A full-dimension local sea clutter simulation method under high-speed motion platform

Abstract

The application provides a full-dimension local sea clutter simulation method under a high-speed motion platform, which comprises the following steps: generating a texture component of a sea area to be simulated, and calculating an instantaneous Doppler shift and an instantaneous Doppler bandwidth of the sea area to be simulated and a Doppler shift and broadening caused by the motion platform, and adding them to obtain a speckle component under the motion platform; calculating a sea spike parameter of the sea area to be simulated to generate a sea spike model; and using the texture component, the speckle component under the motion platform and the sea spike model to construct a full-dimension local sea clutter under the high-speed motion platform. The application simulates the influence of complex sea waves and clutter in the marine environment on the radar on the high-speed motion platform, can help to optimize the radar performance, evaluate the system stability, and find potential problems in practical application in advance, provides a more real test and evaluation environment under the condition of lacking a large amount of measured data, can approach the actual sea state, and realizes the full-dimension simulation of the sea clutter under the high-speed motion platform.

Description

Technical Field

[0001] This invention belongs to the field of radar signal processing technology, specifically relating to a full-dimensional local sea clutter simulation method for high-speed moving platforms. Background Technology

[0002] With the rapid development of the marine economy, the intelligence and automation of marine detection platforms are constantly improving. However, in complex sea conditions, the impact of sea clutter on radar system performance has become a major challenge in maritime target detection and identification. High-speed platforms, in particular, face even more severe interference problems when detecting maritime targets due to factors such as platform movement and sea surface fluctuations. Therefore, research on sea clutter simulation for high-speed platforms is of profound significance.

[0003] Current mainstream methods for modeling sea clutter treat it as a stationary, spherically invariant random process. However, with the improvement of marine radar resolution and the extension of observation time, the amplitude distribution characteristics of sea clutter have gradually shown obvious heavy tailing, and its correlation characteristics have also shown obvious non-stationarity. Among many sea clutter amplitude distribution models, the K-distribution, as a semi-physical and semi-statistical composite Gaussian model, can usually fit the amplitude distribution of sea clutter well. However, the K-distribution fails to take into account the spike effect caused by breaking waves on the sea surface, and therefore cannot effectively describe sea clutter with heavy tailing characteristics at high resolution. To solve this problem, researchers have proposed hybrid distribution models such as the KA distribution and the KK distribution to better model the spike components in sea clutter, but the mathematical expressions of these models are complex and difficult to apply directly in actual simulations. In recent years, some scholars have found that when the composite Gaussian model exhibits the characteristics of an inverse Gamma distribution or an inverse Gaussian distribution, its amplitude exhibits a Pareto distribution or a CGIG distribution, respectively. These distributions can better fit the heavy tailing phenomenon of high-resolution sea clutter. Moreover, the parameter estimation methods and detector designs for these models are relatively mature, and therefore they have been widely used in high-resolution sea clutter modeling.

[0004] On high-speed platforms, the high-speed movement of the platform exacerbates radar signal interference from factors such as frequency shift and phase change, thus affecting the accuracy of target detection. Existing technical solutions for simulating sea clutter on high-speed moving platforms mainly include the following methods: First, using deep neural networks (such as convolutional neural networks and recurrent neural networks) to model sea clutter. By training the model, the characteristics of sea clutter are automatically learned, enabling intelligent processing of radar echo signals. This approach requires a large amount of training data and computational resources, the model training process may be lengthy, and the model's interpretability is poor, potentially leading to uncertainties in practical applications. Second, sea clutter simulation based on statistical models, such as Rayleigh distribution and K-distribution, simulates the amplitude distribution and power spectrum characteristics of sea clutter. A corresponding model is established by analyzing the statistical characteristics of sea clutter. However, since statistical models may not accurately reflect the complexity of actual sea conditions, the simulation results may deviate from the real environment. Summary of the Invention

[0005] To address the aforementioned problems in existing technologies, this invention provides a method for simulating local sea clutter across all dimensions on a high-speed moving platform. The technical problem to be solved by this invention is achieved through the following technical solution:

[0006] A method for simulating local sea clutter in all dimensions under a high-speed moving platform includes:

[0007] S100, determine the sea area to be simulated, and generate the texture components of the sea area to be simulated;

[0008] S200 calculates the instantaneous Doppler shift and instantaneous Doppler bandwidth of the sea area to be simulated, and uses them as speckle components under a stationary platform, as well as to calculate the Doppler shift and broadening caused by a moving platform.

[0009] S300, the instantaneous Doppler shift and instantaneous Doppler bandwidth in the speckle component under the stationary platform are added to the corresponding Doppler shift and broadening caused by the moving platform to obtain the speckle component under the moving platform;

[0010] S400, calculate the sea peak parameters of the sea area to be simulated, and generate a sea peak model using the sea peak parameters;

[0011] S500 utilizes the texture components, speckle components under the motion platform, and sea spike model to construct the local sea clutter of the simulated sea area in all dimensions under the high-speed motion platform.

[0012] Beneficial effects:

[0013] This invention provides a method for simulating local sea clutter in all dimensions under a high-speed moving platform, comprising: determining the sea area to be simulated and generating texture components of the sea area to be simulated; calculating the instantaneous Doppler shift and instantaneous Doppler bandwidth of the sea area to be simulated, and using them as speckle components under a stationary platform, and calculating the Doppler shift and broadening caused by the moving platform; adding the instantaneous Doppler shift and instantaneous Doppler bandwidth of the speckle components under the stationary platform to the corresponding Doppler shift and broadening caused by the moving platform to obtain the speckle components under the moving platform; calculating the sea peak parameters of the sea area to be simulated, and generating a sea peak model using the sea peak parameters; and constructing a local sea clutter simulation of the sea area to be simulated in all dimensions under a high-speed moving platform using the texture components, the speckle components under the moving platform, and the sea peak model. This invention fully considers the clutter spectrum shift and broadening caused by platform motion, as well as the structural trend changes of clutter caused by platform motion. It comprehensively considers the influence of radar platform motion on large-scale dynamic sea surface generation, empirical formulas for sea surface scattering coefficients, sea clutter texture distribution, spatiotemporally varying sea clutter Doppler characteristics, and empirical statistical characteristics of sea peaks. It simulates the impact of complex ocean waves and clutter in the marine environment on radar on high-speed moving platforms (such as aircraft), which can help optimize radar performance, evaluate system stability, and identify potential problems in advance in practical applications. In the absence of a large amount of measured data, it provides a more realistic testing and evaluation environment that can approximate actual sea conditions, realizing full-dimensional simulation of sea clutter on high-speed moving platforms.

[0014] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of a full-dimensional local sea clutter simulation method for a high-speed moving platform provided by the present invention;

[0016] Figure 2 This is a schematic diagram of the motion platform observation provided by the present invention;

[0017] Figure 3 This is a schematic diagram of the orientation division provided by the present invention;

[0018] Figure 4 This is a schematic diagram of the process of the full-dimensional local sea clutter simulation method under the high-speed motion platform provided by the present invention;

[0019] Figure 5 This is a schematic diagram of the simulation results provided by the present invention with a distance resolution of 3m and a platform speed of 0m / s;

[0020] Figure 6 This is a schematic diagram of the simulation results provided by the present invention, with a distance resolution of 3m and a platform speed of 50m / s;

[0021] Figure 7 This is a schematic diagram of the simulation results provided by the present invention, with a distance resolution of 3m and a platform speed of 100m / s. Detailed Implementation

[0022] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0023] This invention proposes a method for modeling and simulating sea clutter on high-speed moving platforms, aiming to simulate the impact of complex ocean waves and clutter on radar on high-speed moving platforms (such as aircraft). This invention can help optimize radar performance, evaluate system stability, and identify potential problems in advance in practical applications. It provides a more realistic testing and evaluation environment, closely approximating actual sea conditions, in situations where a large amount of experimental data is lacking.

[0024] like Figure 1 As shown, this invention provides a full-dimensional local sea clutter simulation method for high-speed moving platforms, including:

[0025] S100, determine the sea area to be simulated, and generate the texture components of the sea area to be simulated;

[0026] In one specific embodiment of the present invention, S100 includes:

[0027] S110, determine the sea area to be simulated and divide it into multiple surface elements;

[0028] This step first determines the sea area to be simulated, and then uses surface analysis to segment the sea area to be simulated to obtain the sea surface at a predetermined sampling interval, wherein the sea surface is composed of multiple surface elements.

[0029] The simulated sea area is segmented using the trigonometric finite element method, commonly used in surface analysis. All triangles in the xy-plane are equilateral triangles with unit side lengths, and their vertices are defined using two-dimensional vectors {x...}. i , i = 1, 2, ..., I}. Related ocean parameters include wind speed, wind direction, and wind zone length. A time-varying sea surface with a sampling interval of Δt can be generated using the cosine superposition method or the Monte Carlo method, denoted as:

[0030] Φ TESS (m, x i ), m=1, 2,…,M, i=1, 2,…,I

[0031] The frozen sea surface at each moment is composed of triangles, and its projection onto the xy-plane is an equilateral triangle with a unit side length. Each surface element is represented by two coordinates on the xy-plane and its height coordinate on the frozen sea surface, i.e.

[0032] facetk ≡Triangle{(A k ,z1),(B k ,z2),(C k ,z3)}

[0033] S120, calculate the RCS of each face element;

[0034] This step includes S121, taking the rubbing angle of each surface element as input to the TCS model, calculating the normalized scattering coefficient through the TCS model, and calculating the RCS of each surface element using the normalized scattering coefficient and the surface area.

[0035] This step can use the TSC model to calculate the radar cross section (RCS) of each surface element.

[0036] The rubbing angle of each facet element can be calculated using the following formula:

[0037]

[0038] in, t represents the normal vector pointing upwards from the surface element. radar This represents the vector pointing from the center of the surface element to the radar, which can be calculated using the azimuth angle of the surface element relative to the radar and the radar altitude. Note that when the surface element is obstructed, its scattering angle is negative; in this case, the normalized scattering coefficient σ is directly applied. 0 Set to zero. Therefore, the RCS of each facet element can be calculated as follows:

[0039]

[0040] The second term of the product represents the area of ​​the surface element, which can be calculated using its projected area on the xy plane and its normal vector.

[0041] S121, the RCS of each spatial resolution unit is calculated using the RCS of each surface element, and they are combined into a time-varying RCS sequence; the time-varying RCS sequence reflects the spatiotemporal variation of the sea clutter texture trend.

[0042] On the radar range-azimuth plane grid, each spatial resolution cell consists of multiple surface elements, therefore the RCS of each spatial resolution cell can be calculated as follows:

[0043]

[0044] The calculated time-varying RCS sequence reflects the spatiotemporal variation of the sea clutter texture trend rather than the numerical values ​​of the texture components themselves.

[0045] S130 uses the RCS of each surface element and a random sequence following an inverse Gamma distribution to generate texture components for the simulated sea area.

[0046] This step includes S131, fusing the order of the time-varying RCS sequence with a random sequence following an inverse Gamma distribution to obtain the texture components of the spatial resolution unit; and S132, assembling the texture components of all spatial resolution units into the texture components of the sea area to be simulated.

[0047] In statistical analysis, the texture components of high-resolution sea clutter can be modeled as an inverse Gamma distribution. Taking an inverse Gamma distribution with scale parameter η and shape parameter λ as an example:

[0048]

[0049] In the formula, τ represents texture.

[0050] To ensure that the texture components of each resolution unit in the simulation, while exhibiting the aforementioned structural trends, also follow an inverse Gamma distribution, this method first generates an independent and identically distributed random sequence x(m), m = 1, 2, ..., M, following an inverse Gamma distribution. Subsequently, the time series RCS(m, n, l), m = 1, 2, ..., M, are fused with the random sequence x(m) following an inverse Gamma distribution to generate the texture sequence in the spatial unit (n, l). Specific details are as follows:

[0051] Arrange the random sequence x(m) in descending order to obtain a new sequence y(m):

[0052] y(1)≥y(2)≥…≥y(M-1)≥y(M).

[0053] It should be noted that the sorting process does not change the probability density function of x(m), therefore the two sequences are identically distributed.

[0054] Sort the sequence w(m) = RCS(m, n, l), where m = 1, 2, ..., M, in descending order, and record the order of each element in the sequence:

[0055] w(rank(1))≥w(rank(2))≥…≥w(rank(M-1))≥w(rank(M))

[0056] Where rank(k) represents the position of the k-th value in the original sequence, sorted from largest to smallest.

[0057] By adjusting the order of sequence y(m), the texture sequence on the spatial resolution unit (n, l) is obtained:

[0058] τ(rank(m),n,l)=y(m),m=1,2,…,M.

[0059] The structural trend changes in clutter caused by platform motion can be explained from a physical perspective as follows: as the radar platform moves, the radar illumination area changes, and the change in the radial distance between the swell and the radar platform is influenced by both swell motion and platform motion. Therefore, when simulating sea clutter from a moving platform, it is necessary to generate a sea surface larger than the radar illumination area, and the radar beam coverage area needs to continuously change with platform motion. Figure 2 As shown. Figure 2 The area enclosed by the solid black line represents the coverage area at a previous time, while the area connected by the dashed gray lines represents the coverage area of ​​the radar beam as it changes with the platform's movement. By dividing the sea surface and the illuminated area into facets and generating sequential textures for each moment, it is possible to simulate a texture sequence with structural trends under a moving platform.

[0060] S200 calculates the instantaneous Doppler shift and instantaneous Doppler bandwidth of the sea area to be simulated, and uses them as speckle components under a stationary platform, as well as to calculate the Doppler shift and broadening caused by a moving platform.

[0061] In one specific embodiment of the present invention, it includes:

[0062] S210 generates a speckle component sequence of the sea area to be simulated by calculating the instantaneous Doppler bandwidth of the speckle component.

[0063] Specifically, S210 includes:

[0064] S211, using the wind speed of the sea area to be simulated, calculate the average half-power Doppler bandwidth of the speckle component;

[0065] For X-band high-resolution radar, the Doppler power spectrum of sea clutter can be modeled as a single-peak bell-shaped function. A first-order autoregressive (AR) process is used for a zero-mean, unit-variance complex Gaussian sequence with zero Doppler offset and instantaneous Doppler bandwidth (IDBW) affected by local wind speed and texture.

[0066] Using empirical formulas, the Mean Half-Power Doppler Bandwidth (MDBW) can be calculated from the local wind speed U (ms). -1 The calculation yielded the following:

[0067] MDBW=0.48U / λ radar

[0068] S212, using the average half-power Doppler bandwidth, calculate the instantaneous Doppler bandwidth and time-varying coefficient of the speckle component;

[0069] As the texture components increase, the IDBW will decrease accordingly. Therefore, the IDBW is modeled as follows:

[0070]

[0071] The calculated time-varying coefficient ρ(m) is:

[0072]

[0073] S213, a low-pass Gaussian distributed random sequence is generated using a first-order AR process; the low-pass Gaussian distributed random sequence includes a hysteresis coefficient;

[0074] When simulating the speckle component, a low-pass Gaussian distributed random sequence is first generated using a first-order AR process, that is, a white complex Gaussian sequence w(m) with zero mean and unit variance is input into the AR model with a one-step hysteresis coefficient ρ:

[0075]

[0076] The output complex Gaussian sequence u0(m) also has zero mean and unit variance.

[0077] When the pulse repetition interval of the radar is Δt, the half-power Doppler bandwidth of the sequence u0(m) can be expressed as:

[0078]

[0079] S214, replace the hysteresis coefficient with the time-varying coefficient to obtain a low-pass Gaussian distributed random sequence whose instantaneous Doppler bandwidth converges to the theoretical value, and use it as the speckle component sequence.

[0080] To ensure that the output of the AR model has IDBW, the fixed coefficient ρ is replaced with the time-varying coefficient ρ(m). Since ρ(m) varies with the texture intensity and is relatively slow to change with the pulse repetition interval, that is, it remains basically unchanged within 0.1s, it can be guaranteed that the IDBW of the sequence u0(m) can converge to the theoretical value.

[0081] S220, calculate the instantaneous Doppler shift of the speckle component sequence.

[0082] This step includes S221, calculating the average Doppler shift of the speckle component sequence, wherein the average Doppler shift is expressed as a function of wind speed and wind direction angle; and S222, calculating the instantaneous Doppler shift using the average Doppler shift.

[0083] According to empirical formulas, the Mean Doppler Shift (MDOS) can be expressed as the wind speed U (ms). -1 and wind direction angle The function.

[0084]

[0085] Here, the superscripts "HH" and "VV" indicate the polarization mode of the radar transmitter and receiver. Furthermore, there is a negative correlation between the instantaneous Doppler shift (IDOS) of the speckle component and the relative texture intensity. The IDOS is modeled as follows:

[0086]

[0087] In the saturation regions with upper and lower bounds, a sigmoid-like nonlinear dependency is used, while a positive factor ε is used to control the degree of volatility. Finally, the complex Gaussian random sequence u(m) with instantaneous Doppler shift can be calculated as follows.

[0088] u(m)=u0(m)exp(2πjf d (m)Δt), m=1, 2,…,M.

[0089] S300, the instantaneous Doppler shift and instantaneous Doppler bandwidth in the speckle component under the stationary platform are added to the corresponding Doppler shift and broadening caused by the moving platform to obtain the speckle component under the moving platform;

[0090] This step adds the instantaneous Doppler shift in the speckle component under the stationary platform to the Doppler shift caused by the moving platform, and adds the bandwidth in the speckle component under the stationary platform to the broadening caused by the moving platform to obtain the speckle component under the moving platform.

[0091] The Doppler shift caused by platform motion is determined by the platform's velocity, beam azimuth and elevation angles, and radar parameters. The Doppler broadening phenomenon arises from the difference in Doppler shift from in-beam scatterers originating from the high-speed moving platform at different azimuth angles. Therefore, the Doppler shift and broadening caused by platform motion can be expressed as follows:

[0092]

[0093] Where f dr B is the Doppler shift caused by the platform motion. d For the Doppler widening caused by platform motion, v r For the platform's movement speed, θ is the elevation angle, θ is the azimuth angle, Δθ is the azimuth beamwidth, and λ is the azimuth beamwidth. radar This is the radar wavelength.

[0094] Adding the instantaneous Doppler offset of the speckle component to the Doppler offset caused by the moving platform yields the Doppler offset of the speckle component under the moving platform:

[0095]

[0096] The sea surface is divided into multiple azimuth-resolved units along the azimuth direction, such as Figure 3 As shown, each azimuth resolution cell on the same range ring corresponds to a different azimuth angle. Based on the platform motion and illumination geometry, the Doppler offset of the echo from each azimuth resolution cell and the Doppler offset and bandwidth of the sea clutter itself are calculated. The speckle component modulated by the platform motion is obtained on each azimuth resolution cell. Then, all azimuth resolution cells on the same range ring are superimposed to obtain the speckle component containing the Doppler offset and bandwidth caused by the platform motion.

[0097] S400, calculate the sea spike parameters of the sea area to be simulated, and generate a sea spike model using the sea spike parameters; wherein, the sea spike parameters include: the probability of n sea spikes appearing in each resolution unit within each pulse repetition interval, the duration of the sea spike, the radial length of the sea spike, the coverage probability of the sea spike, the envelope of the sea spike, the radial velocity, and the location of the sea spike occurrence.

[0098] Sea spikes occur sporadically within resolvable cells, and this phenomenon can be described using a Poisson process. Assuming the average number of sea spikes per unit area and per unit time is λ0, the probability of n sea spikes occurring in each resolvable cell within each pulse repetition interval can be expressed as:

[0099]

[0100] Where S represents the area of ​​the radar spatial resolution cell. For X-band radar, the average probability of sea spike occurrence λ0 is generally less than 10. -5 Furthermore, λ0 increases with increasing sea state and angle of attack. λ0 reaches its maximum value when the radar observes against the wind and its minimum value when observing with the wind. When the area of ​​the spatial resolution cell is 500m²... 2 At this time, the probability of a sea spike occurring within 1 second for each resolution cell is less than 0.005. For high-resolution radar, since sea spikes are caused by near-wave breaking, it is difficult for two or more sea spikes to occur simultaneously within one pulse for each spatial resolution cell. Therefore, we assume that each resolution cell has at most one sea spike within one pulse. When simulating sea spikes, a random integer is generated at each pulse-range grid point (m, n) according to the above formula. If the integer is 0, no sea spike occurs at that grid point; otherwise, a sea spike appears in that resolution cell from the beginning of that pulse.

[0101] After determining the location of the sea crests in the space-time plane, each sea crest can be described by parameters such as its duration, distance spread, slow-time envelope, peak-to-clutter ratio (PCR), and radial velocity. The distribution patterns of some of these parameters can be summarized from measured data. The duration T of the sea crest... spike The peaks are mainly concentrated within 1 to 4 seconds, and the longer the duration, the lower the probability of the peaks occurring. The duration of the peaks can be modeled as having a mean of [value missing].

[0102]

[0103] When radar range resolution reaches meter-level or sub-meter-level, sea spikes often occupy multiple adjacent range cells. The number of range cells occupied by a sea spike depends on the radial length of the corresponding near-break wave. Observational data shows that the radial length of sea spikes is concentrated around a certain fixed value; the probability of excessively long or short sea spikes decreases significantly. Therefore, an Nth-order spline distribution is used to model the radial length of sea spikes. An Nth-order spline distribution can be viewed as the convolution of N uniform probability density functions, or as the distribution of the sum of N independent and identically distributed uniformly distributed random variables, i.e.:

[0104]

[0105] in, This represents the mean radial length of the sea crest, in meters. Because the sea crest expands along the distance and pulse, the probability of the sea crest covering the simulated 3D data is:

[0106]

[0107] Where S / Δr represents the aspect ratio of the radar spatial resolution cell. The coverage probability of the sea crest depends on the unit occurrence probability λ0 and the average duration of the sea crest. average length And the aspect ratio S / Δr of the spatial resolution unit.

[0108] Besides duration and radial length, the radial envelope, temporal envelope, PCR, and radial velocity of the sea spike also require corresponding statistical models. Considering that the sea spike is generated by the mirror structure on the near-breaking wave, we can assume that the sea spike envelope is a tent function along the distance and an isosceles trapezoidal function along the pulse. We assume the support interval of this two-dimensional function is [0, 1]. 2, this function is a tent function on the interval [0, 1] in the instantaneous time slice of a certain pulse, and an isosceles trapezoid function on the interval [0, 1] in the spatial slice of a certain range bin

[0109] Ψ(t,r)=max{0,min{4min(t,1-t),1}}×max{0,1-|2r-1|}.

[0110] It is assumed that the amplitude A of the sea spike follows the interval [a,b](E{τ}) 1 / 2 uniform distribution, where 1<a<b, and a and b need to be selected according to sea state experience during simulation. The speed of breaking waves is closely related to the speed of swells, and the speed of ocean waves increases with the increase of wavelength. It can be found from measured data that the speed of sea spikes is approximately 0.3 times the speed of the dominant wave, and the distribution is relatively concentrated. Therefore, the radial velocity of the sea spike is modeled as:

[0111] υ spike ~Uniform([0.2ν d , 0.4ν d )cosφ,

[0112] where φ represents the wave direction angle relative to the radar line of sight, v d represents the dominant wave phase velocity of the wave field. The dominant wave velocity can be calculated using the time-varying sea surface parameters during the simulation process.

[0113] According to the established simulation model, the mathematical expression of each sea spike with respect to the occurrence position (m0, n0), duration T spike , radial length L, amplitude A and radial velocity v spike , that is, the sea spike model is as follows:

[0114]

[0115] wherein, represents a random initial phase on the interval [-π, π). The first term in the above formula represents the envelope of the sea spike on the two-dimensional range-time plane, and the second term represents the phase sequence, wherein it is assumed that all range bins occupied by a single sea spike share the same phase function. The support domain of the sea spike in the formula is a parallelogram, that is:

[0116]

[0117] S500, constructing full-dimensional local sea clutter of the sea area to be simulated under a high-speed moving platform by using the texture component, the speckle component under the moving platform and the sea spike model.

[0118] When simulating sea clutter from a moving platform, it is necessary to comprehensively consider the large-scale dynamic sea surface generation, empirical formulas for sea surface scattering coefficients, sea clutter texture distribution, spatiotemporally varying Doppler characteristics of sea clutter, empirical statistical characteristics of sea spikes, and the impact of platform motion on these characteristics. The simulation results should be in the following format:

[0119]

[0120] This mainly includes continuous texture components, speckle components, and discrete sea spike components. Here, Δt represents the pulse repetition interval of the radar slow time t, Δr represents the sampling interval of the radial distance r, and Δθ represents the sampling interval of the azimuth angle θ. Each term in the summation corresponds to a random occurrence occurring at the l-th... k Wave position N k A continuous distance unit and lasting M k A series of continuous pulses of sea spikes.

[0121] This invention does not require high levels of measured data and computational resources. Based on a combination of physical formulas and statistical assumptions, the simulation results more closely resemble actual conditions, providing a realistic clutter model environment and supporting the design and performance evaluation of radar systems. This invention presents a simulation model of sea surface echoes with meter-level or sub-meter resolution by adding structural trends and sea spike components to a composite Gaussian model with an inverse Gamma distribution. This is of great significance for the upgrading and modification of moving platform radar systems and the performance testing of detection and tracking algorithms.

[0122] refer to Figure 4 , Figure 4 The simulation flow of the full-dimensional sea clutter simulation method under the high-speed motion platform of the present invention is shown below. To prove the effectiveness of the method, the present invention follows... Figure 4 The simulation process analyzes the simulation results at a set resolutions, and the analysis mainly includes texture trends and the rationality of the Doppler spectrum. Figure 5-7 Power plots and average Doppler spectra of simulation data under several different motion speeds are presented. In the three sets of simulation data with a resolution of 3 m, a clear texture trend is observed, and this trend shows a significant shift caused by platform motion as the platform speed increases. When the platform is stationary, the Doppler bandwidth of the clutter can be calculated to be approximately 70 Hz using the formula in step 2. When the platform motion speed is resolved to 50 m / s and 100 m / s, the Doppler bandwidths can be calculated to be 38 Hz and 76 Hz respectively using the formula for the moving platform in step 2, which are basically consistent with the simulation results.

[0123] It is worth noting that the terms "first" and "second" in this invention are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0124] Although this application has been described herein in conjunction with various embodiments, those skilled in the art will understand and implement other variations of the disclosed embodiments by reviewing the accompanying drawings, the disclosure, and the appended claims in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality.

[0125] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A method for full-dimensional local sea clutter simulation under high-speed motion platform, characterized in that, include: S100, determine the sea area to be simulated, and generate the texture components of the sea area to be simulated; S200 calculates the instantaneous Doppler shift and instantaneous Doppler bandwidth of the sea area to be simulated, and uses them as speckle components under a stationary platform, as well as to calculate the Doppler shift and broadening caused by a moving platform. S300, the instantaneous Doppler shift and instantaneous Doppler bandwidth in the speckle component under the stationary platform are added to the corresponding Doppler shift and broadening caused by the moving platform to obtain the speckle component under the moving platform; S400, calculate the sea peak parameters of the sea area to be simulated, and generate a sea peak model using the sea peak parameters; S500 utilizes the texture component, speckle component under the motion platform, and sea spike model to construct the local sea clutter of the simulated sea area under the high-speed motion platform in all dimensions. S100 includes: S110, determine the sea area to be simulated and divide it into multiple surface elements; S120, calculate the RCS of each face element; S130, using the RCS of each surface element and a random sequence following an inverse Gamma distribution, generates the texture components of the sea area to be simulated; S110 includes: The sea area to be simulated is determined, and the sea area to be simulated is segmented using the surface analysis method to obtain the sea surface at a predetermined sampling interval, wherein the sea surface is composed of multiple surface elements; S120 includes: S121, the rubbing angle of each surface element is used as the input of the TCS model, and the normalized scattering coefficient is calculated through the TCS model. The RCS of each surface element is calculated using the normalized scattering coefficient and the surface area. S121, the RCS of each spatial resolution unit is calculated using the RCS of each surface element, and they are combined into a time-varying RCS sequence; the time-varying RCS sequence reflects the spatiotemporal variation of the sea clutter texture trend; S130 includes: S131, the order of the time-varying RCS sequence is fused with a random sequence following an inverse Gamma distribution to obtain the texture component of the spatial resolution unit; S132, assemble the texture components of all spatial resolution units into the texture components of the sea area to be simulated; S200 includes: S210 generates a speckle component sequence of the sea area to be simulated by calculating the instantaneous Doppler bandwidth of the speckle component. S220, calculate the instantaneous Doppler shift of the speckle component sequence; S210 includes: S211, using the wind speed of the sea area to be simulated, calculate the average half-power Doppler bandwidth of the speckle component; S212, using the average half-power Doppler bandwidth, calculate the instantaneous Doppler bandwidth and time-varying coefficient of the speckle component; S213, a low-pass Gaussian distributed random sequence is generated using a first-order AR process; the low-pass Gaussian distributed random sequence includes a hysteresis coefficient; S214, replace the hysteresis coefficient with the time-varying coefficient to obtain a low-pass Gaussian distributed random sequence whose instantaneous Doppler bandwidth converges to the theoretical value, and use it as the speckle component sequence.

2. The method of claim 1, wherein, S220 includes: S221, Calculate the average Doppler shift of the speckle component sequence, wherein the average Doppler shift is expressed as a function of wind speed and wind direction angle; S222, calculate the instantaneous Doppler shift using the average Doppler shift.

3. The method of claim 1, wherein, The S300 includes: The instantaneous Doppler shift in the speckle component under the stationary platform is added to the Doppler shift caused by the moving platform, and the bandwidth in the speckle component under the stationary platform is added to the broadening caused by the moving platform to obtain the speckle component under the moving platform.

4. The method of claim 1, wherein, The sea peak parameters include: Each resolution cell occurs within each pulse repetition interval n The probability of a sea spike, the duration of a sea spike, the radial length of a sea spike, the coverage probability of a sea spike, the envelope of a sea spike, radial velocity and the location of occurrence of a sea spike.

Images