A real-time estimation method and system for radar signal-to-noise ratio of small targets

Abstract

The application provides a real-time estimation method and system for the signal-to-noise ratio of a small target radar. The application establishes a motion model based on the historical trajectory of a maneuvering target to generate a dynamic position prediction sequence. The scanning beam pointing angle sequence and the timestamp echo signal strength sequence are collected. For each target position point in the prediction sequence, multiple rounds of iterative detection are started, and the detection window is contracted bidirectionally until the boundary is smaller than the beam half-power point width. The signal strength peak value and the background noise intensity in the final iteration window are extracted, and the beam irradiation efficiency factor is generated by combining the spatial matching degree of the scanning beam pointing angle sequence and the position prediction sequence. The signal-to-noise ratio dynamic correction amount is calculated cooperatively. The correction amount, the signal peak value, and the noise intensity are weighted and fused to output the final signal-to-noise ratio value of the maneuvering target in real time. The application improves the real-time estimation accuracy and stability of the signal-to-noise ratio of a small target radar when a high-speed maneuvering target crosses the beam edge.

Description

Technical Field

[0001] This application relates to the field of signal-to-noise ratio estimation technology, and in particular to a real-time estimation method and system for adaptive signal-to-noise ratio of small target radar. Background Technology

[0002] In low-altitude small target surveillance scenarios under complex electromagnetic environments, radar systems need to achieve continuous and stable signal-to-noise ratio estimation for high-speed maneuvering targets. These targets are characterized by strong trajectory abrupt changes and small radar cross-sections. Under the constraint of extremely short beam dwell time of mechanically scanned radar, traditional methods struggle to maintain estimation accuracy when the target crosses the beam edge. It is urgent to solve the problem of core parameter jumps caused by the mismatch between target maneuverability and scanning timing.

[0003] The current mainstream solution adopts a detection mechanism that combines dynamic gate adjustment with historical trajectory extrapolation: a linear prediction model is established based on the motion state of the target in the previous scanning cycle, and an adaptive detection gate is set according to the predicted position during radar scanning. By comparing the statistical difference between the echo signal strength and the background noise in real time, the signal-to-noise ratio output value is dynamically corrected in combination with a preset threshold.

[0004] The scheme has significant limitations when the target suddenly maneuvers: the linear prediction model cannot adapt to nonlinear changes in the trajectory, resulting in detection gate positioning deviation; the preset threshold is difficult to match the changes in beam shape under different scanning elevation angles, causing the beam edge region correction to fail; the signal-to-noise ratio correction process does not consider the spatiotemporal coupling relationship between the actual beam center and the target position, and the final output value exhibits systematic distortion during the target maneuvering and turning phase. Summary of the Invention

[0005] This application provides a real-time estimation method and system for adaptive signal-to-noise ratio of small target radar, which solves the problems of insufficient accuracy and stability of real-time estimation of signal-to-noise ratio of small target radar in the prior art.

[0006] Firstly, this application provides a real-time estimation method for the adaptive signal-to-noise ratio of small target radar, including:

[0007] During the intervals between radar scans, a motion characteristic model is established based on the continuous historical trajectory points of the maneuvering target, and a dynamic position prediction sequence for the target in the next scan cycle is generated.

[0008] A mechanical rotating scanning device acquires radar scanning data packets at a fixed rotation speed, and synchronously records the scanning beam pointing angle sequence and the echo signal strength sequence with the corresponding timestamp. The scanning beam pointing angle sequence includes the azimuth angle and elevation angle of the beam center axis.

[0009] For each target location point in the dynamic position prediction sequence, multiple rounds of iterative detection are initiated within its spatial neighborhood. The first round of detection uses an initial detection window that covers the target's maximum maneuver range. In subsequent rounds, based on the distribution characteristics of the echo signal intensity sequence within the previous round's detection window, the detection window range is shrunk bidirectionally along the range and azimuth directions until the detection window boundary is smaller than the half-power point width of the scanning beam.

[0010] Extract the peak signal intensity and corresponding background noise intensity in the final iteration round, and simultaneously calculate the spatial matching degree between the scanning beam pointing angle sequence and the dynamic position prediction sequence. Convert the spatial matching degree into a beam illumination efficiency factor, and combine the ratio of the peak signal intensity and the corresponding background noise intensity to collaboratively calculate the dynamic correction amount of the signal-to-noise ratio.

[0011] The final signal-to-noise ratio (SNR) value of the moving target within the current scanning cycle is output in real time by weighted fusion based on the dynamic correction amount of the SNR, the peak signal strength, and the corresponding background noise intensity.

[0012] Optionally, for each target location point in the dynamic position prediction sequence, multiple rounds of iterative detection are initiated within its spatial neighborhood. The first round of detection uses an initial detection window covering the target's maximum maneuver range. Subsequent rounds, based on the distribution characteristics of the echo signal intensity sequence within the previous round's detection window, shrink the detection window range bidirectionally along the range and azimuth directions until the detection window boundary is smaller than the half-power point width of the scanning beam. This includes:

[0013] Using each target location point in the dynamic position prediction sequence as the center point, a square initial detection window is formed by expanding along the range and azimuth directions. The side length of the square initial detection window is set to cover the maximum maneuver offset range of the target.

[0014] Within the boundary of the square initial detection window, the intensity distribution of the echo signal intensity sequence is analyzed. When the intensity peak point of the intensity distribution is detected to be located in the edge region of the square initial detection window, the boundary range is proportionally shrunk along the direction of the intensity peak point.

[0015] Repeat the detection window shrinking operation until the boundary range of the initial square detection window is smaller than the half-power point width of the scanning beam, and the intensity peak point is located in the central region of the initial square detection window, at which point the iteration termination condition is satisfied.

[0016] Optionally, the step of extracting the peak signal intensity and corresponding background noise intensity in the final iteration, simultaneously calculating the spatial matching degree between the scanning beam pointing angle sequence and the dynamic position prediction sequence, converting the spatial matching degree into a beam illumination efficiency factor, and collaboratively calculating the signal-to-noise ratio dynamic correction amount by combining the ratio of the peak signal intensity and the corresponding background noise intensity, includes:

[0017] The difference between the coordinates of the center axis pointing point of the scanning beam and the predicted coordinates of the target's dynamic position under the same time mark is calculated to obtain the beam pointing position deviation.

[0018] A beam illumination efficiency factor is generated by mapping the magnitude of the beam pointing position deviation, wherein the mapping relationship satisfies that the beam illumination efficiency factor monotonically decreases as the beam pointing position deviation increases.

[0019] The original signal-to-noise ratio (SNR) value is obtained by dividing the peak signal intensity in the final iteration by the corresponding background noise intensity, and then multiplying it by the beam illumination efficiency factor to generate the dynamic SNR correction.

[0020] Optionally, the step of acquiring radar scan data packets at a fixed rotation speed using a mechanical rotating scanning device, and simultaneously recording the scan beam pointing angle sequence and the echo signal intensity sequence corresponding to the timestamp, wherein the scan beam pointing angle sequence includes the azimuth and elevation angles of the beam center axis, including:

[0021] During the constant angular velocity rotation of the mechanical rotating scanning device, whenever the azimuth angle of the scanning beam center axis reaches the preset angle sampling position, the radar transmitting unit is triggered to send a detection pulse signal and simultaneously receive the target reflection signal.

[0022] The target reflection signal sampled at each azimuth angle is divided into time dimensions, and the echo intensity values ​​of all range cells within the same time slice are combined into two-dimensional distribution data of azimuth and range.

[0023] The maximum signal intensity and its corresponding distance unit coordinates within the coverage area of ​​the main lobe of the scanning beam are extracted from the two-dimensional distribution data of azimuth and distance. An echo signal intensity sequence is generated in chronological order according to the timestamps. At the same time, the azimuth and elevation angle values ​​of the scanning beam center axis at each sampling point are recorded to form a scanning beam pointing angle sequence.

[0024] Optionally, the step of weighted fusion based on the dynamic signal-to-noise ratio correction, the peak signal strength, and the corresponding background noise intensity to output the final signal-to-noise ratio value of the moving target in the current scanning period in real time includes:

[0025] A first fusion weight value is assigned to the peak signal strength, a second fusion weight value is assigned to the corresponding background noise strength, and a third fusion weight value is assigned to the dynamic correction amount of the signal-to-noise ratio. The sum of all fusion weight values ​​is a constant.

[0026] The signal strength peak value is multiplied by the first fusion weight value to form the main component, the corresponding background noise intensity is multiplied by the second fusion weight value to form the noise component, and the signal-to-noise ratio dynamic correction amount is multiplied by the third fusion weight value to form the correction component;

[0027] The main component, the noise component, and the correction component are superimposed to calculate the final signal-to-noise ratio of the maneuvering target in the current scanning cycle.

[0028] Optionally, the step of analyzing the intensity distribution of the echo signal intensity sequence within the boundary range of the square initial detection window, and when the intensity peak point of the intensity distribution is detected to be located in the edge region of the square initial detection window, proportionally shrinking the boundary range along the direction of the intensity peak point, includes:

[0029] An edge region is defined within the initial square detection window, and the edge region is defined as a narrow side region adjacent to the boundary of the detection window;

[0030] Locate the spatial position point corresponding to the maximum signal intensity within the current detection window range in the echo signal intensity sequence. If the spatial position point falls into the narrow edge region, it is determined that the intensity peak point is in an edge position state.

[0031] Along the line connecting the intensity peak point at the edge position and the center point of the detection window, the boundary of the detection window is moved towards the center point of the detection window by a preset shrinkage distance to form a new shrinkage detection window boundary range.

[0032] Optionally, the step of calculating the difference between the coordinates of the scanning beam center axis pointing point and the predicted coordinates of the target dynamic position under the same time mark to obtain the beam pointing position deviation includes:

[0033] The azimuth and elevation angles of the scanning beam's center axis are converted into beam pointing spatial coordinates in a three-dimensional spatial coordinate system.

[0034] Extract the target predicted spatial coordinates corresponding to the same time marker from the dynamic location prediction sequence;

[0035] Calculate the spatial straight-line distance between the beam pointing spatial coordinates and the target predicted spatial coordinates, and output the beam pointing position deviation.

[0036] Secondly, this application provides a real-time estimation system for adaptive signal-to-noise ratio of small target radar, comprising:

[0037] The scanning module is used to establish a motion characteristic model based on the continuous historical trajectory points of the maneuvering target during the interval between radar scans, and generate a dynamic position prediction sequence of the target in the next scan cycle.

[0038] The recording module is used to acquire radar scanning data packets at a fixed rotation speed through a mechanical rotating scanning device, and synchronously record the scanning beam pointing angle sequence and the echo signal strength sequence of the corresponding timestamp. The scanning beam pointing angle sequence includes the azimuth angle and elevation angle of the beam center axis.

[0039] The detection module is used to initiate multiple rounds of iterative detection within the spatial neighborhood of each target location point in the dynamic position prediction sequence. The first round of detection adopts an initial detection window that covers the maximum maneuver range of the target. In subsequent rounds, the detection window range is shrunk in both the range and azimuth directions according to the distribution characteristics of the echo signal intensity sequence within the previous detection window, until the detection window boundary is smaller than the half-power point width of the scanning beam.

[0040] The calculation module is used to extract the peak signal intensity and the corresponding background noise intensity in the final iteration round, and at the same time calculate the spatial matching degree between the scanning beam pointing angle sequence and the dynamic position prediction sequence, convert the spatial matching degree into the beam illumination efficiency factor, and combine the ratio of the peak signal intensity and the corresponding background noise intensity to calculate the dynamic correction amount of the signal-to-noise ratio.

[0041] The output module is used to perform weighted fusion based on the dynamic correction amount of the signal-to-noise ratio, the peak signal strength and the corresponding background noise intensity, and output the final signal-to-noise ratio value of the moving target in the current scanning cycle in real time.

[0042] Thirdly, this application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a real-time estimation method for adaptive signal-to-noise ratio of small target radar as described in the first aspect above.

[0043] Fourthly, this application provides a computer storage medium storing a computer program, which, when executed by a computer, implements a real-time estimation method for adaptive signal-to-noise ratio of small target radar as described in the first aspect.

[0044] In this embodiment, during the radar scanning interval, a motion characteristic model is established based on the continuous historical trajectory points of the maneuvering target, generating a dynamic position prediction sequence for the target in the next scanning cycle. A mechanical rotating scanning device acquires radar scanning data packets at a fixed rotation speed, simultaneously recording the scanning beam pointing angle sequence and the echo signal strength sequence corresponding to the timestamp. The scanning beam pointing angle sequence includes the azimuth and elevation angles of the beam center axis. For each target position point in the dynamic position prediction sequence, multiple rounds of iterative detection are initiated within its spatial neighborhood. The first round of detection uses an initial detection window covering the target's maximum maneuvering range. Subsequent rounds are based on the echo signal strength sequence from the previous round... The distribution characteristics within the detection window are analyzed, and the detection window range is narrowed bidirectionally along the range and azimuth directions until the detection window boundary is smaller than the half-power point width of the scanning beam. The peak signal intensity and corresponding background noise intensity in the final iteration are extracted, and the spatial matching degree between the scanning beam pointing angle sequence and the dynamic position prediction sequence is calculated. The spatial matching degree is converted into a beam illumination efficiency factor, and the dynamic signal-to-noise ratio correction is calculated in conjunction with the ratio of the peak signal intensity and the corresponding background noise intensity. Based on the dynamic signal-to-noise ratio correction, the peak signal intensity, and the corresponding background noise intensity, a weighted fusion is performed to output the final signal-to-noise ratio value of the maneuvering target in the current scanning cycle in real time.

[0045] The technical solution of this application has the following beneficial effects:

[0046] A motion model based on historical trajectories generates a dynamic position prediction sequence, resolving trajectory mismatch issues caused by sudden changes in target maneuvering and providing highly timely prior position information for subsequent scanning and focusing. A mechanical scanning device simultaneously acquires beam pointing angle and echo intensity sequences, ensuring spatiotemporal alignment between scanning parameters and signal data, providing a precise physical scanning foundation for iterative detection. A multi-round iterative detection window dynamic shrinkage mechanism adaptively adjusts the detection range based on echo intensity distribution, overcoming signal acquisition failure at beam edges for small targets and achieving precise locking of the core signal region of maneuvering targets. A beam illumination efficiency factor is generated through spatial matching degree, collaboratively correcting the basic signal-to-noise ratio (SNR) value and compensating for signal attenuation errors caused by target maneuvering deviating from the beam center. Weighted fusion of multi-source parameters outputs the final SNR value, suppressing random fluctuations in single scans and improving the estimation stability of high-speed maneuvering targets in complex interference environments.

[0047] Furthermore, an initial detection window covering the maximum maneuver range is formed by expanding the target's predicted location point as the center, and the echo intensity distribution is analyzed within the window. When a peak intensity point is detected in the window edge region, the detection window boundary is proportionally shrunk along the direction from the peak point to the center. This shrinking operation is repeated until the detection window is smaller than the beam's half-power point width and the peak point moves into the window's center region, at which point the iteration terminates. Through the trajectory prediction-guided iterative window dynamic shrinking mechanism, the spatial coupling relationship between the target maneuver trajectory and the scanning beam is adapted in real time, solving the signal intensity acquisition distortion problem caused by small targets deviating from the beam center during sudden turns, and significantly improving the robustness of signal-to-noise ratio estimation in the beam edge region.

[0048] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 A flowchart of a real-time estimation method for adaptive signal-to-noise ratio of small target radar provided in this application is shown;

[0051] Figure 2 A scene diagram is shown, illustrating a real-time estimation method for adaptive signal-to-noise ratio of small target radar provided in this application;

[0052] Figure 3 A schematic diagram of the structure of a real-time adaptive signal-to-noise ratio estimation system for small target radar provided in this application is shown.

[0053] Figure 4 A schematic diagram of the structure of a computing device provided in this application is shown. Detailed Implementation

[0054] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0055] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0056] Existing dynamic gate adjustment schemes suffer from three fundamental limitations when dealing with low-altitude, high-speed maneuvering targets: First, linear trajectory prediction models cannot adapt to sudden nonlinear maneuvers of the target, causing the detection gate to continuously deviate from its true position. Second, the fixed correction threshold is severely mismatched with the dynamic broadening characteristic of the mechanically scanned beam with elevation angle, resulting in complete failure of signal-to-noise ratio (SNR) correction in the beam edge region. Third, the SNR calculation process completely ignores the spatial coupling relationship between the beam center axis and the actual target position, causing systematic distortion of the output value during the target's maneuvering and turning phase. These defects collectively lead to uncontrollable jumps in the SNR estimate when small targets cross the beam critical region, severely restricting the tracking stability of the radar system.

[0057] To address the aforementioned shortcomings, this invention proposes a real-time signal-to-noise ratio (SNR) estimation method based on a three-element synergy of trajectory, scanning, and iteration. The core of this method lies in establishing a dynamic closed-loop coupling mechanism between trajectory prediction, mechanical scanning, and iterative detection. Specifically, a spatiotemporally synchronized position prediction sequence is generated through nonlinear motion feature modeling, driving multi-round iterative detection with bidirectional adaptive contraction of the range and azimuth window edges to lock onto the core region of the target signal in real time. Simultaneously, a spatial matching degree model between the beam pointing and the predicted position is constructed, generating a beam efficiency correction factor that dynamically evolves with position deviation. Finally, the optimized value is output by fusing signal strength, noise floor, and correction factor. This method significantly enhances the continuous tracking capability of small target radar in complex environments through the triple synergy of trajectory data-guided scanning focusing, iterative detection adapting to beam shape, and spatial matching degree compensating for positioning errors.

[0058] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0059] Figure 1A flowchart illustrating a real-time estimation method for adaptive signal-to-noise ratio of small target radar, as provided in this application embodiment, is shown below. Figure 1 As shown, the method includes:

[0060] 101. During the intervals between radar scans, establish a motion characteristic model based on the continuous historical trajectory points of the maneuvering target, and generate a dynamic position prediction sequence for the target in the next scan cycle;

[0061] In the above scheme, the continuous historical trajectory point of the maneuvering target refers to the target position record continuously detected by the radar system over multiple previous scanning cycles. Each record includes the time point of target detection, the straight-line distance from the radar, the azimuth angle, and the elevation angle. The motion feature model is a data relationship structure used to describe the target's motion patterns. This model establishes prediction rules for target position changes over time by analyzing the motion change patterns implicit in the historical trajectory points. The dynamic position prediction sequence is a set of future target positions calculated based on the motion feature model. Each position point includes the predicted time point, predicted distance, predicted azimuth angle, and predicted elevation angle, and these points are arranged in chronological order to form a sequence.

[0062] In this embodiment, firstly, all historical trajectory point data of the target within the most recent few radar scan cycles are collected. For example, the precise location and corresponding timestamp of the target when it was detected each time in the last 5 scans are extracted. Secondly, the motion change characteristics of the historical trajectory points are analyzed. Specifically, this involves calculating the displacement change, direction change, and time interval between adjacent trajectory points; identifying abrupt changes in displacement and direction; and statistically analyzing acceleration patterns.

[0063] Next, a nonlinear prediction model is constructed based on the identified motion characteristics: if the target motion exhibits regular speed changes, a polynomial fitting method is used to establish the relationship between position and time; if the target motion has random abrupt changes, a state transition probability model is used to describe the position jump rules.

[0064] Then, taking the current scan end time as the starting point, the model is input to deduce the time segment of the next scan cycle: the model calculation is performed for each time node, and the predicted distance value, predicted azimuth angle value, and predicted elevation angle value of the target at that time are output. The predicted positions of all time nodes are arranged in chronological order to form a dynamic position prediction sequence.

[0065] Finally, the completeness of the predicted sequence is verified: check whether the distance change between adjacent predicted points exceeds the maximum detection range of the radar; verify whether the azimuth / elevation angle changes conform to the mechanical scanning rotation speed constraints; eliminate abnormal predicted points that do not meet the physical laws, and generate the final usable dynamic position prediction sequence.

[0066] 102. A radar scanning data packet is acquired at a fixed rotation speed using a mechanical rotating scanning device, and the scanning beam pointing angle sequence and the echo signal strength sequence of the corresponding timestamp are recorded synchronously. The scanning beam pointing angle sequence includes the azimuth angle and elevation angle of the beam center axis.

[0067] Optionally, step 102 may specifically include the following steps:

[0068] 1021. During the rotation of the mechanical rotating scanning device at a constant angular velocity, whenever the azimuth angle of the scanning beam center axis reaches the preset angle sampling position, the radar transmitting unit is triggered to send a detection pulse signal and simultaneously receive the target reflection signal.

[0069] 1022. Divide the target reflection signal sampled at each azimuth angle into time dimensions, and combine the echo intensity values ​​of all range cells within the same time slice into two-dimensional distribution data of azimuth and range;

[0070] 1023. Extract the maximum signal intensity value and its corresponding distance unit coordinates within the coverage area of ​​the main lobe of the scanning beam from the two-dimensional distribution data of azimuth and distance. Generate an echo signal intensity sequence in chronological order of timestamps. At the same time, record the azimuth and elevation angle values ​​of the scanning beam center axis at each sampling point to form a scanning beam pointing angle sequence.

[0071] In the above scheme, the mechanical rotating scanning device refers to a radar antenna assembly driven by a motor to rotate at a constant speed. The scanning beam pointing angle sequence is a data sequence recording the direction of the beam centerline; each data point includes both horizontal and vertical angles. The echo signal intensity sequence is a set of target reflected signal intensity values ​​arranged in chronological order, with each intensity value corresponding to a specific range cell. The azimuth and range two-dimensional distribution data is tabular data arranged by horizontal angle and distance. The main lobe coverage area is the central radiation region where radar wave energy is most concentrated.

[0072] In this embodiment, firstly, in step 1021, as the radar antenna rotates at a constant speed, the angle encoder monitors the azimuth angle of the beam center axis in real time. When the azimuth angle reaches a preset sampling point, the trigger circuit synchronously starts the pulse transmission and reception modules. The transmission unit generates a high-frequency electromagnetic pulse, which is radiated into space through the antenna; the reception unit opens the signal acquisition window after pulse transmission, captures the electromagnetic wave energy reflected by the target, and converts it into an electrical signal. For example, the antenna rotates at 30 degrees per second, and sampling is preset to occur every 1 degree. When the azimuth angle reaches 45 degrees, a pulse is immediately transmitted and the reception channel is opened.

[0073] Next, in step 1022, the reflected signals acquired at each azimuth angle are segmented at fixed time intervals, determined by the radar pulse repetition period. Each signal segment corresponds to a specific range cell, where distance = speed of light × time interval / 2. The square of the signal amplitude within that range cell is extracted as the echo intensity. The intensity values ​​of all range cells are arranged in order of distance, forming a "distance and intensity" data column for the current azimuth angle. Data columns from multiple azimuth angles are combined into a two-dimensional data matrix. For example, the reflected signal is sliced ​​at 1-microsecond intervals, with each slice corresponding to a 150-meter range cell. At a 45-degree azimuth angle, the intensity values ​​of cells 0-150 meters are recorded in the first column, cells 150-300 meters are recorded in the second column, and so on, up to the maximum detection range.

[0074] Finally, step 1023 locates the main lobe region of the beam in the two-dimensional data matrix: along the azimuth direction, find regions where the intensity of three or more consecutive angle sampling points is significantly higher than the background value; along the range direction, find cell segments where the intensity continuously exceeds the threshold. Traverse all cells within this region, recording the maximum intensity and its corresponding range cell location. Simultaneously save the azimuth and elevation angles of the beam center axis at the current moment. Accumulate peak intensity, range coordinates, and pointing angle data in timestamp order to form three synchronized sequences. For example, in the 45-degree azimuth data sequence, it is found that the intensity in the 60-65 km range exceeds the threshold, with the highest intensity of 95 units at 62 km. Record the peak value of 95 and the range of 62 km, while simultaneously saving the azimuth angle of 45 degrees and the elevation angle of 3 degrees.

[0075] In practical applications, assuming a small UAV suddenly makes a 90-degree turn against a background of strong sea clutter, the radar needs to acquire a valid signal at the instant of the target's maneuver. Through step 1021, the mechanically scanned antenna rotates at a constant speed of 6 revolutions per second, setting an angle sampling point every 0.2 degrees. When the target suddenly turns, the antenna rotates to an azimuth angle of 122.3 degrees, triggering pulse transmission. The transmitting unit generates an X-band pulse with a pulse width of 0.5 microseconds, and the receiving unit simultaneously activates, capturing the target's reflected signal against the strong sea clutter background.

[0076] In step 1022, the reflected signal is sliced ​​at 0.1 microsecond intervals, with each slice corresponding to a distance resolution of 15 meters. The signal at a 122.3-degree azimuth angle is divided into 600 distance units, covering 0-9 kilometers, with each unit recording the square of the signal voltage amplitude. A two-dimensional data column corresponding to 122.3 degrees is generated: distance unit 1 [0-15 meters] intensity 32 units... distance unit 547 [8.2 kilometers] intensity 158 units... distance unit 600 (9 kilometers) intensity 28 units.

[0077] Step 1023, analysis of the two-dimensional data sequence revealed a sudden increase in the intensity of the cell cluster at 8.19-8.24 km, with a background average of 35 units and a target area intensity exceeding 150 units. The peak intensity of 158 units was located in range cell 547, corresponding to 8.205 km, confirming that this area is within the main lobe coverage. The current time was recorded as follows: peak intensity 158, range 8.205 km, azimuth 122.3 degrees, elevation 1.8 degrees. This data point was added to the sequence, and the timestamp was set to 123456.789 milliseconds of the radar system clock.

[0078] The overall scheme of step 102 above, through the coordinated use of three technologies—angle-triggered pulse synchronization mechanism, distance slice two-dimensional data reconstruction, and main lobe region peak positioning—accurately captures instantaneous signal characteristics and associates them with spatial pointing information in scenarios of sudden target maneuvering and strong background interference. This provides a spatiotemporally strictly aligned dataset of signal strength and beam pointing for subsequent processing, laying the data foundation for real-time signal-to-noise ratio estimation.

[0079] 103. For each target location point in the dynamic position prediction sequence, multiple rounds of iterative detection are initiated within its spatial neighborhood. The first round of detection adopts an initial detection window that covers the maximum maneuver range of the target. In subsequent rounds, based on the distribution characteristics of the echo signal intensity sequence within the previous round detection window, the detection window range is shrunk bidirectionally along the range and azimuth directions until the detection window boundary is smaller than the half-power point width of the scanning beam.

[0080] Optionally, step 103 may specifically include the following steps:

[0081] 1031. Taking each target location point in the dynamic position prediction sequence as the center point, a square initial detection window is formed by expanding along the distance and azimuth directions. The side length of the square initial detection window is set to cover the maximum maneuver offset range of the target.

[0082] 1032. Within the boundary range of the square initial detection window, analyze the intensity distribution state of the echo signal intensity sequence. When the intensity peak point of the intensity distribution state is detected to be located in the edge region of the square initial detection window, shrink the boundary range proportionally along the direction of the intensity peak point.

[0083] Specifically, step 1032 may include the following process: dividing the square initial detection window into an edge region, the edge region being defined as a narrow side region adjacent to the boundary of the detection window; locating the spatial position point corresponding to the maximum signal intensity within the current detection window range in the echo signal intensity sequence; if the spatial position point falls into the narrow side region, it is determined that the intensity peak point is in an edge position state; moving the detection window boundary towards the detection window center point by a preset shrinkage distance along the line connecting the intensity peak point in the edge position state and the detection window center point, forming a new shrinkage detection window boundary range.

[0084] 1033. Repeat the detection window shrinkage operation until the boundary range of the square initial detection window is smaller than the half-power point width of the scanning beam, and the intensity peak point is located in the central region of the square initial detection window, and the iteration termination condition is satisfied.

[0085] In the above scheme, the spatial neighborhood is the surrounding spatial region centered on the predicted target location. The initial detection window is a square area used for the first round of detection, with a side length covering the target's maximum possible displacement. Range and azimuth are defined as follows: range refers to the straight-line distance between the radar and the target, and azimuth refers to the horizontal angular direction. The intensity distribution is the numerical distribution of signal intensity at various locations within the detection window. The edge region is a ring-shaped strip near the boundary of the detection window. The half-power point width is the physical width when the radar beam energy drops to half of its peak value.

[0086] In this embodiment, step 1031 first extracts the spatial coordinates of the target prediction point from the dynamic position prediction sequence, including the straight-line distance to the radar and the horizontal azimuth angle. Using this coordinate point as the central reference, the coordinates are extended to both sides by the same distance, which is twice the product of the target's maximum velocity and the scan period; simultaneously, the coordinates are extended to both sides by the same angle, which is twice the product of the target's maximum angular velocity and the scan period. The larger of the distance extension and the angle extension is taken as the uniform side length of the initial square detection window, ensuring that the window can cover the target's maximum possible displacement range. For example, if the target's maximum flight speed is 400 m / s and the radar scan period is 0.1 seconds, then the distance extension is 80 meters; if the target's maximum turning angular velocity is 50 degrees / s, then the azimuth extension is 10 degrees. 10 degrees is selected as the initial window side length.

[0087] Next, in step 1032, within the spatial range covered by the initial detection window, all echo signal intensity data points are traversed, and the specific coordinates of the maximum signal intensity are determined by comparing the numerical values. The nearest distance from the peak point to the four boundaries of the detection window is calculated. If this distance is less than one-fifth of the current window side length, the peak point is determined to be located in the edge region. At this time, a spatial vector connecting the peak point position and the window center point is established. Along the direction pointed to by this vector, the four boundaries of the detection window are moved synchronously towards the center, with a moving distance of one-tenth of the current window side length, forming a new reduced detection window. For example, if the center of the initial window with a side length of 10 degrees is at an azimuth angle of 120 degrees, and the detected peak point is 1.8 degrees away from the boundary (<10 / 5 = 2 degrees), the window is shrunk along the direction from the center 120° to the peak point 121.8°, with each boundary moving inward by 10 degrees × 0.1 = 1 degree.

[0088] Finally, step 1033 repeats the peak detection and boundary shrinkage operations on the newly shrunk detection window. After each iteration, it checks whether the current window side length is less than the radar beam half-power point width and calculates whether the distance between the peak point and the window center is less than one-quarter of the window side length. The iteration terminates when both "window size is less than the physical beam width" and "peak point is located in the core area of ​​the window" are satisfied; otherwise, the boundary shrinkage continues at the same ratio. For example, the initial window is 10 degrees, and after three rounds of shrinkage, it is reduced to 7.3 degrees, which is still greater than the beam width of 2 degrees. The distance from the peak point to the center is 1.5 degrees > 7.3 / 4 = 1.825 degrees, so shrinkage continues; in the fifth round, the window is reduced to 1.8 degrees < 2 degrees, and the distance from the peak point to the center is 0.4 degrees < 1.8 / 4 = 0.45 degrees, so the iteration terminates.

[0089] In practical applications, a small UAV suddenly makes a 90-degree turn from 10 kilometers away against a background of dense buildings, with a radar scan cycle of 0.1 seconds and a beamwidth of 1.5 degrees. Step 1031 provides the target's position through a dynamic position prediction sequence: azimuth 85 degrees, range 10.2 kilometers. The target's maximum speed is 350 m / s, range spread = 350 × 0.1 × 2 = 70 meters; maximum angular velocity is 60 degrees / s, azimuth spread = 60 × 0.1 × 2 = 12 degrees. Choosing 12 degrees as the initial window side length, a square window is formed centered at (85°, 10.2 km), with a range of azimuth 79°-91° and range 9.9-10.5 km.

[0090] Step 1032 analyzes the signal strength data within the window. A peak intensity of 210 units is found at an azimuth of 88.7° and a distance of 10.38 km, with a building clutter average of 80 units. This point is 2.3 degrees from the nearest window boundary, between 88.7° and 91°. Since 2.3 degrees is less than 12 / 5 = 2.4 degrees, the actual calculation shows that 91° - 88.7° = 2.3 degrees < 2.4 degrees, thus classifying it as an edge region. Moving the boundary inward by 12 degrees × 0.1 = 1.2 degrees along the line connecting the center at 85° to the peak point at 88.7°, the new window range is reduced to an azimuth of 80.2°-89.8° and a distance of 10.1-10.45 km.

[0091] In step 1033, the second round detected the peak value shifted to 87.5° and 10.32km within the new window. The distance from the boundary (87.5° - 80.2° = 7.3°) is greater than the new window length (9.6 / 5 = 1.92 degrees). The window length is 89.8 - 80.2 = 9.6 degrees. The distance from 87.5° to the left boundary (80.2°) is 7.3 degrees, and the distance from the right boundary (89.8°) is 2.3 degrees, less than 1.92 degrees – still in the edge zone. The contraction continues along the direction from the center (85°) to the new peak value (87.5°) by 1.2 degrees, resulting in an actual contraction of 9.6 × 0.1 = 0.96 degrees. After four rounds of contraction, the window edge length is reduced to 1.6 degrees, greater than 1.5 degrees. The peak value is located at 86.1° and 10.25km, with an azimuth difference of 0.6 degrees from the window center (85.5°, 10.28km), greater than 1.6 / 4 = 0.4 degrees. The fifth round of contraction reduced the temperature to 1.44 degrees < 1.5 degrees, and the peak value was 0.35 degrees from the center < 1.44 / 4 = 0.36 degrees, thus satisfying the termination condition.

[0092] The overall scheme of step 103 above, through the dynamic adaptation of the initial window to the target's maneuverability, the intelligent boundary contraction strategy based on the peak position, and the iterative termination mechanism of dual physical constraints, enables the detection window to accurately converge to the target's real signal area under the environment of sudden target maneuvering and strong clutter interference, effectively overcoming the signal attenuation problem caused by beam edge effect, and providing a reliable guarantee for accurate signal-to-noise ratio estimation.

[0093] 104. Extract the peak signal intensity and corresponding background noise intensity in the final iteration round, and simultaneously calculate the spatial matching degree between the scanning beam pointing angle sequence and the dynamic position prediction sequence. Convert the spatial matching degree into a beam illumination efficiency factor, and combine the ratio of the peak signal intensity and the corresponding background noise intensity to collaboratively calculate the dynamic correction amount of the signal-to-noise ratio.

[0094] Optionally, step 104 may specifically include the following steps:

[0095] 1041. Calculate the difference between the coordinates of the center axis pointing point of the scanning beam and the predicted coordinates of the target's dynamic position under the same time mark to obtain the beam pointing position deviation.

[0096] Specifically, step 1041 may include the following processes: converting the azimuth and elevation angles of the scanning beam center axis into beam pointing spatial coordinates in a three-dimensional spatial coordinate system; extracting the target prediction spatial coordinates corresponding to the same time marker from the dynamic position prediction sequence; calculating the spatial straight-line distance between the beam pointing spatial coordinates and the target prediction spatial coordinates, and outputting it as the beam pointing position deviation.

[0097] 1042. A beam illumination efficiency factor is generated by mapping the magnitude of the beam pointing position deviation, wherein the mapping relationship satisfies that the beam illumination efficiency factor monotonically decreases as the beam pointing position deviation increases.

[0098] 1043. Divide the peak signal intensity in the final iteration by the corresponding background noise intensity to obtain the original signal-to-noise ratio value, and then multiply it by the beam illumination efficiency factor to generate the dynamic correction amount of the signal-to-noise ratio.

[0099] In the above scheme, spatial matching degree is a quantitative indicator of the degree of spatial overlap between the actual radar beam pointing direction and the predicted target position. Beam illumination efficiency factor is a system characterizing the effectiveness of beam energy coverage of the target. Beam pointing position deviation is the three-dimensional straight-line distance between the beam center point and the predicted target point. Background noise intensity is the average energy value of interference signals in the vicinity of the target signal.

[0100] In this embodiment, two sets of spatial coordinate data at the same time point are first obtained through step 1041: the first set is the actual pointing coordinates of the scanning beam's central axis, recorded by the azimuth and elevation angle values ​​through the antenna angle encoder, and converted into the spatial point position in a three-dimensional Cartesian coordinate system based on the radar station's geographical location; the second set is the target prediction coordinates at that moment in the dynamic position prediction sequence, including distance and azimuth angle values, which are also converted into a three-dimensional Cartesian coordinate system spatial point. The straight-line distance between the two spatial points is calculated, and this distance value is the beam pointing position deviation. For example, the beam pointing angle (120.5°, 2.1°) is converted into spatial coordinate point A (X=8120, Y=150, Z=300), and the target prediction point (121.3°, 8.2km) is converted into spatial coordinate point B (X=8145, Y=185, Z=305). The distance between points A and B is calculated [(8145-8120)]. 2 +(185-150) 2 +(305-300) 2 ]^(1 / 2)≈43.3 meters.

[0101] Next, a physical mapping model between the position deviation and the illumination efficiency factor is established through step 1042. This model follows the radar beam energy distribution law: when the deviation is zero, the efficiency factor is 1, and the beam center completely covers the target; when the deviation equals the beam half-power point radius, the efficiency factor drops to 0.5, and the target is at the beam energy half-decay point; when the deviation exceeds the beam width, the efficiency factor approaches 0. The measured deviation is converted into the efficiency factor using a preset discrete mapping table or continuous function relationship. For example, if the beam half-power point radius is 150 meters and the measured deviation is 43.3 meters, a linear mapping rule is adopted: Efficiency factor = 1 - Deviation / (2 × Half-power radius) = 1 - 43.3 / 300 ≈ 0.855.

[0102] Finally, in step 1043, two key parameters are extracted from the final iterative detection window: 1) the maximum value among all signal intensity values ​​within the window is taken as the peak signal intensity; 2) a background region is defined around the peak position, excluding neighboring cells, and the arithmetic mean of the signal intensity within this region is calculated as the background noise intensity. The peak value is divided by the noise intensity to obtain the original signal-to-noise ratio (SNR) value, which is then multiplied by the beam illumination efficiency factor output in step 1042 to generate the final dynamic SNR correction. For example, if the peak intensity within the detection window is 205 units, the background region distance is 8.12-8.18 km, excluding 8.15 km ± 50 meters, and the noise mean is 38 units, the original SNR = 205 / 38 ≈ 5.39. Multiplying by the efficiency factor 0.855, the corrected SNR ≈ 4.61.

[0103] In practical applications, high-speed UAVs perform 90-degree sharp turns in dense high-rise buildings, with a radar beam half-power radius of 120 meters and strong background interference from building reflections. Through step 1041, at the current timestamp of 123456.000 milliseconds, the scanning beam pointing angle is recorded as: azimuth 85.7°, elevation 3.2°. Through coordinate transformation: the spatial coordinates of the beam pointing point = (radar station coordinates X0 + 8150 × cos85.7° × cos3.2°, Y0 + 8150 × sin85.7° × cos3.2°, Z0 + 8150 × sin3.2°) are calculated to obtain (X = 8150.2, Y = 122.5, Z = 455.3); the target dynamic position prediction point: azimuth 86.3°, distance 8.15km, converted to (X = 8140.8, Y = 215.6, Z = 450.1); the deviation = [(8150.2 - 8140.8)]. 2 +(122.5-215.6) 2 +(455.3-450.1) 2 ]^(1 / 2)≈93.7 meters

[0104] Using the nonlinear mapping rule in step 1042: Deviation ≤ 60 meters: Efficiency factor = 1 - (deviation / 120)^2; 60 meters < deviation ≤ 240 meters: Efficiency factor = 0.75 - 0.25 × (deviation - 60) / 180; The current deviation is 93.7 meters > 60 meters, so we substitute it into the calculation: Efficiency factor = 0.75 - 0.25 × (93.7 - 60) / 180 = 0.75 - 0.25 × 0.1872 ≈ 0.703.

[0105] The final iteration window parameters in step 1043 are as follows: Peak position: azimuth 86.3°, distance 8.15km; Peak signal strength: 218 units, UAV reflected signal; Background area: distance 8.12-8.18km, excluding 8.145-8.155km, a total of 40 distance units; Background noise intensity calculation: Unit intensity value list: [35,42,38...33], including building reflection interference; Arithmetic mean = (35+42+...+33) / 40 = 39.6 units; Original signal-to-noise ratio = 218 / 39.6≈5.51; Dynamic correction = 5.51×0.703≈3.87.

[0106] The overall scheme in step 104 above effectively compensates for the beam illumination energy loss caused by target maneuvering by accurately quantifying the spatial deviation between the beam pointing and the target position, establishing an efficiency factor mapping model that conforms to physical laws, and collaboratively correcting the original signal-to-noise ratio value, thus outputting an optimized signal-to-noise ratio value that is closer to the real signal strength under strong background interference.

[0107] 105. Based on the dynamic correction amount of the signal-to-noise ratio, the peak signal strength and the corresponding background noise intensity, a weighted fusion is performed to output the final signal-to-noise ratio value of the moving target in the current scanning cycle in real time.

[0108] Optionally, step 105 may specifically include the following steps:

[0109] 1051. Assign a first fusion weight value to the peak signal strength, assign a second fusion weight value to the corresponding background noise strength, and assign a third fusion weight value to the dynamic correction amount of the signal-to-noise ratio. The sum of all fusion weight values ​​is a constant.

[0110] 1052. The peak signal strength is multiplied by the first fusion weight value to form the main component, the corresponding background noise intensity is multiplied by the second fusion weight value to form the noise component, and the dynamic correction amount of the signal-to-noise ratio is multiplied by the third fusion weight value to form the correction component.

[0111] 1053. Perform superposition calculation on the main component, the noise component and the correction component, and output the final signal-to-noise ratio value of the maneuvering target in the current scanning cycle.

[0112] In the above scheme, the first fusion weight value is the core importance coefficient assigned to the peak signal strength in the final calculation. The second fusion weight value is the interference suppression coefficient assigned to the background noise intensity in the final calculation. The third fusion weight value is the spatial compensation coefficient assigned to the dynamic signal-to-noise ratio correction in the final calculation. The principal component is the core signal characterization value after weight scaling of the peak signal strength. The noise component is the environmental interference characterization value after weight scaling of the background noise intensity. The correction component is the beam error compensation value after weight scaling of the dynamic signal-to-noise ratio correction.

[0113] In this embodiment, firstly, step 1051 presets a weight allocation rule based on the target type and environmental characteristics: the peak signal strength is assigned the highest weight as a direct observation value, the background noise intensity is assigned a suppressive weight as an interference term, and the dynamic signal-to-noise ratio correction is assigned a balanced weight as a spatial compensation term. The sum of the three weights is strictly maintained at 1. The specific weight values ​​are dynamically adjusted according to the radar operating mode: the correction weight is increased when the target is highly maneuverable, and the noise weight is decreased when the background interference is large. For example, for a high-speed target in a complex background, the peak weight is set to 0.6, the noise weight to 0.15, and the correction weight to 0.25.

[0114] Next, a weighted calculation is performed in step 1052: the peak signal strength value is multiplied by the first weight value to obtain the main component, which carries the core energy information of the target; the background noise intensity value is multiplied by the second weight value to obtain the noise component, which quantifies the intensity of environmental interference; and the signal-to-noise ratio dynamic correction value is multiplied by the third weight value to obtain the correction component, which reflects the beam space compensation value. The three components are calculated independently but maintain consistent dimensions. For example, the peak value is 205 units × 0.6 = 123, the noise value is 38 units × 0.15 = 5.7, and the correction value is 3.87 × 0.25 = 0.9675.

[0115] Finally, step 1053 performs algebraic fusion on the three components: the noise component is subtracted from the main component to suppress environmental interference, and the correction component is added to incorporate spatial compensation, ultimately outputting a signal-to-noise ratio value with complete physical meaning. This calculation is completed before the end of the scan cycle to ensure real-time performance. For example: 123 (main component) - 5.7 (noise component) + 0.9675 (correction component) = 118.2675 → output 118.3.

[0116] In practical applications, in the scenario of UAV ultra-low-altitude penetration, through step 1051, the high-speed UAV rapidly ascends against a background of strong sea clutter during the current scanning cycle. The peak signal strength of 218 units reflects the true target energy but exhibits fluctuations. The background noise intensity of 39.6 units indicates strong sea surface reflection interference, and the dynamic correction of the signal-to-noise ratio of 3.87 indicates a slight beam deviation. According to the maneuvering target weighting rule: the peak value is assigned a base confidence weight of 0.65, the noise is assigned a suppression weight of 0.15 due to strong interference, and the correction is assigned a compensation weight of 0.20 due to the target's ascent. The sum verifies that 0.65 + 0.15 + 0.20 = 1.00 meets the requirement of a constant constant.

[0117] In step 1052, the main component is calculated by multiplying the peak value of 218 units by the first weight of 0.65 to obtain 141.70 carrying target core energy; the noise component is calculated by multiplying the background noise of 39.6 units by the second weight of 0.15 to obtain 5.94 quantized sea clutter interference; the correction component is calculated by multiplying the dynamic correction amount of 3.87 by the third weight of 0.20 to obtain 0.774 reflecting the beam space compensation value. The dimensions of the three components are unified into dimensionless values.

[0118] In step 1053, the fusion calculation uses the formula "major component minus noise component plus correction component": 141.70 - 5.94 = 135.76 to eliminate environmental interference, 135.76 + 0.774 = 136.534 to incorporate spatial compensation. After rounding, the final signal-to-noise ratio (SNR) value of 136.5 is output. This value is output to the radar tracking system in real time at the end of the scan cycle. The original peak SNR of 218 / 39.6 ≈ 5.51 is overestimated because spatial attenuation is not considered. Even after dynamic correction of 3.87, it is still too low. The weighted fusion output of 136.5 is equivalent to an SNR of 13.65, which is closer to the actual reflection characteristics of the target and achieves balanced optimization under target maneuvering and strong interference.

[0119] The overall scheme in step 105 above, through a triple weight dynamic allocation mechanism, component independent weighted calculation and algebraic fusion optimization, effectively suppresses background interference while retaining the core signal characteristics of the target, accurately integrates beam space compensation, and outputs a final signal-to-noise ratio value with clear physical meaning and strong environmental adaptability, significantly improving the reliability of parameter estimation for high-speed maneuvering targets in complex scenarios.

[0120] The following is a complete embodiment for steps 101-105:

[0121] like Figure 2As shown, in the UAV canyon penetration scenario, step 101 is executed as follows: During the scanning interval, the radar system extracts 10 consecutive historical trajectory points of the target, with timestamp sequence T1-T10, corresponding to azimuth sequence [85.1°, 85.3°, 85.8°...86.7°], and distance sequence [8.15km, 8.17km, 8.21km...8.35km]. Analysis of the trajectory points reveals that within a 0.1-second time interval, the azimuth change increases from 0.2° to 0.9°, and the distance increment increases from 20 meters to 80 meters, indicating that the target is performing an acceleration and turning maneuver. A nonlinear motion model is established: azimuth change rate = 0.5 × t 2 +0.3×t, where t is the time increment, and the rate of change of distance = 4×t + 0.2×t 2 Starting from T10, the following three prediction points are extrapolated within 0.1 seconds of the next cycle: t = 0.033 seconds → azimuth 87.2°, distance 8.38km; t = 0.067 seconds → azimuth 87.9°, distance 8.42km; t = 0.1 seconds → azimuth 88.6°, distance 8.47km, generating a dynamic position prediction sequence.

[0122] Step 102 is executed as follows: The mechanical scanning antenna rotates at a constant speed of 6 revolutions per second, triggering pulse transmission every 0.1° azimuth angle. When the target turns to azimuth 87.5°, a Ku-band pulse with a pulse width of 0.5μs is transmitted. The receiver synchronously acquires the reflected signal and divides it into 800 range cells at 0.1μs intervals, corresponding to a distance resolution of 15 meters. A two-dimensional data table for azimuth 87.5° is generated: cell 1 (0-15m) intensity 32... cell 553 (8.295km) intensity 185... cell 800 (12km) intensity 28. The peak intensity of 185 units (i.e., 553 units) at azimuth 87.3°-87.7° and distance 8.25-8.35km in the main lobe region is extracted, and the current pointing angle of 87.5° / elevation of 2.1° is recorded and added to the sequence.

[0123] Step 103 is executed as follows: For the predicted point t = 0.067 seconds, azimuth 87.9° / distance 8.42km, an initial detection window is set, with a side length covering the maximum maneuver range: azimuth ±6° and distance ±0.6km. The first round of detection window range: azimuth 84.9°-90.9°, distance 7.82-9.02km. Analysis of the echo sequence within the window reveals a peak point at azimuth 88.7° / distance 8.38km, 1.2° from the boundary (<6 / 5 = 1.2°), thus determining the edge. The boundary is contracted by 10% along the direction from the center (87.9°) to the peak (88.7°), reducing the window length to 10.8°. After four rounds of iteration: In the third round, with a window length of 2.5°, the peak distance from the center was 0.6° > 2.5 / 4 = 0.625°, and the window continued to shrink; In the fourth round, with a window length of 1.9° < beam half-power width of 2.1° and a peak distance from the center of 0.4° < 1.9 / 4 = 0.475°, the final output window was terminated, with an azimuth of 87.5°-88.3° and a distance of 8.32-8.48km.

[0124] Step 104 is executed as follows: The final signal peak value extracted within the window is 205 units, with an azimuth of 88.1° and a range of 8.41km. Background noise is taken from adjacent cells, excluding peak points between 8.38-8.44km, with a mean of 42 units. Spatial matching degree is calculated: at the current moment, the beam pointing is 87.5° / 2.1° → spatial coordinates (8145, 135, 452), the predicted point is 87.9° / 8.42km → (8148, 218, 449), and the deviation is [(8145-8148)]. 2 +(135-218) 2 +(452-449) 2 ]^(1 / 2)≈83.1 meters. Mapped beam illumination efficiency factor: beam half-power radius 120 meters, deviation 83.1 meters → factor=0.75-0.25×(83.1-60) / 180≈0.72. Original signal-to-noise ratio=205 / 42≈4.88, dynamic correction amount=4.88×0.72≈3.51.

[0125] Weight allocation is performed in step 105: peak weight 0.65 / high confidence, noise weight 0.15 / strong suppression, correction weight 0.20 / compensation maneuver. Component calculations: principal component = 205 × 0.65 = 133.25; noise component = 42 × 0.15 = 6.3; correction component = 3.51 × 0.20 = 0.702. Fusion output: 133.25 - 6.3 + 0.702 = 127.652 → final signal-to-noise ratio 127.7.

[0126] This scheme significantly overcomes the signal-to-noise ratio estimation distortion problem in the beam edge region of traditional methods in scenarios with sudden target maneuvering and strong terrain clutter interference. It achieves stable tracking of high-speed small targets throughout the entire process and improves the target characteristic identification capability of radar system in complex electromagnetic environments. Through trajectory data-driven dynamic position prediction, precise spatiotemporal synchronization of mechanical scanning and signal acquisition, adaptive shrinking and locking of iterative detection window, physical correction of beam spatial matching degree, and multi-parameter weighted fusion optimization fifth-order collaborative mechanism, this scheme achieves stable tracking of high-speed small targets throughout the entire process.

[0127] Figure 3 This application provides a schematic diagram of the structure of a real-time adaptive signal-to-noise ratio estimation system for small target radar, as shown in the embodiments of this application. Figure 3 As shown, the system includes:

[0128] The scanning module 31 is used to establish a motion characteristic model based on the continuous historical trajectory points of the maneuvering target during the radar scanning interval, and generate a dynamic position prediction sequence of the target in the next scanning cycle.

[0129] Recording module 32 is used to acquire radar scanning data packets at a fixed rotation speed through a mechanical rotating scanning device, and synchronously record the scanning beam pointing angle sequence and the echo signal strength sequence of the corresponding timestamp. The scanning beam pointing angle sequence includes the azimuth angle and elevation angle of the beam center axis.

[0130] Detection module 33 is used to initiate multiple rounds of iterative detection within the spatial neighborhood of each target location point in the dynamic position prediction sequence. The first round of detection adopts an initial detection window that covers the maximum maneuver range of the target. In subsequent rounds, the detection window range is shrunk in both the range and azimuth directions according to the distribution characteristics of the echo signal intensity sequence within the previous detection window, until the detection window boundary is smaller than the half-power point width of the scanning beam.

[0131] The calculation module 34 is used to extract the peak signal intensity and the corresponding background noise intensity in the final iteration round, and at the same time calculate the spatial matching degree between the scanning beam pointing angle sequence and the dynamic position prediction sequence, convert the spatial matching degree into the beam illumination efficiency factor, and combine the ratio of the peak signal intensity and the corresponding background noise intensity to calculate the dynamic correction amount of the signal-to-noise ratio.

[0132] The output module 35 is used to perform weighted fusion based on the dynamic correction amount of the signal-to-noise ratio, the peak signal strength and the corresponding background noise intensity, and output the final signal-to-noise ratio value of the moving target in the current scanning cycle in real time.

[0133] Figure 3 The aforementioned real-time estimation system for the adaptive signal-to-noise ratio of small target radar can perform... Figure 1The implementation principle and technical effects of the real-time adaptive signal-to-noise ratio estimation method for small target radar described in the illustrated embodiment will not be repeated here. The specific operation methods of each module and unit in the real-time adaptive signal-to-noise ratio estimation system for small target radar in the above embodiments have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0134] In one possible design, Figure 3 The real-time signal-to-noise ratio estimation system for small target radar shown in the embodiment can be implemented as a computing device, such as... Figure 4 As shown, the computing device may include a storage component 41 and a processing component 42;

[0135] The storage component 41 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 42.

[0136] The processing component 42 is used for the above Figure 1 The embodiment describes a real-time adaptive signal-to-noise ratio estimation method for small target radar.

[0137] The processing component 42 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0138] Storage component 41 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0139] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0140] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0141] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0142] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0143] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown illustrates a real-time method for adaptive signal-to-noise ratio estimation in small target radar.

[0144] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0145] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0146] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0147] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A real-time adaptive signal-to-noise ratio estimation method for small target radar, characterized in that, include: During the intervals between radar scans, a motion characteristic model is established based on the continuous historical trajectory points of the maneuvering target, and a dynamic position prediction sequence for the target in the next scan cycle is generated. A mechanical rotating scanning device acquires radar scanning data packets at a fixed rotation speed, and synchronously records the scanning beam pointing angle sequence and the echo signal strength sequence with the corresponding timestamp. The scanning beam pointing angle sequence includes the azimuth angle and elevation angle of the beam center axis. For each target location point in the dynamic position prediction sequence, multiple rounds of iterative detection are initiated within its spatial neighborhood. The first round of detection uses an initial detection window that covers the target's maximum maneuver range. In subsequent rounds, based on the distribution characteristics of the echo signal intensity sequence within the previous round's detection window, the detection window range is shrunk bidirectionally along the range and azimuth directions until the detection window boundary is smaller than the half-power point width of the scanning beam. Extract the peak signal intensity and corresponding background noise intensity in the final iteration round, and simultaneously calculate the spatial matching degree between the scanning beam pointing angle sequence and the dynamic position prediction sequence. Convert the spatial matching degree into a beam illumination efficiency factor, and combine the ratio of the peak signal intensity and the corresponding background noise intensity to collaboratively calculate the dynamic correction amount of the signal-to-noise ratio. The final signal-to-noise ratio (SNR) value of the moving target within the current scanning cycle is output in real time by weighted fusion based on the dynamic correction amount of the SNR, the peak signal strength, and the corresponding background noise intensity.

2. The method according to claim 1, characterized in that, For each target location point in the dynamic position prediction sequence, multiple rounds of iterative detection are initiated within its spatial neighborhood. The first round of detection uses an initial detection window covering the target's maximum maneuver range. Subsequent rounds, based on the distribution characteristics of the echo signal intensity sequence within the previous round's detection window, shrink the detection window range bidirectionally along the range and azimuth directions until the detection window boundary is smaller than the half-power point width of the scanning beam, including: Using each target location point in the dynamic position prediction sequence as the center point, a square initial detection window is formed by expanding along the range and azimuth directions. The side length of the square initial detection window is set to cover the maximum maneuver offset range of the target. Within the boundary of the square initial detection window, the intensity distribution of the echo signal intensity sequence is analyzed. When the intensity peak point of the intensity distribution is detected to be located in the edge region of the square initial detection window, the boundary range is proportionally shrunk along the direction of the intensity peak point. Repeat the detection window shrinking operation until the boundary range of the initial square detection window is smaller than the half-power point width of the scanning beam, and the intensity peak point is located in the central region of the initial square detection window, at which point the iteration termination condition is satisfied.

3. The method according to claim 1, characterized in that, The process involves extracting the peak signal intensity and corresponding background noise intensity from the final iteration, simultaneously calculating the spatial matching degree between the scanning beam pointing angle sequence and the dynamic position prediction sequence, converting the spatial matching degree into a beam illumination efficiency factor, and collaboratively calculating the signal-to-noise ratio dynamic correction amount by combining the ratio of the peak signal intensity to the corresponding background noise intensity. This includes: The difference between the coordinates of the center axis pointing point of the scanning beam and the predicted coordinates of the target's dynamic position under the same time mark is calculated to obtain the beam pointing position deviation. A beam illumination efficiency factor is generated by mapping the magnitude of the beam pointing position deviation, wherein the mapping relationship satisfies that the beam illumination efficiency factor monotonically decreases as the beam pointing position deviation increases. The original signal-to-noise ratio (SNR) value is obtained by dividing the peak signal strength in the final iteration by the corresponding background noise intensity, and then multiplying it by the beam illumination efficiency factor to generate the dynamic SNR correction.

4. The method according to claim 1, characterized in that, The method involves acquiring radar scanning data packets at a fixed rotation speed using a mechanical rotating scanning device, and simultaneously recording the scanning beam pointing angle sequence and the echo signal strength sequence corresponding to the timestamp. The scanning beam pointing angle sequence includes the azimuth and elevation angles of the beam center axis, including: During the constant angular velocity rotation of the mechanical rotating scanning device, whenever the azimuth angle of the scanning beam center axis reaches the preset angle sampling position, the radar transmitting unit is triggered to send a detection pulse signal and simultaneously receive the target reflection signal. The target reflection signal sampled at each azimuth angle is divided into time dimensions, and the echo intensity values ​​of all range cells within the same time slice are combined into two-dimensional distribution data of azimuth and range; The maximum signal intensity and its corresponding distance unit coordinates within the coverage area of ​​the main lobe of the scanning beam are extracted from the two-dimensional distribution data of azimuth and distance. An echo signal intensity sequence is generated in chronological order according to the timestamps. At the same time, the azimuth and elevation angle values ​​of the scanning beam center axis at each sampling point are recorded to form a scanning beam pointing angle sequence.

5. The method according to claim 1, characterized in that, The step of weighted fusion based on the dynamic signal-to-noise ratio correction, the peak signal strength, and the corresponding background noise intensity, to output the final signal-to-noise ratio value of the moving target in the current scanning period in real time includes: A first fusion weight value is assigned to the peak signal strength, a second fusion weight value is assigned to the corresponding background noise strength, and a third fusion weight value is assigned to the dynamic correction amount of the signal-to-noise ratio. The sum of all fusion weight values ​​is a constant. The signal strength peak value is multiplied by the first fusion weight value to form the main component, the corresponding background noise intensity is multiplied by the second fusion weight value to form the noise component, and the signal-to-noise ratio dynamic correction amount is multiplied by the third fusion weight value to form the correction component; The main component, the noise component, and the correction component are superimposed to calculate the final signal-to-noise ratio of the maneuvering target in the current scanning cycle.

6. The method according to claim 2, characterized in that, Within the boundary of the initial square detection window, the intensity distribution of the echo signal intensity sequence is analyzed. When the intensity peak point of the intensity distribution is detected to be located in the edge region of the initial square detection window, the boundary range is proportionally reduced along the direction of the intensity peak point, including: An edge region is defined within the initial square detection window, and the edge region is defined as a narrow side region adjacent to the boundary of the detection window; Locate the spatial position point corresponding to the maximum signal intensity within the current detection window range in the echo signal intensity sequence. If the spatial position point falls into the narrow edge region, it is determined that the intensity peak point is in an edge position state. Along the line connecting the intensity peak point at the edge position and the center point of the detection window, the boundary of the detection window is moved towards the center point of the detection window by a preset shrinkage distance to form a new shrinkage detection window boundary range.

7. The method according to claim 3, characterized in that, The step of calculating the difference between the coordinates of the scanning beam center axis pointing point and the predicted coordinates of the target dynamic position under the same time mark to obtain the beam pointing position deviation includes: The azimuth and elevation angles of the scanning beam's center axis are converted into beam pointing spatial coordinates in a three-dimensional spatial coordinate system. Extract the target predicted spatial coordinates corresponding to the same time marker from the dynamic location prediction sequence; Calculate the spatial straight-line distance between the beam pointing spatial coordinates and the target predicted spatial coordinates, and output the beam pointing position deviation.

8. A real-time adaptive signal-to-noise ratio estimation system for small target radar, characterized in that, include: The scanning module is used to establish a motion characteristic model based on the continuous historical trajectory points of the maneuvering target during the interval between radar scans, and generate a dynamic position prediction sequence of the target in the next scan cycle. The recording module is used to acquire radar scanning data packets at a fixed rotation speed through a mechanical rotating scanning device, and synchronously record the scanning beam pointing angle sequence and the echo signal strength sequence of the corresponding timestamp. The scanning beam pointing angle sequence includes the azimuth angle and elevation angle of the beam center axis. The detection module is used to initiate multiple rounds of iterative detection within the spatial neighborhood of each target location point in the dynamic position prediction sequence. The first round of detection adopts an initial detection window that covers the maximum maneuver range of the target. In subsequent rounds, the detection window range is shrunk in both the range and azimuth directions according to the distribution characteristics of the echo signal intensity sequence within the previous detection window, until the detection window boundary is smaller than the half-power point width of the scanning beam. The calculation module is used to extract the peak signal intensity and the corresponding background noise intensity in the final iteration round, and at the same time calculate the spatial matching degree between the scanning beam pointing angle sequence and the dynamic position prediction sequence, convert the spatial matching degree into the beam illumination efficiency factor, and combine the ratio of the peak signal intensity and the corresponding background noise intensity to calculate the dynamic correction amount of the signal-to-noise ratio. The output module is used to perform weighted fusion based on the dynamic correction amount of the signal-to-noise ratio, the peak signal strength and the corresponding background noise intensity, and output the final signal-to-noise ratio value of the moving target in the current scanning cycle in real time.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a real-time estimation method for adaptive signal-to-noise ratio of small target radar as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The system contains a computer program that, when executed by a computer, implements a real-time estimation method for adaptive signal-to-noise ratio of small target radar as described in any one of claims 1 to 7.

Images