly improving the target detection probability and stabilizing the false alarm rate, and finally achieving reliable output of clear and continuous target points.

[0108] In summary, to address the problem of excessive clutter interference during airborne radar target detection, this application provides a clutter suppression method based on windowed mean filtering. This method utilizes inertial navigation data provided by the inertial navigation system to perform real-time modeling and calculation of the clutter distribution. Through theoretical clutter distribution estimation, data windowing is performed near the clutter channel in the radar echo range-Doppler pair. The windowed data is then used for mean filtering of the airborne clutter, resulting in the following beneficial effects:

[0109] 1. Simple calculation and convenient engineering implementation:

[0110] This method avoids the complex covariance matrix inversion and large-scale matrix operations found in algorithms such as Space-Time Adaptive Processing (STAP). It only utilizes the aircraft's inertial navigation information for clutter model estimation and achieves clutter suppression through simple operations of traversal windowing and local mean filtering in the range-Doppler matrix. The process has clear logic and is easy to implement in parallel on embedded hardware platforms such as FPGAs and DSPs, greatly reducing the difficulty and cost of engineering development.

[0111] 2. Highly adaptable with minimal restrictions on the maneuverability of the carrier aircraft:

[0112] Unlike techniques such as the Offset Phase Center Antenna (DPCA), which impose stringent requirements on the aircraft's strictly uniform linear motion, this method offers greater tolerance for varying aircraft motion states. Because radar frame processing time is extremely short (microseconds), the aircraft's variable-speed motion during this period can be approximated as uniform. Therefore, this method can update the clutter theory model frame-by-frame based on real-time inertial navigation information, effectively adapting to speed changes and maneuvers during actual flight, significantly improving the method's practicality and scenario adaptability.

[0113] 3. Significant inhibitory effect:

[0114] This method, through striped region determination guided by theoretical curves and local adaptive filtering, can effectively and fully suppress the main body of ground clutter and significantly reduce the background clutter level.

[0115] Based on the same inventive concept, please refer to Figure 11 This application also provides a clutter suppression device based on windowed mean filtering, the device comprising:

[0116] The processing module is used to perform pulse compression and coherent accumulation processing on the echo signal received by the radar to obtain the echo range-Doppler matrix containing the measured clutter.

[0117] The generation module is used to establish a range-Doppler two-dimensional distribution model of theoretical clutter based on real-time inertial navigation data and radar system parameters, and generate theoretical clutter distribution curves.

[0118] The determination module is used to determine a strip-shaped clutter suppression region for each range cell in the echo range-Doppler matrix, centered on the theoretical value of the theoretical clutter distribution curve at the current range cell.

[0119] The replacement module is used to perform a mean filter replacement operation with a guard unit on each clutter unit in the clutter suppression area to obtain the echo distance-Doppler matrix after suppression processing.

[0120] Optionally, the generation module is specifically used for:

[0121] For each range cell in the echo range-Doppler matrix, based on the real-time inertial navigation data of the inertial navigation system and the radar system parameters, the theoretical clutter Doppler frequency of the current range cell is calculated using the radar Doppler principle; the inertial navigation data includes the aircraft speed and aircraft altitude; the radar system parameters include the radar wavelength, radar beam azimuth angle, and radar beam elevation angle.

[0122] The theoretical clutter Doppler frequencies of each distance cell are mapped to the corresponding Doppler channel indices in the echo distance-Doppler matrix, and all mapping points are connected to form the theoretical clutter distribution curve.

[0123] Optionally, the generation module is specifically used for:

[0124] For each range cell in the echo range-Doppler matrix, the slant range from the radar to the ground scattering point corresponding to the current range cell is calculated based on the aircraft altitude and radar beam elevation angle.

[0125] Based on the slant range, aircraft altitude, and radar beam azimuth, calculate the angle between the vector of the aircraft speed and the line-of-sight direction from the radar to the ground scattering point corresponding to the current range cell;

[0126] Based on the aircraft speed, radar wavelength, and included angle, calculate the theoretical clutter Doppler frequency of the current range cell.

[0127] Optionally, the processing module is specifically used for:

[0128] Using a reference signal designed with radar timing correlation parameters, pulse compression processing is performed on the echo signal received by the radar to obtain a pulse compressed signal;

[0129] Coherent accumulation processing is performed on multiple pulse compression signals within a coherent processing time to obtain the echo distance-Doppler matrix containing measured clutter.

[0130] Optionally, determine the specific use of the module for:

[0131] Divide the minimum relative velocity between the target and the carrier aircraft by the velocity resolution of a single Doppler channel to obtain a quotient.

[0132] The quotient is rounded up to obtain the number of Doppler channels that the clutter suppression region extends to each side of the theoretical clutter distribution curve, which is used as the preset number.

[0133] For each range cell in the echo range-Doppler matrix, a preset number of Doppler channels are extended to both sides of the Doppler channel index corresponding to the clutter distribution curve processed by the current range cell as the center, and the covered Doppler channel interval is determined as the clutter suppression region of the current range cell.

[0134] Optionally, the replacement module is specifically used for:

[0135] Centered on each clutter cell, a local neighborhood window of size M×M is selected in the echo range-Doppler matrix, where M is an odd number greater than 1;

[0136] Set the center region of the local neighborhood window as a protection cell; the protection cell contains the current clutter cell.

[0137] Calculate the statistical mean of the reference cells excluding the protection cells within the local neighborhood window;

[0138] Replace the original data value of the current clutter cell with the statistical mean.

[0139] Optionally, the device also includes a detection module, which is used for:

[0140] For each clutter cell within the clutter suppression region, a mean filtering replacement operation with guard cells is performed to obtain the suppressed echo distance-Doppler matrix. Then, constant false alarm rate detection is performed on the suppressed echo distance-Doppler matrix to obtain cells that exceed the detection threshold, which are used as preliminary detection points.

[0141] The initial detection points are aggregated to obtain the target points.

[0142] It should be noted that each module in the clutter suppression device based on windowed mean filtering in this embodiment corresponds one-to-one with each step in the clutter suppression method based on windowed mean filtering in the aforementioned embodiment. Therefore, the specific implementation of this embodiment can refer to the implementation of the clutter suppression method based on windowed mean filtering described above, and will not be repeated here.

[0143] Based on the same inventive concept, this application also provides a computer device, which includes a processor, a memory, and a computer program stored in the memory. The computer program is executed by the processor to implement the aforementioned clutter suppression method based on windowed mean filtering.

[0144] Based on the same inventive concept, this application also provides a computer storage medium storing a computer program, which is executed by a processor to implement the aforementioned clutter suppression method based on windowed mean filtering.

[0145] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or it may be a device including one or any combination of the above-mentioned memories. The computer may be a variety of computing devices, including smart terminals and servers.

[0146] In some embodiments, executable instructions may take the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a standalone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0147] As an example, executable instructions may, but do not necessarily, correspond to files in the file system. They may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a Hyper Text Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple collaborative files (e.g., a file that stores one or more modules, subroutines, or code sections).

[0148] As an example, executable instructions can be deployed to execute on a single computing device, or on multiple computing devices located in one location, or on multiple computing devices distributed across multiple locations and interconnected via a communication network.

[0149] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0150] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0151] The above specific embodiments further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A clutter suppression method based on windowed mean filtering, characterized in that, include: The echo signal received by the radar is processed by pulse compression and coherent accumulation to obtain the echo range-Doppler matrix containing the measured clutter. Based on real-time inertial navigation data and radar system parameters, a range-Doppler two-dimensional distribution model of theoretical clutter is established, and theoretical clutter distribution curves are generated. For each range cell in the echo range-Doppler matrix, a strip-shaped clutter suppression region is determined centered on the theoretical value of the theoretical clutter distribution curve at the current range cell. For each clutter unit within the clutter suppression region, a mean filter replacement operation with a guard unit is performed to obtain the echo distance-Doppler matrix after suppression processing.

2. The clutter suppression method based on windowed mean filtering according to claim 1, characterized in that, Based on the real-time inertial navigation system data and radar system parameters, a range-Doppler two-dimensional distribution model of theoretical clutter is established, generating theoretical clutter distribution curves, including: For each range cell in the echo range-Doppler matrix, based on the real-time inertial navigation data of the inertial navigation system and the radar system parameters, the theoretical clutter Doppler frequency of the current range cell is calculated using the radar Doppler principle; the inertial navigation data includes the aircraft speed and aircraft altitude; the radar system parameters include the radar wavelength, radar beam azimuth angle, and radar beam elevation angle. The theoretical clutter Doppler frequency of each distance cell is mapped to the corresponding Doppler channel index in the echo distance-Doppler matrix, and all mapping points are connected to form the theoretical clutter distribution curve.

3. The clutter suppression method based on windowed mean filtering according to claim 2, characterized in that, For each range cell in the echo range-Doppler matrix, based on real-time inertial navigation data and radar system parameters, and using the radar Doppler principle, the theoretical clutter Doppler frequency of the current range cell is calculated, including: For each range cell in the echo range-Doppler matrix, the slant range from the radar to the ground scattering point corresponding to the current range cell is calculated based on the aircraft altitude and the radar beam elevation angle. Based on the slant range, the aircraft altitude, and the radar beam azimuth, calculate the angle between the vector of the aircraft speed and the line-of-sight direction from the radar to the ground scattering point corresponding to the current range cell; Based on the aircraft speed, the radar wavelength, and the included angle, the theoretical clutter Doppler frequency of the current range cell is calculated.

4. The clutter suppression method based on windowed mean filtering according to claim 1, characterized in that, The pulse compression and coherent accumulation processing of the radar-received echo signal yields an echo range-Doppler matrix containing measured clutter, including: Using a reference signal designed with radar timing correlation parameters, pulse compression processing is performed on the echo signal received by the radar to obtain a pulse compressed signal; Coherent accumulation processing is performed on multiple pulse compression signals within a coherent processing time to obtain an echo distance-Doppler matrix containing measured clutter.

5. A clutter suppression method based on windowed mean filtering according to claim 1, characterized in that, For each range cell in the echo range-Doppler matrix, a strip-shaped clutter suppression region is determined centered on the theoretical value of the theoretical clutter distribution curve at the current range cell, including: Divide the minimum relative velocity between the target and the carrier aircraft by the velocity resolution of a single Doppler channel to obtain a quotient. The quotient is rounded up to obtain the number of Doppler channels that the clutter suppression region extends to each side of the theoretical clutter distribution curve, which is used as a preset number. For each distance cell in the echo distance-Doppler matrix, taking the Doppler channel index corresponding to the theoretical clutter distribution curve where the current distance cell is located as the center, extend the preset number of Doppler channels to both sides, and determine the covered Doppler channel interval as the clutter suppression region of the current distance cell.

6. The clutter suppression method based on windowed mean filtering according to claim 1, characterized in that, The mean filtering replacement operation with protection unit includes: Centered on each clutter cell, a local neighborhood window of size M×M is selected in the echo distance-Doppler matrix, where M is an odd number greater than 1; The central region of the local neighborhood window is set as a protection unit; the protection unit contains the current clutter unit. Calculate the average statistical data of the reference units outside the protection unit in the local neighborhood window; Replace the original data value of the current clutter cell with the statistical mean.

7. The clutter suppression method based on windowed mean filtering according to claim 1, characterized in that, After performing a mean filtering replacement operation with guard cells on each clutter cell within the clutter suppression region to obtain the suppressed echo distance-Doppler matrix, the method further includes: The echo distance-Doppler matrix after suppression processing is subjected to constant false alarm rate detection, and the cells that exceed the detection threshold are used as preliminary detection points; The preliminary detection points are aggregated to obtain the target points.

8. A clutter suppression device based on windowed mean filtering, characterized in that, include: The processing module is used to perform pulse compression and coherent accumulation processing on the echo signal received by the radar to obtain the echo range-Doppler matrix containing the measured clutter. The generation module is used to establish a range-Doppler two-dimensional distribution model of theoretical clutter based on real-time inertial navigation data and radar system parameters, and generate theoretical clutter distribution curves. The determination module is used to determine a strip-shaped clutter suppression region for each range cell in the echo range-Doppler matrix, centered on the theoretical value of the theoretical clutter distribution curve at the current range cell. The replacement module is used to perform a mean filter replacement operation with a guard unit on each clutter unit in the clutter suppression area to obtain the echo distance-Doppler matrix after suppression processing.

9. A computer device, characterized in that, The computer device includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement a clutter suppression method based on windowed mean filtering as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and the processor executes the computer program to implement a clutter suppression method based on windowed mean filtering as described in any one of claims 1-7.

Images