Bistatic end-fire array airborne radar clutter suppression method and system

Abstract

This invention discloses a clutter suppression method and system for a dual-base-end array airborne radar, relating to the field of airborne radar technology. The method includes the following steps: acquiring attitude angles, position coordinates, and velocity vector data of the transmitting and receiving ends; reconstructing the geometric baseline between the two ends under a unified reference frame; performing spatiotemporal joint analysis on the echo signal under geometric baseline constraints to generate a clutter spectrum; constructing a spatiotemporal filtering weight vector based on the angle and frequency distribution characteristics of the clutter spectrum; applying local sparse constraints to suppress noise energy in conjunction with the filtered output; and performing multi-channel coherent superposition of the processing results to achieve signal energy aggregation. This invention, through geometric baseline constraints under a unified reference frame and clutter spectrum construction, enables the spatiotemporal filtering direction to adaptively match the clutter structure, improving clutter suppression stability under asymmetric configurations; and by combining local sparse constraints and multi-channel coherent energy aggregation, it enhances target echo prominence and improves the detection capability of weak targets in complex clutter environments.

Description

Technical Field

[0001] This invention relates to the field of airborne radar technology, specifically to a clutter suppression method and system for dual-base-array airborne radar. Background Technology

[0002] A bistatic end-array airborne radar refers to an airborne radar system where the transmitter and receiver are separately arranged. The transmitter radiates electromagnetic waves into the target area through an array antenna, while the receiver receives echo signals from the target and environment via an independent platform or a relatively separate antenna array. Due to the inconsistent spatial distribution of the transmitter and receiver, the system geometrically forms a bistatic configuration, resulting in asymmetric characteristics in target scattering properties, propagation paths, and Doppler distribution. Clutter suppression in bistatic end-array airborne radar refers to the entire process of identifying, modeling, and attenuating the large number of background echo signals from non-target scattering sources such as the ground, sea surface, and atmosphere under these asymmetric spatial geometric conditions. Specifically, by constructing a bistatic geometric and motion parameter model, analyzing the phase coupling relationship of each array element in the spatial and temporal domains, and using techniques such as space-time adaptive processing (STAP), multi-channel beamforming, subspace projection, or sparse reconstruction, clutter components are dynamically separated, allowing the target signal to stand out clearly in a complex background, thereby improving the detection sensitivity and positioning accuracy of weak targets.

[0003] The existing technology has the following shortcomings:

[0004] In dual-base airborne radar, the spatiotemporal distribution of clutter is highly dependent on range or elevation angle, exhibiting significant range nonstationarity. This means that as the range or elevation angle changes, the angle-Doppler structure and covariance characteristics of the clutter continuously change, making traditional two-dimensional spatiotemporal adaptive processing methods prone to failure and significantly reducing clutter suppression performance. Furthermore, the geometric relationship and motion state between the transmitter and receiver are complex, especially when the transmitter is maneuvering. The coupling relationships of clutter in spatial, temporal, Doppler, and angular dimensions become even more complex and difficult to accurately characterize with a fixed model. In this situation, estimating clutter statistical characteristics, such as the covariance matrix, requires a large number of independent and identically distributed training samples. However, in real-world scenarios with rapidly changing airborne platforms and frequent clutter disturbances, sufficient samples are often difficult to obtain, easily leading to model estimation bias and thus significantly reducing clutter suppression performance.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a clutter suppression method and system for dual-base-array airborne radar to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a clutter suppression method for dual-base-array airborne radar, comprising the following steps:

[0008] Step 1: Based on the spatial distribution characteristics of the dual-base-end array airborne radar, acquire the attitude angle, position coordinates and velocity vector data of the transmitter and receiver. Establish the dynamic spatial mapping relationship between the transmitter and receiver under a unified reference system. Reconstruct the geometric baseline between the transmitter and receiver according to the time series to provide stable spatial constraints for subsequent signal processing.

[0009] Step 2: Under the established geometric baseline constraints, perform spatiotemporal joint analysis on the echo signal obtained by the receiver, and couple and map the angular parameters of the echo signal with the Doppler frequency shift to generate a clutter spectrum with real geometric projection characteristics, which is used to characterize the distribution characteristics of clutter energy in the angular and frequency dimensions.

[0010] Step 3: Based on the angle and frequency coupling distribution of the generated clutter spectrum, and according to the energy continuity characteristics between adjacent distance cells, perform rotational weighting control on the clutter spectrum to generate a space-time filtering weight vector that adaptively changes along the distance dimension, so that the space-time filtering direction continuously matches the local clutter structure, thereby maintaining the directional weakening of clutter energy.

[0011] Step 4: Using the generated spatiotemporal filtering weight vector, apply local sparsity constraints, set weights based on the clutter spectral similarity between adjacent distance cells, dynamically suppress noise energy far from the main spectrum direction, and perform selective updates in clutter characteristic abrupt regions to ensure the continuity and stability of the filtering output in the distance dimension.

[0012] Step 5: Based on the filtered output after sparse constraint processing, geometric phase correction and energy aggregation are performed on the multi-channel received signals. Coherent superposition of the multi-channel received signals is achieved under a unified reference frame. Through directional energy enhancement control, the target echo is effectively highlighted in the fused signal, thereby achieving prominent detection of weak targets and deep suppression of background clutter.

[0013] Preferably, reconstructing the geometric baseline between the transmitter and receiver according to a time series includes the following steps:

[0014] The spatial state information of the transmitter and receiver is collected to obtain the attitude angle, position coordinates and velocity vector data of the transmitter and receiver under a unified time reference, and stored in the form of time series.

[0015] Based on attitude angle, position coordinates and velocity vector data, coordinate transformation and geometric registration are performed on the spatial state of the transmitter and receiver under a unified reference system, so that the spatial position and attitude relationship between the transmitter and receiver can be expressed in a unified coordinate system.

[0016] In a unified reference frame, the spatial position difference vector between the transmitter and receiver is calculated point by point based on time series data, forming a spatial line segment connecting the transmitter and receiver, and a geometric baseline that changes continuously with time is constructed using this spatial line segment.

[0017] Based on the temporal variation characteristics of the geometric baseline, the geometric baseline is introduced as a spatial constraint into the subsequent signal processing to describe the spatial propagation relationship between the transmitter and receiver, and to provide a spatial reference for subsequent spatiotemporal joint analysis and spatiotemporal filtering weight vector generation.

[0018] Preferably, the method of coupling and mapping the angular parameters of the echo signal with the Doppler frequency shift to generate a clutter spectrum includes the following steps:

[0019] Under the established geometric baseline constraints, the spatial direction of the echo signal acquired by the receiver is correlated, and the angle parameters of the echo signal of each channel are matched with the geometric baseline direction at the corresponding time, so that the spatial angle of the echo signal is redefined in a unified reference system.

[0020] Based on the spatial orientation correlation results, the echo signals are organized and matched in the time dimension according to the temporal variation characteristics of the geometric baseline, so that the temporal characteristics of the echo signals in each time period are continuously correlated with the changes in the direction and length of the geometric baseline.

[0021] Based on the unified mapping of spatial direction and time dimension, and according to the directionality and variation law of geometric baseline, the angular parameters of the echo signal and the Doppler frequency shift are coupled and mapped, and a clutter spectrum reflecting the clutter energy distribution characteristics in the angular and frequency dimensions is constructed in a unified reference frame.

[0022] Preferably, in the process of constructing the clutter spectrum, the spatial direction of the geometric baseline in the unified reference frame is used as the projection reference of the spectrum. The angular parameters of the echo signals in each time period are projected to the corresponding spatial direction positions, and the Doppler frequency shift distribution obtained by time dimension matching is synchronously mapped to the corresponding spatial direction positions. This ensures that the clutter spectrum maintains a spatial projection relationship consistent with the geometric baseline direction in different time periods, thereby continuously characterizing the distribution state of clutter energy in the angular and frequency dimensions.

[0023] Preferably, the clutter spectrum is subjected to rotational weighting to generate a space-time filtering weight vector that adaptively varies along the distance dimension, including the following steps:

[0024] Energy distribution analysis is performed on the generated clutter spectrum under a unified reference frame to extract the energy concentration regions of each distance cell in the angular and frequency dimensions, and to determine the energy continuity characteristics between adjacent distance cells.

[0025] Based on the energy continuity characteristic, the clutter spectrum of adjacent distance cells is rotated and weighted according to the relationship of its main energy direction, so that the clutter spectrum maintains a continuous transition in the spatial angular direction.

[0026] Based on the clutter spectrum after rotational weighting, the main direction of clutter energy is extracted in each range cell, and combined with the relationship of the main directions of adjacent range cells, a space-time filtering weight vector that varies continuously along the range dimension is generated.

[0027] The space-time filtering weight vector is directionally matched with the clutter spectrum of the corresponding range cell in a unified reference frame and applied to echo signal processing. This ensures that the space-time filtering direction continuously matches the local clutter structure, thereby achieving directional attenuation of clutter energy.

[0028] Preferably, when generating the space-time filtering weight vector that changes continuously along the distance dimension, the main energy direction of the clutter spectrum corresponding to each distance cell is used as the direction reference of the weight vector, and the space-time filtering weight vector is continuously connected according to the energy distribution change relationship of the clutter spectrum between adjacent distance cells, so that the space-time filtering weight vector maintains directional consistency in the distance dimension and is adjusted synchronously with the change of clutter spectrum structure.

[0029] Preferably, the local sparsity constraint is applied using the filtering output result corresponding to the space-time filtering weight vector, including the following steps:

[0030] Based on the filtering output results corresponding to the space-time filtering weight vector, the energy distribution of each distance unit is extracted along the distance dimension to distinguish the energy band in the main spectrum direction from the noise energy band.

[0031] Based on energy distribution, the clutter spectral structure of adjacent distance cells is compared to determine the clutter spectral similarity between adjacent distance cells and set corresponding weights.

[0032] Based on similarity weights, local sparsity constraints are applied to the filtered output to dynamically suppress noise energy far from the main spectrum direction;

[0033] Based on the changes in similarity weights, selective updates are performed within the distance cells where clutter spectral similarity abruptly changes, adjusting the local sparsity constraint strength.

[0034] Based on local sparsity constraints and selective update results, the filtering output results of each distance unit are integrated to maintain the continuity and stability of the filtering output results in the distance dimension.

[0035] Preferably, selective updates are performed within the distance cells where the clutter spectral similarity changes abruptly. This includes adjusting the scope of the local sparsity constraint based on the relationship between the main spectral direction of the distance cell and the main spectral direction of adjacent distance cells, so that the energy distribution corresponding to the main spectral direction remains continuous and the noise energy far from the main spectral direction is continuously suppressed.

[0036] Preferably, based on the filtered output result after sparse constraint processing, geometric phase correction and energy aggregation are performed on the multi-channel received signal, including the following steps:

[0037] Based on the filtered output results after sparse constraint processing, combined with the spatial geometric relationship between the transmitter and receiver, geometric phase correction is performed on the multi-channel received signals to ensure that the received signals of each channel maintain spatial phase consistency under a unified reference frame.

[0038] Based on the multi-channel received signal with geometric phase correction, the energy amplitude of the received signal of each channel is balanced and continuously matched to ensure that the energy distribution of different channels is consistent within the same distance cell.

[0039] Based on the multi-channel received signals that have achieved energy balance, the signals of each channel are coherently superimposed according to the geometric baseline direction under a unified reference frame to realize the energy aggregation of the multi-channel received signals.

[0040] Based on the fused signal formed by coherent superposition, directional energy enhancement control is applied to the fused signal to make the target echo stand out in the fused signal and suppress background clutter.

[0041] The dual-base-end array airborne radar clutter suppression system includes a geometric baseline construction module, a space-time spectrum generation module, an adaptive filter weight generation module, a local sparse constraint module, and a multi-channel coherent fusion module.

[0042] Geometric baseline construction module: acquires attitude angles, position coordinates and velocity vector data of the transmitter and receiver, establishes dynamic spatial mapping relationship between the transmitter and receiver under a unified reference frame, and reconstructs the geometric baseline between the transmitter and receiver according to the time series;

[0043] Space-time spectrum generation module: Under the constraints of the established geometric baseline, the module performs joint space-time analysis on the echo signal obtained by the receiver, and couples and maps the angular parameters of the echo signal with the Doppler frequency shift to generate clutter spectrum.

[0044] Adaptive filter weight generation module: Based on the angle and frequency coupling distribution of the clutter spectrum, and according to the energy continuity characteristics between adjacent distance cells, the clutter spectrum is subjected to rotational weighting control to generate a space-time filter weight vector that adaptively varies along the distance dimension.

[0045] Local sparsity constraint module: Utilizes the filtering output corresponding to the space-time filtering weight vector to apply local sparsity constraint processing, sets weights based on the clutter spectral similarity between adjacent distance cells, and dynamically suppresses noise energy far from the main spectrum direction;

[0046] Multi-channel coherent fusion module: Based on the filtered output results after sparse constraint processing, geometric phase correction and energy aggregation are performed on the multi-channel received signals, and coherent superposition of the multi-channel received signals is realized under a unified reference frame.

[0047] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0048] This invention reconstructs the time-varying geometric baseline between the transmitter and receiver in a unified reference frame, ensuring that the spatial characteristics of the echo signal are always guided by spatial constraints. This fundamentally alleviates the non-stationary problem caused by clutter variations with range or elevation angle in dual-baseline airborne radars. The clutter spectrum generated under geometric baseline constraints accurately reflects the clutter distribution in the angular and frequency dimensions, ensuring that the spacetime filtering direction remains consistent with the local clutter structure. This maintains stable clutter suppression even under complex maneuvering conditions, improving the radar's background suppression capability in asymmetric spatial configurations.

[0049] This invention introduces local sparsity constraints and selective update mechanisms on the basis of spatiotemporal filtering output, and combines geometric phase correction and coherent energy aggregation of multi-channel received signals to maintain a continuous and stable energy distribution in the range dimension of the processing result. Through directional energy enhancement control, the target echo is concentrated and enhanced in the fused signal, while background clutter and noise energy are suppressed in non-dominant directional regions. This improves the discriminability of weak targets in complex clutter environments, providing a clearer and more reliable signal foundation for subsequent target detection and localization. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0051] Figure 1 This is a flowchart of the clutter suppression method for dual-base-end array airborne radar of the present invention;

[0052] Figure 2 This is a flowchart illustrating the reconstruction of the geometric baseline between the transmitter and receiver according to a time series in this invention.

[0053] Figure 3This is a flowchart illustrating the process of performing rotational weighting on the clutter spectrum and generating a space-time filtering weight vector according to the present invention.

[0054] Figure 4 This is a schematic diagram of the modules of the dual-base-end array airborne radar clutter suppression system of the present invention. Detailed Implementation

[0055] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0056] This invention provides, for example Figures 1 to 3 The clutter suppression method for dual-base-array airborne radar shown includes the following steps:

[0057] Step 1: Based on the spatial distribution characteristics of the dual-base-end array airborne radar, acquire the attitude angle, position coordinates and velocity vector data of the transmitter and receiver. Establish the dynamic spatial mapping relationship between the transmitter and receiver under a unified reference system. Reconstruct the geometric baseline between the transmitter and receiver according to the time series to provide stable spatial constraints for subsequent signal processing.

[0058] The specific steps for reconstructing the geometric baseline between the transmitter and receiver based on time series data are as follows:

[0059] First, precise spatial state information is collected from both the transmitter and receiver. This process involves deploying high-precision inertial measurement units (IMUs) and global positioning systems (GPS) on the transmitter platform to record real-time three-dimensional position coordinates, heading angles, pitch angles, and roll angles during the transmitter's flight. Simultaneously, velocity sensors acquire the transmitter's three-dimensional velocity vector. The receiver platform is similarly equipped with attitude measurement and positioning devices to synchronously acquire the receiver's three-dimensional position coordinates, attitude angle parameters, and instantaneous velocity vector. To ensure temporal consistency of the collected data, both the transmitter and receiver use a unified time base for data recording, and a time synchronization signal ensures that the data from both ends are aligned within milliseconds. All acquired attitude angle, position coordinate, and velocity vector data are stored in time-series format for subsequent spatial coordinate unification and geometric baseline reconstruction.

[0060] After acquiring the spatial state information of the transmitter and receiver, coordinate transformation and geometric registration are performed on the attitude angles, position coordinates, and velocity vectors of both ends based on a unified reference system. This process begins by selecting a global reference coordinate system as the unified spatial benchmark, such as a fixed ground coordinate system with geographic coordinates as its origin. Then, based on the attitude angle parameters of the transmitter and receiver, the three-dimensional position coordinates of the transmitter and receiver in their respective local coordinate systems are transformed to the global reference coordinate system using an attitude matrix. This ensures that the spatial position, orientation, and velocity direction of the transmitter and receiver are uniformly expressed within the same coordinate system. Through this coordinate unification, the absolute position vectors and velocity vectors of the transmitter and receiver in the unified reference system can be obtained. At this point, the relative orientation of the transmitter and receiver in space is clear; the direction from the transmitter to the receiver, the height difference between the two ends, and the horizontal distance can all be directly reflected by this unified coordinate data. By continuously performing this spatial coordinate transformation and attitude compensation, the geometric relationship between the transmitter and receiver can be accurately described within the same coordinate system at different points in time.

[0061] After establishing the spatial coordinates and attitude relationships between the transmitter and receiver in a unified reference frame, the geometric baseline between them is further reconstructed based on time-series data. Specifically, by calculating the spatial position difference vector between the transmitter and receiver in the unified reference frame at each time point, a spatial line segment connecting the two ends is obtained. The direction and length of this line segment at different time points reflect the spatial geometric changes between the transmitter and receiver. As time progresses, the spatial positions of the transmitter and receiver are continuously updated, thus forming a continuously changing spatial trajectory in the time dimension, which is the geometric baseline. By recording the time series of this geometric baseline, the directional change trend, altitude change characteristics, and distance change patterns of the relative motion between the transmitter and receiver during flight can be obtained. In this process, the starting point of the geometric baseline is the spatial position of the transmitter at a certain moment, and the ending point is the spatial position of the receiver at the same moment. The direction of the line segment is from the transmitter to the receiver, and the length represents the spatial distance between the transmitter and receiver. The continuous calculation and updating of this geometric baseline enables the system to maintain the continuity of the spatial geometric description even when the attitudes of the transmitter and receiver are constantly changing. By concatenating geometric baselines at consecutive moments, a three-dimensional dynamic trajectory of the relative motion between the transmitter and receiver can be obtained, which can be used to reflect the changes in spatial relationships during platform maneuvers.

[0062] After the geometric baseline is reconstructed, it serves as a spatial constraint, providing a stable spatial reference for subsequent signal processing. In this stage, by combining the geometric baseline at each moment with the corresponding attitude information of the transmitter and receiver, the propagation direction and path of the electromagnetic wave during transmission and reception can be clearly defined. Based on the directional and length variation characteristics of the geometric baseline, the propagation path of the echo signal in the unified reference frame can be accurately calibrated, ensuring that the subsequent analysis of physical parameters such as angle, pitch, distance, and time delay in signal processing is based on the real spatial relationships. Simultaneously, the temporal continuity of the geometric baseline allows for continuous spatial compensation of the signal propagation path during platform maneuvers, attitude adjustments, or speed changes, preventing the accumulation of spatial deviations caused by platform motion. In this way, the spatial geometric relationship between the transmitter and receiver remains consistent in the dynamic environment, and the spatial constraints of signal propagation are continuously maintained, providing a spatial reference for subsequent spatiotemporal joint analysis and the generation of spatiotemporal filter weight vectors. Furthermore, by updating the geometric baseline in real time and continuously referencing this geometric baseline information, it is possible to ensure that the spatial mapping relationship between the transmitter and receiver is not affected by external disturbances, even when the flight trajectory of the airborne platform is complex and the attitude changes frequently, thereby maintaining the spatial consistency and accuracy of geometric constraints throughout the entire processing.

[0063] Through the execution of the above steps, the attitude angles, position coordinates, and velocity vector data of the transmitting and receiving ends are fully utilized, achieving dynamic unification of spatial position and attitude information under a unified reference frame. By establishing a dynamic spatial mapping relationship and continuously updating the geometric baseline under a unified reference frame, the dual-base end-array airborne radar has a reliable spatial geometric constraint foundation under complex airborne operating conditions, providing a guarantee for the stability and accuracy of subsequent space-time signal processing from the source.

[0064] Step 2: Under the established geometric baseline constraints, perform spatiotemporal joint analysis on the echo signal obtained by the receiver, and couple and map the angular parameters of the echo signal with the Doppler frequency shift to generate a clutter spectrum with real geometric projection characteristics, which is used to characterize the distribution characteristics of clutter energy in the angular and frequency dimensions.

[0065] The clutter spectrum is generated by coupling and mapping the angular parameters of the echo signal with the Doppler frequency shift. The specific steps are as follows:

[0066] Under the constraint of the geometric baseline, the echo signals acquired by the receiver are precisely correlated in spatial direction. This step uses the reconstructed geometric baseline as a spatial reference and projects the signals of each channel of the receiver directionally according to the spatial direction of the geometric baseline. In specific implementation, the receiver determines the signal reception direction during flight by its installation position and the direction of the antenna array, with each receiving channel corresponding to a specific spatial reception angle. Since the transmitter and receiver are arranged separately and their positions and attitudes are constantly changing, the spatial incident direction of the received signal and the direction of the geometric baseline have a dynamic relationship. Therefore, it is necessary to correlate the reception angle parameter at each time point with the geometric baseline direction at that time. To achieve this correlation, the echo signal data recorded by the receiver at each moment must be matched with the coordinates of the starting and ending points of the geometric baseline at the corresponding moment. The spatial propagation direction from the transmitter to the receiver at that moment is determined by the vector difference of the spatial positions, and the spatial angle of the received signal is redefined based on this direction. In this way, the spatial parameters of the received signal no longer depend on the attitude reference of the receiving platform itself, but establish a spatial direction correlation with the transmitter through the geometric baseline, so that all received data are accurately located under a unified spatial geometry. Through this process, the spatial relationship between the angular properties of the echo signal and the geometric baseline is clearly established, laying the spatial direction foundation for subsequent time mapping and angle-frequency joint analysis.

[0067] After spatial correlation is established, the received signals are organized and matched in the time dimension based on the temporal variation characteristics of the geometric baseline, thereby achieving a continuous correspondence between spatial direction and temporal information. Since both the transmitter and receiver are in motion during actual flight, the length, direction, and rate of change of the geometric baseline all change over time. This change directly affects the time delay characteristics and energy distribution of the received signal. To ensure accurate correlation in the time dimension, the receiver maps the echo signal at each sampling moment to the length, direction, and relative motion state of the geometric baseline at that time. Specifically, under a unified reference frame, the received signal data is first grouped according to timestamps, so that each group corresponds to a specific geometric baseline state. Then, combining the velocity direction and attitude angle changes of the transmitter and receiver, the relative directional change trend of the signal propagation path at that moment is determined, and the time series of the signal is rearranged so that each segment of the signal in time can accurately reflect the continuous evolution of the spatial propagation path. In this way, the temporal variation characteristics of the received signal and the spatial direction changes of the geometric baseline achieve a one-to-one correspondence, thus maintaining the continuity of the geometric relationship on the time axis. At this point, the signal's temporal characteristics not only reflect the echo delay information but also, in a spatial geometric sense, the relative motion trajectories of the transmitter and receiver. This continuous matching allows subsequent analysis of the Doppler frequency shift to be performed under conditions of temporal and spatial consistency, ensuring that the frequency characteristics originate from genuine spatial propagation effects rather than platform motion errors.

[0068] After achieving unified mapping in both spatial and temporal dimensions, the angle parameters of the received signal are gradually coupled with the Doppler frequency shift based on the directionality and variation law of the geometric baseline, thereby generating a clutter spectrum with true geometric projection characteristics. In this process, the spatial direction of the geometric baseline in the unified reference frame is first used as the projection reference for the spectrum. Using this as a reference, the angle parameters of the received signal at each moment are projected onto the unified spatial plane, ensuring a continuous spatial distribution of echo signals sampled at different times. Next, based on the signal frequency variation trend obtained during time matching, the frequency shift in each spatial direction is correlated with the rate of change of the geometric baseline in the temporal dimension, giving each spatial direction its specific frequency distribution characteristics. Through this correspondence between angle and frequency, a two-dimensional energy distribution plane is constructed in the unified reference frame. The horizontal coordinate of each point in this two-dimensional energy distribution plane represents the spatial angle of the signal, the vertical coordinate represents the frequency shift of the signal, and the energy intensity at the corresponding position reflects the clutter distribution intensity in that direction. To ensure consistency between the clutter spectrum and the geometry, spatial projection is performed centered on the current geometric baseline direction at each time interval, ensuring that the directionality of the clutter spectrum remains consistent with the spatial relationship between the transmitter and receiver. When the platform undergoes attitude changes or velocity adjustments, the geometric baseline parameters are updated in real time, and the angle-frequency mapping is re-executed, enabling the generated clutter spectrum to continuously reflect the distribution of clutter energy in actual space. This clutter spectrum not only reveals the concentration of clutter energy at different angles but also reflects the energy distribution pattern at different Doppler frequencies, thus forming a clutter energy projection structure with real spatial significance. Through this clutter spectrum, the expansion characteristics of clutter in the angle and frequency dimensions can be clearly reflected.

[0069] The Doppler frequency shift distribution obtained by time-dimension matching is synchronously mapped to the corresponding spatial direction position. The specific implementation process is as follows:

[0070] First, in the preprocessing stage, the echo signal at each moment is time-matched with the corresponding geometric baseline state, ensuring that each Doppler frequency shift value has a clear time label. Simultaneously, the spatial orientation of the echo signal at that moment is determined by the spatial orientation of the geometric baseline in a unified reference frame. Then, using this spatial orientation as a mapping index, the Doppler frequency shift distribution extracted under the same time label is attached to the corresponding spatial orientation position. This ensures that frequency information no longer exists independently on the time axis but forms a one-to-one correspondence with the spatial orientation indicated by the geometric baseline. As the time series progresses, the Doppler frequency shift distribution at different time points is continuously mapped to its corresponding spatial orientation position, thus forming an energy structure that unfolds directionally and is distributed according to frequency in a unified spatial plane, achieving synchronous mapping of time-dimensional frequency shift information to the spatial orientation dimension.

[0071] This approach ensures that the Doppler frequency shift originates from the actual spatial propagation relationship, enabling the frequency characteristics to accurately reflect the clutter or target motion features in the corresponding spatial direction.

[0072] Through the above steps, under the spatial constraints of the geometric baseline, the echo signal obtained by the receiver is gradually mapped onto a unified spatial and temporal framework. Its angular parameters and Doppler frequency shift are organically combined geometrically, thereby generating a clutter spectrum that reflects the actual spatial propagation relationship. This clutter spectrum realistically reflects the distribution law of clutter energy in different angular and frequency directions, enabling the dual-baseline airborne radar to accurately describe the spatial distribution characteristics of clutter in complex motion environments. By completing the joint spatial-temporal analysis of the echo signal under real geometric constraints, the entire process maintains consistent spatial reference and temporal continuity in a physical sense, allowing the radar to maintain accurate understanding and stable processing of clutter distribution characteristics even under multi-platform collaboration, maneuvering flight, and complex terrain conditions.

[0073] Step 3: Based on the angle and frequency coupling distribution of the generated clutter spectrum, and according to the energy continuity characteristics between adjacent distance cells, perform rotational weighting control on the clutter spectrum to generate a space-time filtering weight vector that adaptively changes along the distance dimension, so that the space-time filtering direction continuously matches the local clutter structure, thereby maintaining the directional weakening of clutter energy.

[0074] The clutter spectrum is subjected to rotational weighting to generate a space-time filtering weight vector that adaptively varies along the range dimension. The specific steps are as follows:

[0075] After the clutter spectrum is generated, a detailed spatial analysis of its energy distribution in both angle and frequency dimensions is performed to clarify the energy concentration regions and energy continuity characteristics of the clutter at different distance cells. This process is conducted in a unified reference frame. The clutter spectrum formed by the receiver after acquiring the echo signal consists of a large number of spatial points representing energy intensity, each corresponding to a specific angle and frequency position. To accurately identify energy distribution characteristics, these energy points are spatially traversed and their directions are statistically analyzed. The energy variation trend within different angle ranges is recorded, and the main diffusion direction of clutter energy is determined based on the energy extension in the frequency dimension. During this process, by comparing the energy intensity distribution cell by cell, the connection relationship between adjacent distance cells in the main energy direction can be discovered. For example, if the energy distribution of two adjacent distance cells exists continuously within a similar angle range and the energy variation amplitude is small, it indicates that the spatial distribution of clutter in that region is continuous; conversely, if the energy concentration direction between adjacent distance cells changes abruptly, it indicates that the clutter structure at that location has undergone a directional abrupt change. By quantitatively recording these changes, the clutter spectrum can be divided into continuous and abrupt regions, providing a reference for subsequent rotation-weighted processing.

[0076] After clarifying the energy continuity characteristics of the clutter spectrum, a rotational weighted adjustment is performed on the clutter spectrum based on the changing patterns of energy distribution directions between distance cells, enabling a smooth spatial transition of energy directions in each distance cell. This process is also performed within a unified reference frame. Specifically, the rotation is based on the difference in principal energy directions between adjacent distance cells. When an angular shift in the principal energy directions of adjacent distance cells is detected, the spatial angular difference between the principal energy directions of the two distance cells is first calculated. The energy distribution of the latter distance cell is then adjusted spatially, gradually bringing its principal energy direction closer to that of the former distance cell. This rotation operation uses the geometric baseline as the axis of rotation and achieves directional adjustment of the clutter spectrum through spatial angular compensation, thereby maintaining the continuity of the clutter energy principal axis in space. For regions with significant abrupt energy differences, the direction is gradually adjusted according to the energy change gradient, allowing the energy distribution in that region to transition to the direction of adjacent regions within a certain range, thus avoiding discontinuities in the filtering direction caused by abrupt energy changes. During the rotation adjustment, the energy intensity is weighted and balanced according to the frequency distribution characteristics to maintain energy balance in the frequency dimension of the rotated clutter spectrum, avoiding energy shifts caused by excessive directional adjustment. Through this continuous rotational weighting process, the entire clutter spectrum forms a smooth energy transition structure from near to far in the spatial angular direction, providing a directional reference for subsequent spatiotemporal filtering direction extraction.

[0077] It should be noted that:

[0078] During the rotation adjustment process, in order to maintain energy balance in the frequency dimension of the rotated clutter spectrum, weighted compensation is performed on the energy intensity in different frequency ranges while performing directional adjustment.

[0079] The following examples will illustrate this point.

[0080] Suppose that at a certain moment, the main energy in the clutter spectrum is concentrated in the region with an angular direction of 15°–25°, corresponding to a Doppler frequency range of 150Hz–250Hz, with a peak energy of 0.92 (normalized value). However, after rotation adjustment, the main energy direction of adjacent range cells shifts to 30°–40°, with a frequency range of 200Hz–300Hz, and the peak energy drops to 0.75. To prevent excessive rotation from weakening the high-frequency energy range, during rotation weighting, the energy weighting coefficient for this cell in the 200Hz–300Hz frequency range is set to 1.2, while the weighting coefficient in the low-frequency range of 100Hz–200Hz is set to 0.8. This differentiated weighting ensures that the total energy of the rotated spectrum across different frequency ranges remains close to that of the original spectrum, thus achieving a balanced energy distribution along the frequency dimension.

[0081] In this way, the clutter spectrum after rotation adjustment not only achieves spatial continuity with adjacent distance cells in the angular direction, but also maintains the balance of energy intensity in the frequency direction, avoiding energy shift caused by excessive directional adjustment.

[0082] After performing rotational weighting of the clutter spectrum, a space-time filtering weight vector that adaptively varies along the range dimension is extracted based on the adjusted energy distribution structure. This step uses the rotated clutter spectrum as input, determines the dominant energy direction of the clutter on a range-cell basis, and continuously matches this direction information with adjacent range cells to form a set of range-varying direction vectors. In the specific implementation, firstly, the region with the highest energy intensity is found on the clutter spectrum of each range cell; the angular direction of this region is the dominant energy direction of that range cell. Subsequently, this dominant energy direction is compared with the dominant energy directions of adjacent range cells. If the two directions are close, they are smoothly connected; if the directions differ significantly, a transition direction is inserted in the middle direction according to the energy distribution trend, so that the space-time filtering weight vector forms a smooth variation curve in space. In this way, a continuously varying space-time filtering weight vector trajectory is obtained along the entire range dimension. This trajectory can automatically adjust its direction according to the clutter distribution in different range cells, ensuring that the filter remains consistent with the local clutter structure in space. Furthermore, based on this trajectory, the spatiotemporal filter weight vectors are arranged in spatial order to form a sequence of spatiotemporal filter weight vectors covering the entire range. This sequence can dynamically adapt to changes in the directionality of clutter, ensuring that the filtering direction always corresponds to the local clutter distribution in different spatial regions, thus laying the foundation for continuous suppression.

[0083] Finally, the space-time filtering weight vector generated along the range dimension is directionally matched with the rotated weighted clutter spectrum and applied to the echo signal processing under a unified reference frame to achieve directional attenuation of clutter energy. Specifically, within the clutter spectrum region corresponding to each range cell, the space-time filtering weight vector of that range cell is applied to its clutter energy distribution direction, causing the filtering response to form an energy attenuation band along the main energy direction. Since the space-time filtering weight vectors of adjacent range cells maintain continuity in the spatial direction, the filtering response forms a smooth directional attenuation structure across the entire range, gradually weakening clutter energy in space without abrupt changes. Simultaneously, by maintaining the consistency between the geometric baseline and the filtering direction under the unified reference frame, the directionality of the filtering response always corresponds to the spatial geometric distribution of the transmitter and receiver, ensuring that the filtering output truly reflects the attenuation of clutter energy in space. As the platform attitude or geometric baseline changes, the space-time filtering weight vector can be dynamically updated, allowing the filtering direction to follow the changes in the main clutter direction in real time, thus maintaining effective clutter suppression throughout the entire radar scan process. Ultimately, through this continuous filtering process, clutter energy is effectively weakened in the spatial direction, background interference is significantly reduced, while the target signal is preserved and highlighted in the non-clutter direction, providing a clean input signal for subsequent sparse constraint processing and energy fusion.

[0084] By executing the above steps, based on the angular and frequency coupling distribution characteristics of the clutter spectrum under a unified reference frame, and combined with the energy continuity characteristics between adjacent range cells, a complete process from energy direction identification, clutter spectrum rotation weighting, spatiotemporal filtering weight vector extraction to filtering direction application is achieved. The entire process maintains the spatial continuity of the clutter energy distribution and the consistency of the filtering direction, enabling the spatiotemporal filtering weight vector to adaptively change along the range dimension, continuously matching the local clutter structure, thereby achieving directional attenuation of clutter energy under complex spatial geometry conditions.

[0085] Step 4: Using the generated spatiotemporal filtering weight vector, apply local sparsity constraints, set weights based on the clutter spectral similarity between adjacent distance cells, dynamically suppress noise energy far from the main spectrum direction, and perform selective updates in clutter characteristic abrupt regions to ensure the continuity and stability of the filtering output in the distance dimension.

[0086] Using the filtered output results corresponding to the generated space-time filter weight vector, local sparsity constraints are applied. The specific steps are as follows:

[0087] After obtaining the filtered output corresponding to the space-time filtering weight vector, a detailed energy distribution extraction is performed on the filtered output signal along the distance dimension to identify the initial spatial distribution of clutter and noise energy. Specifically, each distance unit of the filtered output signal corresponds to a spatial location range. Then, the signal energy in each distance unit is calculated point by point, and its energy amplitude and directional characteristics are recorded, resulting in a series of spatial curves reflecting the energy distribution. By analyzing these spatial curves, the variation in energy intensity at different distances can be observed. Typically, clutter energy is concentrated in areas aligned with the geometric baselines of the transmitter and receiver, while noise energy exhibits directional dispersion and significant intensity fluctuations. To ensure the targeted nature of subsequent processing, these energy data are smoothed and organized, with regions of strong energy continuity designated as the main spectrum energy band, and regions with frequent energy fluctuations designated as noise energy bands. This energy distribution extraction process provides the basic data for subsequent weight setting and constraint application, enabling targeted and structured adjustments to the energy distribution in subsequent operations.

[0088] After the energy distribution is extracted and classified, the clutter spectra between adjacent distance cells are compared to calculate the similarity of their energy structures, thereby determining the weight distribution required for sparsity constraints. This step uses the previously generated clutter spectra as a reference, comparing the energy direction, main peak position, energy spread range, and frequency distribution pattern of adjacent distance cells item by item. If the main energy directions of two distance cells almost overlap and the shape and range of their energy concentration regions are similar, they are considered to belong to the same continuous clutter structure in space; if the main energy direction shifts or the energy peak intensity differs significantly, it indicates that their clutter characteristics are different. Based on this comparison, a similarity weight is assigned to each pair of adjacent distance cells. Regions with high similarity are assigned higher weights, representing that their energy structure needs to remain continuous; regions with low similarity are assigned lower weights, representing that their energy distribution can be moderately adjusted. In this way, a continuous similarity weight curve is formed across the entire distance dimension. The similarity weight curve reflects the spatial consistency of the clutter spectrum in a physical sense, and also provides a directional basis for subsequent energy suppression and constraint application, so that the sparse constraint no longer simply acts on a single distance cell, but can be differentiated according to the structural relationship between adjacent distance cells.

[0089] After determining the similarity weight distribution, a local sparsity constraint is applied to the filtered output based on this similarity weight to dynamically suppress noise energy far from the main spectrum direction. This step is performed in a unified reference frame, where the energy distribution of the filtered output is mapped one-to-one with the corresponding similarity weight, giving each distance cell its own independent energy control factor. When the energy direction of a distance cell deviates significantly from the main spectrum direction, the energy response intensity in that direction is automatically reduced based on its low similarity weight, thus spatially weakening the contribution of noise energy. Conversely, when the energy direction of a distance cell is consistent with the main spectrum direction and has a high weight, the energy distribution in that direction remains unchanged to protect the spatial integrity of the clutter main structure. The core of this local sparsity constraint lies in using energy weighting to keep the energy distribution of the filtered output concentrated in the main spectrum direction while gently suppressing areas where energy is dispersed. After this process, the filtered output signal exhibits a spatial energy distribution pattern that gradually decays outward from the main spectrum direction, effectively suppressing the diffusion of noise energy far from the main spectrum direction and enhancing the prominence of the main energy direction.

[0090] After local sparsity constraint processing, a selective update operation is performed on regions where clutter characteristics exhibit abrupt spatial changes to avoid excessive attenuation of the main spectrum energy in these areas. This step primarily focuses on distance ranges where clutter characteristics change drastically, such as at terrain boundary changes, areas containing highly reflective objects, or energy jump locations caused by rapid changes in platform attitude. In these regions, the energy structure similarity of adjacent distance cells decreases sharply. Continuing to use the global sparsity constraint parameters might lead to undue suppression of the main spectrum energy. To avoid this problem, the strength of the sparsity constraint is adjusted by recalculating the main spectrum direction and energy peak position within the region. Specifically, the constraint strength is relaxed along the main spectrum direction to maintain the natural distribution of clutter energy; while stronger constraints are applied to noise energy regions far from the main spectrum direction to ensure energy suppression. In this way, the energy distribution in abrupt regions achieves both preservation of the main spectrum and suppression of noise, maintaining the spatial coherence of the overall energy distribution while ensuring the authenticity of the local structure. This step prevents energy discontinuities in the filtered output at spatial abrupt points, maintaining continuous and stable energy changes along the distance dimension.

[0091] After selective update operations, all filtered outputs processed with local sparsity constraints are integrated to ensure energy continuity and spatial stability across the entire range dimension. This integration process is performed in a unified reference frame, rearranging the outputs of all range cells in spatial order and smoothing the energy boundaries between adjacent range cells to create a natural spatial connection in the energy distribution of each range cell. During integration, referencing previously established similarity weight curves, different degrees of smoothing are applied to areas of energy abrupt change and high-weight continuum regions, ensuring both overall output continuity and preservation of local structural features. The final filtered output exhibits a continuous energy distribution along the range dimension, with concentrated and coherent energy in the main spectrum direction. Noise energy away from the main spectrum direction is effectively weakened, resulting in a stable overall energy structure with spatial consistency. This result provides an accurate, smooth, and stable input basis for subsequent geometric phase correction and energy aggregation of multi-channel received signals, enabling radar signal processing to maintain continuous energy performance and reliable spatial constraints under complex maneuvering platform conditions.

[0092] Through the implementation of the above steps, under a unified reference frame, local sparsity constraints are applied using the filtered output results corresponding to the spatiotemporal filtering weight vector. Weights are set based on the clutter spectral similarity between adjacent distance cells, achieving dynamic suppression of noise energy and preservation of the main spectrum energy structure. Selective updates and result integration ensure the continuity and stability of the filtered output results in the distance dimension. The entire process achieves adaptive adjustment of the local energy structure under spatial geometric constraints, enabling the filtered results to accurately reflect the physical characteristics of the clutter distribution.

[0093] Step 5: Based on the filtered output after sparse constraint processing, geometric phase correction and energy aggregation are performed on the multi-channel received signals. Coherent superposition of the multi-channel received signals is achieved under a unified reference system. Through directional energy enhancement control, the target echo is effectively highlighted in the fused signal, thereby achieving prominent detection of weak targets and deep suppression of background clutter.

[0094] Based on the filtered output after sparse constraint processing, geometric phase correction and energy aggregation are performed on the multi-channel received signals. The specific steps are as follows:

[0095] Geometric phase correction is performed on the filtered output after sparse constraint processing of the multi-channel received signals. The specific implementation process is as follows: By accurately describing the spatial geometric relationship between the transmitter and each receiver, the phase difference of each channel signal along its propagation path is determined. Since the antennas of each receiving channel are not perfectly uniform in space, and the transmitter and receiver continuously undergo relative displacement during flight, the path length of each channel signal reaching the receiver is different, resulting in a slight phase shift in the echo signal. Directly superimposing these uncorrected signals can easily cause energy cancellation or directional shift due to phase mismatch, thereby weakening the target energy. To solve this problem, the propagation path length of each channel is first determined based on the real-time spatial coordinate data of the transmitter and receiver, and the phase shift compensation required for each channel is calculated based on the path difference. Then, the phase of the received signal for each channel is corrected time-slice by time slice, ensuring that the wavefronts of all channel signals maintain consistent direction under a unified reference frame, thus ensuring spatial coherence of the signals. Through this process, the phases of all receiving channels are unified, forming a spatially geometrically aligned set of multi-channel signals, laying an accurate spatial foundation for subsequent energy balancing and coherent superposition.

[0096] After geometric phase correction, the corrected multi-channel received signals undergo energy amplitude equalization and continuous matching to eliminate energy inconsistencies between channels caused by differences in receiver gain, platform attitude, or spatial orientation. The specific process is as follows: First, the energy response curves of each receiving channel within the same range cell are analyzed, comparing the peak energy differences of each channel in the target direction and clutter direction. When some channels are found to have higher or lower energy responses, the signal amplitude is adjusted based on their relative deviations, making the energy distribution of all channels at the same spatial location more consistent. To ensure the continuity of energy matching, this adjustment is not only applied to a single range cell but also smooths the energy response curves across the entire range, resulting in a gradual transition of energy changes in space. Simultaneously, considering the attitude changes generated by the flight platform during maneuvering, the energy adjustment is synchronized in time by continuously acquiring attitude angle and position change information from both the transmitter and receiver, ensuring that energy equalization remains consistent at different time points. Through this continuous matching process, all channel signals are corrected to the same reference level in terms of energy amplitude, maintaining a balance in the energy contribution of subsequent signal superposition and avoiding fusion distortion caused by energy deviations in individual channels.

[0097] After phase correction and energy equalization, all channel signals are coherently superimposed to achieve energy aggregation and target signal enhancement. The specific process is as follows: First, based on the geometric baseline directions of the transmitter and receiver, the signals of each channel are aligned in the same spatial direction, allowing echo signals in the same target direction to spatially overlap. Since the phase has already been corrected, the wavefronts of all channel signals remain consistent in space. Therefore, during superposition, the main wave components of each channel signal can achieve complete coherent superposition, while noise and clutter are partially canceled out due to directional differences and phase inconsistencies. In the actual superposition process, the signals of each channel are weighted and summed according to their energy weights, resulting in more uniform energy aggregation. As the number of channels increases, the energy in the target direction rapidly increases after superposition, while clutter energy attenuates due to phase inconsistencies. The resulting superimposed signal forms a significant energy concentration area in the target direction, while the energy density decreases in other directions, thus achieving spatial enhancement of the target signal and initial suppression of background clutter. After this coherent superposition process, the radar received signal is transformed from the original dispersed channel signals into a highly concentrated fused signal with higher signal-to-noise ratio and directional resolution.

[0098] After obtaining the fused signal through coherent superposition, directional energy enhancement control is applied to the fused signal to further enhance the prominence of the target echo and achieve deep suppression of background clutter. The specific operation process is as follows: First, the energy distribution characteristics of the fused signal in the angular direction are analyzed to identify the energy gradient between the energy peak in the target direction and the surrounding clutter energy region. To make the energy in the target direction more concentrated and prominent, energy enhancement control is applied in the target direction to increase the energy response amplitude in that direction. Simultaneously, energy suppression is applied in the clutter direction to further weaken the clutter energy in the fused signal. During the energy enhancement control process, the geometric baseline direction and target angle parameters are combined to ensure that the energy enhancement always proceeds along a direction consistent with the spatial relationship between the transmitter and receiver, thus guaranteeing the geometric consistency of the spatial distribution. At the same time, to maintain the continuity of the energy distribution, a smoothing buffer is set in the transition region between energy enhancement and suppression, allowing the target energy to gradually transition from the main direction to the periphery, preventing abrupt changes or energy gaps. Through this directional energy enhancement control, the target echo energy is further concentrated in the fused signal, while clutter energy is weakened in spatial distribution, forming a clear energy contrast that enables small targets to be reliably identified and detected in the background. The final output fused signal maintains the directional characteristics consistent with the bistatic configuration in spatial geometry and exhibits a stable structure with concentrated energy at the center and attenuated energy at the periphery, providing a high-quality signal foundation for subsequent target localization and identification.

[0099] It should be noted that the specific operations of the above steps are all carried out under a unified reference frame.

[0100] Through the above steps, geometric phase correction, energy matching, coherent superposition, and directional energy enhancement control were performed on the multi-channel received signals under a unified reference frame. The entire process maintained phase alignment and energy continuity under spatial geometric constraints, enabling the fused multi-channel received signals to form a unified output with converged energy. Geometric phase correction achieved spatial directionality, energy equalization eliminated amplitude differences between channels, coherent superposition achieved concentrated signal energy enhancement, and directional energy control achieved target energy prominence and deep suppression of clutter energy. This effectively improved the ability of the dual-base-array airborne radar to detect weak targets in complex environments and provided a reliable energy foundation and spatial reference for subsequent high-resolution imaging and precise positioning.

[0101] This invention reconstructs the time-varying geometric baseline between the transmitter and receiver in a unified reference frame, ensuring that the spatial characteristics of the echo signal are always guided by spatial constraints. This fundamentally alleviates the non-stationary problem caused by clutter variations with range or elevation angle in dual-baseline airborne radars. The clutter spectrum generated under geometric baseline constraints accurately reflects the clutter distribution in the angular and frequency dimensions, ensuring that the spacetime filtering direction remains consistent with the local clutter structure. This maintains stable clutter suppression even under complex maneuvering conditions, improving the radar's background suppression capability in asymmetric spatial configurations.

[0102] This invention introduces local sparsity constraints and selective update mechanisms on the basis of spatiotemporal filtering output, and combines geometric phase correction and coherent energy aggregation of multi-channel received signals to maintain a continuous and stable energy distribution in the range dimension of the processing result. Through directional energy enhancement control, the target echo is concentrated and enhanced in the fused signal, while background clutter and noise energy are suppressed in non-dominant directional regions. This improves the discriminability of weak targets in complex clutter environments, providing a clearer and more reliable signal foundation for subsequent target detection and localization.

[0103] This invention provides, for example Figure 4 The dual-base-end array airborne radar clutter suppression system shown includes a geometric baseline construction module, a space-time spectrum generation module, an adaptive filter weight generation module, a local sparse constraint module, and a multi-channel coherent fusion module.

[0104] Geometric baseline construction module: acquires attitude angles, position coordinates and velocity vector data of the transmitter and receiver, establishes dynamic spatial mapping relationship between the transmitter and receiver under a unified reference frame, and reconstructs the geometric baseline between the transmitter and receiver according to the time series;

[0105] Space-time spectrum generation module: Under the constraints of the established geometric baseline, the module performs joint space-time analysis on the echo signal obtained by the receiver, and couples and maps the angular parameters of the echo signal with the Doppler frequency shift to generate clutter spectrum.

[0106] Adaptive filter weight generation module: Based on the angle and frequency coupling distribution of the clutter spectrum, and according to the energy continuity characteristics between adjacent distance cells, the clutter spectrum is subjected to rotational weighting control to generate a space-time filter weight vector that adaptively varies along the distance dimension.

[0107] Local sparsity constraint module: Utilizes the filtering output corresponding to the space-time filtering weight vector to apply local sparsity constraint processing, sets weights based on the clutter spectral similarity between adjacent distance cells, and dynamically suppresses noise energy far from the main spectrum direction;

[0108] Multi-channel coherent fusion module: Based on the filtered output results after sparse constraint processing, geometric phase correction and energy aggregation are performed on the multi-channel received signals, and coherent superposition of the multi-channel received signals is realized under a unified reference frame.

[0109] The clutter suppression method for dual-base-array airborne radar provided in this embodiment of the invention is implemented through the aforementioned dual-base-array airborne radar clutter suppression system. For details of the specific methods and procedures of the dual-base-array airborne radar clutter suppression system, please refer to the embodiments of the aforementioned dual-base-array airborne radar clutter suppression method, which will not be repeated here.

[0110] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A clutter suppression method for a dual-base-array airborne radar, characterized in that, Includes the following steps: Step 1: Obtain the attitude angles, position coordinates, and velocity vector data of the transmitter and receiver, establish the dynamic spatial mapping relationship between the transmitter and receiver under a unified reference frame, and reconstruct the geometric baseline between the transmitter and receiver according to the time series. Step 2: Under the established geometric baseline constraints, perform spatiotemporal joint analysis on the echo signal obtained at the receiving end, and couple and map the angular parameters of the echo signal with the Doppler frequency shift to generate clutter spectrum. Step 3: Based on the angle and frequency coupling distribution of the clutter spectrum, and according to the energy continuity characteristics between adjacent distance cells, perform rotational weighting control on the clutter spectrum to generate a space-time filtering weight vector that adaptively varies along the distance dimension; Step 4: Using the filtering output results corresponding to the space-time filtering weight vector, apply local sparsity constraint processing, set weights based on the clutter spectral similarity between adjacent distance cells, and dynamically suppress noise energy far from the main spectrum direction; Step 5: Based on the filtered output after sparse constraint processing, perform geometric phase correction and energy aggregation on the multi-channel received signals to achieve coherent superposition of the multi-channel received signals under a unified reference frame.

2. The clutter suppression method for dual-base-end array airborne radar according to claim 1, characterized in that, Reconstructing the geometric baseline between the transmitter and receiver based on time series data includes the following steps: The spatial state information of the transmitter and receiver is collected to obtain the attitude angle, position coordinates and velocity vector data of the transmitter and receiver under a unified time reference, and stored in the form of time series. Based on attitude angle, position coordinates and velocity vector data, coordinate transformation and geometric registration of the spatial state of the transmitter and receiver are performed in a unified reference frame. In a unified reference frame, the spatial position difference vector between the transmitter and receiver is calculated point by point based on time series data to form a spatial line segment connecting the transmitter and receiver, and a geometric baseline that changes continuously with time is constructed using the spatial line segment. Based on the temporal variation characteristics of the geometric baseline, the geometric baseline is introduced as a spatial constraint into the signal processing process to describe the spatial propagation relationship between the transmitter and receiver.

3. The clutter suppression method for dual-base-end array airborne radar according to claim 2, characterized in that, The clutter spectrum is generated by coupling and mapping the angular parameters of the echo signal with the Doppler frequency shift, including the following steps: Under the established geometric baseline constraints, the spatial orientation of the echo signals acquired by the receiver is correlated, and the angle parameters of the echo signals of each channel are matched with the geometric baseline orientation at the corresponding time. Based on the spatial orientation correlation results, the echo signals are organized and matched in the time dimension according to the temporal variation characteristics of the geometric baseline, so that the temporal characteristics of the echo signals in each time period are continuously correlated with the changes in the direction and length of the geometric baseline. Based on the unified mapping of spatial direction and time dimension, and according to the directionality and variation law of geometric baseline, the angular parameters of the echo signal and the Doppler frequency shift are coupled and mapped, and a clutter spectrum reflecting the clutter energy distribution characteristics in the angular and frequency dimensions is constructed in a unified reference frame.

4. The clutter suppression method for dual-base-end array airborne radar according to claim 3, characterized in that, In the process of constructing the clutter spectrum, the spatial direction of the geometric baseline in the unified reference system is used as the projection reference of the spectrum. The angular parameters of the echo signals in each time period are projected to the corresponding spatial direction positions, and the Doppler frequency shift distribution obtained by time dimension matching is synchronously mapped to the corresponding spatial direction positions.

5. The clutter suppression method for dual-base-end array airborne radar according to claim 3, characterized in that, Performing rotational weighting on the clutter spectrum to generate a space-time filtering weight vector that adaptively varies along the range dimension includes the following steps: Energy distribution analysis is performed on the generated clutter spectrum under a unified reference frame to extract the energy concentration regions of each distance cell in the angular and frequency dimensions, and to determine the energy continuity characteristics between adjacent distance cells. Based on the energy continuity characteristic, the clutter spectrum of adjacent distance cells is subjected to rotational weighting control according to the relationship of their main energy direction variation; Based on the clutter spectrum after rotational weighting, the main direction of clutter energy is extracted in each range cell, and combined with the relationship of the main directions of adjacent range cells, a space-time filtering weight vector that varies continuously along the range dimension is generated. The space-time filtering weight vector is directionally matched with the clutter spectrum of the corresponding range cell in a unified reference frame and then applied to echo signal processing.

6. The clutter suppression method for dual-base-end array airborne radar according to claim 5, characterized in that, When generating the space-time filtering weight vector that varies continuously along the distance dimension, the principal energy direction of the clutter spectrum corresponding to each distance cell is used as the direction reference of the weight vector, and the space-time filtering weight vector is continuously connected according to the energy distribution variation relationship of the clutter spectrum between adjacent distance cells.

7. The clutter suppression method for dual-base-end array airborne radar according to claim 5, characterized in that, Using the filtering output corresponding to the space-time filtering weight vector, local sparsity constraints are applied, including the following steps: Based on the filtering output results corresponding to the space-time filtering weight vector, the energy distribution of each distance unit is extracted along the distance dimension to distinguish the energy band in the main spectrum direction from the noise energy band. Based on energy distribution, the clutter spectral structure of adjacent distance cells is compared to determine the clutter spectral similarity between adjacent distance cells and similarity weights are set. Based on similarity weights, local sparsity constraints are applied to the filtered output to dynamically suppress noise energy far from the main spectrum direction; Based on the changes in similarity weights, selective updates are performed within the distance cells where clutter spectral similarity abruptly changes, adjusting the local sparsity constraint strength. Based on local sparse constraints and selective update results, the filtering output results of each distance cell are integrated.

8. The clutter suppression method for dual-base-end array airborne radar according to claim 7, characterized in that, Selective updates are performed within the distance cells where the clutter spectral similarity abruptly changes. This includes adjusting the scope of local sparsity constraints based on the relationship between the main spectral direction of the distance cell and the main spectral direction of adjacent distance cells.

9. The clutter suppression method for dual-base-end array airborne radar according to claim 7, characterized in that, Based on the filtered output after sparse constraint processing, geometric phase correction and energy aggregation are performed on the multi-channel received signals, including the following steps: Based on the filtered output results after sparse constraint processing, and combined with the spatial geometric relationship between the transmitter and receiver, geometric phase correction is performed on the multi-channel received signal. Based on the multi-channel received signals that have undergone geometric phase correction, the energy amplitude of the received signals of each channel is equalized and continuously matched. Based on the multi-channel received signals that have achieved energy equalization, the signals of each channel are coherently superimposed according to the geometric baseline direction under a unified reference frame; Directional energy enhancement control is applied to the fused signal based on the coherent superposition.

10. A dual-base-end airborne radar clutter suppression system, used to implement the dual-base-end airborne radar clutter suppression method according to any one of claims 1-9, characterized in that, It includes a geometric baseline construction module, a spatiotemporal spectrum generation module, an adaptive filter weight generation module, a local sparse constraint module, and a multi-channel coherent fusion module; Geometric baseline construction module: acquires attitude angles, position coordinates and velocity vector data of the transmitter and receiver, establishes dynamic spatial mapping relationship between the transmitter and receiver under a unified reference frame, and reconstructs the geometric baseline between the transmitter and receiver according to the time series; Space-time spectrum generation module: Under the constraints of the established geometric baseline, the module performs joint space-time analysis on the echo signal obtained by the receiver, and couples and maps the angular parameters of the echo signal with the Doppler frequency shift to generate clutter spectrum. Adaptive filter weight generation module: Based on the angle and frequency coupling distribution of the clutter spectrum, and according to the energy continuity characteristics between adjacent distance cells, the clutter spectrum is subjected to rotational weighting control to generate a space-time filter weight vector that adaptively varies along the distance dimension. Local sparsity constraint module: Utilizes the filtering output corresponding to the space-time filtering weight vector to apply local sparsity constraint processing, sets weights based on the clutter spectral similarity between adjacent distance cells, and dynamically suppresses noise energy far from the main spectrum direction; Multi-channel coherent fusion module: Based on the filtered output results after sparse constraint processing, geometric phase correction and energy aggregation are performed on the multi-channel received signals, and coherent superposition of the multi-channel received signals is realized under a unified reference frame.

Images