Full-dimensional local sea clutter simulation method under high-speed motion platform

Abstract

The invention provides a full-dimensional local sea clutter simulation method under a high-speed motion platform. The full-dimensional local sea clutter simulation method comprises the following steps: generating a texture component of a sea area to be simulated, and calculating instantaneous Doppler shift and instantaneous Doppler bandwidth of the sea area to be simulated and Doppler shift and broadening caused by the motion platform; correspondingly adding the speckle components to obtain speckle components under the motion platform; calculating sea peak parameters of the sea area to be simulated so as to generate a sea peak model; and constructing a full-dimensional local sea clutter under the high-speed motion platform by using the texture component, the speckle component under the motion platform and a sea peak model. The method simulates the influence of complex sea waves and clutters in the marine environment on the radar on the high-speed motion platform, can help to optimize the radar performance and evaluate the system stability, discovers potential problems in advance in practical application, provides a more real test and evaluation environment under the condition of lacking a large amount of actual measurement data, can approach the actual sea condition, and has a good application prospect. And the sea clutter full-dimension simulation of the high-speed motion platform is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of radar signal processing, and particularly relates to a full-dimensional local sea clutter simulation method under a high-speed moving platform. Background Art

[0002] With the rapid development of the marine economy, the intelligence and automation levels of marine detection platforms have been continuously improved. However, under complex sea conditions, the impact of sea clutter on the performance of radar systems has become a major challenge in marine target detection and recognition. Especially for high-speed platforms, due to factors such as platform movement and sea surface fluctuations, their radar systems face more severe interference problems when detecting marine targets. Therefore, it is of great significance to study the sea clutter simulation of high-speed platforms.

[0003] The current mainstream sea clutter modeling method regards it as a stationary spherically invariant random process. However, with the improvement of the resolution of marine radars and the extension of the observation time, the amplitude distribution characteristics of sea clutter gradually show obvious heavy tailing phenomena, and its correlation characteristics also show obvious non-stationarity. Among many sea clutter amplitude distribution models, the K-distribution, as a semi-physical and semi-statistical compound Gaussian model, can usually better fit the amplitude distribution of sea clutter. However, the K-distribution fails to consider the spike effect caused by breaking waves on the sea surface, so it cannot effectively describe the sea clutter with heavy tailing characteristics at high resolutions. 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 directly apply in actual simulations. In recent years, some scholars have found that when the compound Gaussian model exhibits the characteristics of the inverse Gamma distribution or the inverse Gaussian distribution, its amplitude shows the Pareto distribution or the CGIG distribution respectively, and these distributions can better fit the heavy tailing phenomena of high-resolution sea clutter. Moreover, the parameter estimation methods and detector designs of these models have been relatively mature, so they have been widely used in high-resolution sea clutter modeling.

[0004] On a high-speed platform, due to the high-speed movement of the platform, factors such as the frequency shift and phase change of radar signals will exacerbate the interference of sea clutter, thereby affecting the accuracy of target detection. For the simulation of sea clutter on a high-speed moving platform, the existing technical solutions mainly include the following methods: First, use deep neural networks (such as convolutional neural networks, recurrent neural networks, etc.) to model sea clutter. By training the model, automatically learn the characteristics of sea clutter and achieve intelligent processing of radar echo signals. This solution requires a large amount of training data and computing resources, the model training process may be long, and the interpretability of the model is poor, which may lead to uncertainties in practical applications. Second, the simulation of sea clutter based on statistical models, using statistical models such as Rayleigh distribution, K distribution, etc., to simulate the amplitude distribution and power spectrum characteristics of sea clutter. By analyzing the statistical characteristics of sea clutter, establish the corresponding model. Since the statistical model may not accurately reflect the complexity of the actual sea conditions, there is a deviation between the simulation results and the real environment. Summary of the Invention

[0005] To solve the above problems existing in the prior art, the present invention provides a full-dimensional local sea clutter simulation method under a high-speed moving platform. The technical problems to be solved by the present invention are realized through the following technical solutions:

[0006] A full-dimensional local sea clutter simulation method under a high-speed moving platform includes:

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

[0008] S200, calculate the instantaneous Doppler shift and instantaneous Doppler bandwidth of the sea area to be simulated, and use them as the speckle component under a stationary platform and calculate the Doppler shift and broadening caused by the moving platform;

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

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

[0011] S500, use the texture component, the speckle component under the moving platform and the sea spike model to construct the full-dimensional local sea clutter of the sea area to be simulated under the high-speed moving platform.

[0012] Beneficial Effects:

[0013] The present invention provides a full - dimensional local sea clutter simulation method under a high - speed moving platform, which includes: determining the sea area to be simulated and generating the texture component 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 the speckle component under a stationary platform, as well as calculating the Doppler shift and broadening caused by the moving platform; adding the instantaneous Doppler shift and instantaneous Doppler bandwidth in the speckle component under the stationary platform and the Doppler shift and broadening caused by the moving platform correspondingly to obtain the speckle component under the moving platform; calculating the sea spike parameters of the sea area to be simulated and generating a sea spike model using the sea spike parameters; constructing the full - dimensional local sea clutter of the sea area to be simulated under the high - speed moving platform by using the texture component, the speckle component under the moving platform, and the sea spike model. The present invention fully considers problems such as the clutter spectrum shift and spectrum broadening caused by platform movement, and the change of clutter structural trend caused by platform movement. The method comprehensively considers the influence of radar platform movement on large - scale dynamic sea surface generation, the empirical formula of sea surface scattering coefficient, sea clutter texture distribution, spatio - temporal variable sea clutter Doppler characteristics, and the empirical statistical characteristics of sea spikes, simulates the influence of complex sea waves and clutter in the ocean environment on the radar on a high - speed moving platform (such as an aircraft, etc.), can help optimize radar performance, evaluate system stability, and discover potential problems in advance in practical applications. In the case of lacking a large amount of measured data, it provides a more realistic test and evaluation environment, can approach the actual sea conditions, and realizes the full - dimensional simulation of sea clutter under a high - speed moving platform.

[0014] The following will further elaborate on the present invention in detail with reference to the drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

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

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

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

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

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

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

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

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

[0023] The present invention proposes a method for modeling and simulating sea clutter under a high-speed moving platform, aiming to simulate the influence of complex sea waves and clutter in the ocean environment on a radar on a high-speed moving platform (such as an aircraft, etc.). The invention can help optimize radar performance, evaluate system stability, and discover potential problems in advance in practical applications. In the case of a lack of a large amount of measured data, it provides a more realistic test and evaluation environment and can approximate the actual sea conditions.

[0024] As Figure 1 shown, the present invention provides a full-dimensional local sea clutter simulation method under a high-speed moving platform, including:

[0025] S100, determining the sea area to be simulated and generating the texture component of the sea area to be simulated;

[0026] In a specific implementation manner of the present invention, S100 includes:

[0027] S110, determining the sea area to be simulated and dividing it into multiple facets;

[0028] In this step, the sea area to be simulated is first determined, and the sea area to be simulated is divided using the surface analysis method to obtain a sea surface with a predetermined sampling interval, where the sea surface is composed of multiple facets.

[0029] The sea area to be simulated is divided using the commonly used triangular finite element method 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}. The relevant ocean parameters include: wind speed, wind direction, and fetch length. The time-varying sea surface with a sampling interval of Δt can be generated according to 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 on the xy plane is an equilateral triangle with unit side lengths. Each facet is represented by two coordinates in the xy plane and the height coordinate on the frozen sea surface, that is

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

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

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

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

[0036] The grazing angle of each surface element can be calculated according to the following formula:

[0037]

[0038] where, represents the normal vector pointing above the surface element, and t radar represents the vector pointing from the center of the surface element to the radar, which can be calculated by using the azimuth angle of the surface element relative to the radar and the radar height. It should be noted that when the surface element is blocked, its grazing angle is negative, and at this time, the normalized scattering coefficient σ 0 is directly set to zero. Therefore, the RCS of each surface element can be calculated as follows:

[0039]

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

[0041] S121, calculate the RCS of each spatial resolution unit by using the RCS of each surface element, and form a time-varying RCS sequence; the time-varying RCS sequence reflects the spatio-temporal variation of the sea clutter texture trend.

[0042] On the grid of the radar range-azimuth plane, each spatial resolution unit is composed of multiple surface elements. Therefore, the RCS of each spatial resolution unit can be calculated as follows:

[0043]

[0044] The calculated time-varying RCS sequence reflects the spatio-temporal variation of the sea clutter texture trend rather than the numerical value of the texture component itself.

[0045] S130, generate the texture component of the sea area to be simulated by using the RCS of each surface element and a random sequence obeying the inverse Gamma distribution.

[0046] This step includes S131, where the order of the time-varying RCS sequence is fused with a random sequence obeying the inverse Gamma distribution to obtain the texture component of the spatial resolution unit; S132, where the texture components of all spatial resolution units are combined to form the texture component of the sea area to be simulated.

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

[0048]

[0049] In the formula, τ represents the texture.

[0050] In order to ensure that the texture component of each resolution unit in the simulation follows the inverse Gamma distribution while having the structural trend as described above. In this method, an independent and identically distributed random sequence x(m), m = 1, 2,..., M that obeys the inverse Gamma distribution is first generated. Subsequently, the order of the time series RCS(m, n, l), m = 1, 2,..., M is fused with the random sequence x(m) that obeys the inverse Gamma distribution to generate the texture sequence in the spatial unit (n, l). The specific details are as follows:

[0051] The random sequence x(m) is sorted 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), so the two sequences are identically distributed.

[0054] The sequence w(m) = RCS(m, n, l), m = 1, 2,..., M is sorted in descending order, and the order of each element in the sequence is recorded:

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

[0056] Among them, rank(k) represents the position of the value ranked k from largest to smallest in the original sequence.

[0057] The order of the sequence y(m) is adjusted to obtain the texture sequence on the spatial resolution unit (n, l):

[0058] τ(rank(m), n, l) = y(m), m = 1, 2,..., M.

[0059] Regarding the structural trend change of clutter caused by platform movement, it can be physically explained as follows: As the radar platform moves, the area irradiated by the radar changes, and the change in the radial distance between the swell and the radar platform is affected by both the movement of the swell and the movement of the platform. Therefore, when simulating sea clutter of a moving platform, it is necessary to generate a sea surface larger than the radar irradiation area, and the coverage area of the radar beam changes continuously with the movement of the platform. As Figure 2 shown. Figure 2 In the figure, the area enclosed by the black solid line represents the coverage range at a previous time, and the area formed by connecting the gray dotted lines represents the coverage range of the radar beam changing with the movement of the platform. For the sea surface and the irradiation area at each moment, by performing surface element division and sequence texture generation, the simulation of the texture sequence with a structural trend under a moving platform can be achieved.

[0060] S200, calculate the instantaneous Doppler shift and instantaneous Doppler bandwidth of the sea area to be simulated, and use them as the speckle component under a stationary platform and calculate the Doppler shift and broadening caused by the moving platform;

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

[0062] S210, generate a sequence of speckle components of the sea area to be simulated by calculating the instantaneous Doppler bandwidth of the speckle component;

[0063] Specifically, S210 includes:

[0064] S211, use the wind speed of the sea area to be simulated to 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, and a first-order autoregressive (AR) process is used for a zero-mean, unit-variance complex Gaussian sequence with zero Doppler shift and an instantaneous Doppler bandwidth (Instantaneous Doppler Bandwith, IDBW) affected by the local wind speed and texture.

[0066] Using an empirical formula, the average half-power Doppler bandwidth (Mean Half-Power Doppler Bandwidth, MDBW) can be calculated from the local wind speed U (m / s -1 ):

[0067] MDBW = 0.48U / λ radar

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

[0069] As the texture component increases, the IDBW will decrease accordingly. Therefore, IDBW is modeled as:

[0070]

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

[0072]

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

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

[0075]

[0076] where 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 result in the AR model has IDBW, replace the fixed coefficient ρ with the time-varying coefficient ρ(m). Since ρ(m) that varies with the texture intensity is slow-varying relative to the pulse repetition interval, that is, it remains basically unchanged within 0.1 s, it can ensure 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. Calculate the average Doppler shift of the speckle component sequence, and 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.

[0083] According to the empirical formula, the average Doppler shift (Mean Doppler Shift, MDOS) can be expressed as the wind speed U (m / s -1 ) and the wind direction angle function

[0084]

[0085] Among them, the superscripts "HH" and "VV" represent the polarization modes of the radar transmitter and receiver. In addition, 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 upper and lower bound saturation regions, a sigmoid-like non-linear dependence relationship is used, and at the same time, a positive value factor ε is used to control the degree of fluctuation. 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, add the instantaneous Doppler shift and instantaneous Doppler bandwidth in the speckle component under the stationary platform to the Doppler shift and broadening caused by the moving platform correspondingly to obtain the speckle component under the moving platform;

[0090] In this step, 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.

[0091] The Doppler shift caused by platform motion is jointly determined by the platform motion speed, the azimuth angle, the pitch angle of the beam, and the radar parameters. The appearance of the Doppler broadening phenomenon is caused by the difference in the Doppler shift of the scatterers in the beam from the high-speed moving platform at different azimuth angles. Therefore, the Doppler shift and broadening caused by platform motion can be expressed as

[0092]

[0093] where f dr is the Doppler shift caused by platform motion, B d is the Doppler broadening caused by platform motion, v r is the platform motion speed, is the pitch angle, θ is the azimuth angle, Δθ is the azimuth beam width, λ radar is the radar wavelength.

[0094] Adding the instantaneous Doppler shift of the speckle component and the Doppler shift caused by the moving platform, the Doppler shift of the speckle component under the moving platform can be obtained:

[0095]

[0096] The sea surface is divided into multiple azimuth resolution cells along the azimuth direction, as Figure 3 shown. Each azimuth resolution cell on the same range ring corresponds to a different azimuth angle. According to the platform motion and illumination geometric relationship, the Doppler shift of the echo of each azimuth resolution cell and the Doppler shift 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 the azimuth resolution cells on the same range ring are superimposed, and the speckle component containing the Doppler shift and bandwidth caused by the platform motion can be obtained.

[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 that n sea spikes appear in each resolution cell 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 occurrence position of the sea spike.

[0098] Sea spikes occur in sporadic resolution cells, and this phenomenon can be described by a Poisson process. Assuming that the average number of sea spike occurrences per unit area and per unit time is λ0, the probability that n sea spikes appear in each resolution 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 occurrence probability λ0 of sea spikes is generally less than 10 -5 , and increases with the increase of sea state and grazing angle. λ0 reaches the maximum value when the radar observes against the wind and the minimum value when observing with the wind. When the area of the spatial resolution cell is 500m 2 , the occurrence probability of sea spikes in each resolution cell within 1s is less than 0.005. For high-resolution radar, since sea spikes are caused by near-breaking waves, it is difficult for two or more sea spikes to appear simultaneously in each spatial resolution cell within one pulse. Therefore, we assume that there is at most one sea spike in each resolution cell within one pulse. When simulating sea spikes, at each pulse-range grid point (m, n), a random integer is generated according to the above formula. If the integer is 0, no sea spike occurs at this grid point; otherwise, a sea spike appears in this resolution cell starting from this pulse.

[0101] After determining the occurrence location of sea spikes in the spatio-temporal plane, each sea spike can be described by parameters such as its duration, range spread, slow-time envelope, Peak to Clutter Ratio (PCR), and radial velocity. The distribution laws of some of these parameters can be summarized from measured data. The duration T of the sea spike spike is mainly concentrated within 1 s to 4 s, and the longer the duration of the sea spike, the lower the occurrence probability. The duration of the sea spike can be modeled with a mean of

[0102]

[0103] When the range resolution of the radar reaches the meter level or sub-meter level, sea spikes often occupy multiple adjacent range cells. The number of range cells occupied by the sea spike depends on the radial length of the corresponding near-breaking wave. It can be observed from the measured data that the radial length of the sea spike is concentrated around a certain fixed value, and the occurrence probabilities of sea spikes that are too long or too short both decrease significantly. Therefore, the Nth-order spline distribution is used to model the radial length of the sea spike. The Nth-order spline distribution can be regarded as obtained by the convolution of N uniformly distributed probability density functions, or regarded as the distribution of the sum of N independent and identically distributed uniformly distributed random variables, that is:

[0104]

[0105] where represents the mean of the radial length of the sea spike, with the unit of m. Since the sea spike expands along the range and pulses, the coverage probability of the sea spike in the simulated 3D data is:

[0106]

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

[0108] In addition to the duration and radial length, corresponding statistical models also need to be established for the radial envelope, time envelope, PCR, and radial velocity of the sea spike. Considering that the sea spike is generated by the specular structure on the near-breaking wave, it can be assumed that the envelope of the sea spike is a tent function along the range and an isosceles trapezoid function along the pulse. Assume that the support interval of this two-dimensional function is [0, 1] 2When the function is at a certain instant time slice of a pulse, it is a tent function on the interval [0, 1], and the spatial slice at a certain range cell is an isosceles trapezoid function on the interval [0, 1].

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

[0110] The amplitude A of the sea spike is assumed to follow a uniform distribution on the interval [a, b](E{τ}), 1 / 2 where 1 < a < b, and a and b need to be selected according to sea condition experience during the simulation. The speed of the near-breaking wave is closely related to the swell speed, and the speed of the ocean wave increases with the increase of the wavelength. It can be found from the measured data that the speed of the sea spike is about 0.3 times the speed of the main 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, and v d represents the phase velocity of the main wave in the wave field. The main wave speed can be calculated using the time-varying sea surface parameters during the simulation.

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

[0114]

[0115] where, represents the random initial phase on the interval [-π, π). The first term in the above formula represents the envelope of the sea spike in the two-dimensional range-time plane, and the second term represents the phase sequence, where it is assumed that all range cells occupied by each 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, using the texture component, the speckle component under the moving platform, and the sea spike model, construct the full-dimensional local sea clutter in the sea area to be simulated under the high-speed moving platform.

[0118] When performing sea clutter simulation for a moving platform, it is necessary to comprehensively consider large-scale dynamic sea surface generation, empirical formulas for sea surface scattering coefficients, sea clutter texture distribution, spatio-temporal varying sea clutter Doppler characteristics, empirical statistical characteristics of sea spikes, and the influence of platform motion on the above characteristics. The format of the simulation results is

[0119]

[0120] which mainly includes continuous texture components, speckle components, and discrete sea spike components. Among them, 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 sea spike that randomly occurs in the l k th wave position N k continuous range cells and lasts for M k continuous pulses.

[0121] The present invention does not have high requirements for measured data and computing resources. Based on the combination of physical formulas and statistical assumptions, the simulation results are more in line with the actual situation, can provide a realistic clutter model environment, and support the design and performance evaluation of radar systems. By adding structural trends and sea spike components to the inverse Gamma distribution composite Gaussian model, the present invention gives a simulation model for sea surface echo description with meter-level or sub-meter-level resolution, which is of great significance for the upgrade and transformation of moving platform radar systems and the performance test of detection and tracking algorithms.

[0122] Reference Figure 4 , Figure 4 is the simulation process of the full-dimensional sea clutter simulation method for the high-speed moving platform of the present invention. In order to prove the effectiveness of the method, the present invention analyzes the simulation results at a set of resolutions according to the Figure 4 simulation process, and the analysis content mainly includes the rationality of texture trends and Doppler spectra. Figures 5 - 7 The power diagrams and average Doppler spectra of the simulation data in several cases with different moving speeds are given respectively. In the three groups of simulation data with a resolution of 3m, obvious texture trends can be seen, and with the increase of the platform speed, the texture trends show obvious offsets caused by platform motion. When the platform is stationary, the Doppler bandwidth of the clutter can be calculated to be about 70Hz by the formula in step 2. When the platform moving speeds are 50m / s and 100m / s respectively, the Doppler bandwidths can be calculated to be 38Hz and 76Hz by the formula for the moving platform in step 2, which is basically consistent with the simulation results.

[0123] It should be noted that the terms "first" and "second" in the present invention are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0124] Although the present application has been described in conjunction with various embodiments herein, however, in the process of implementing the claimed present application, those skilled in the art can understand and achieve other variations of the disclosed embodiments by viewing the accompanying drawings, the disclosure content, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "one" does not exclude a plurality of cases.

[0125] The above content is a further detailed description of the present invention in conjunction with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is only limited to these descriptions. For those of ordinary skill in the technical field to which the present invention pertains, without departing from the concept of the present invention, several simple deductions or substitutions can be made, and all should be regarded as belonging to the protection scope of the present invention.

Claims

1. A full-dimensional local sea clutter simulation method under a high-speed motion platform, characterized in that: include: S100, determining a sea area to be simulated, and generating a texture component of the sea area to be simulated; S200, calculating the instantaneous Doppler shift and instantaneous Doppler bandwidth of the sea area to be simulated, and using them as the speckle component under the stationary platform and calculating the Doppler shift and broadening caused by the moving platform; S300, correspondingly adding the instantaneous Doppler shift and the instantaneous Doppler bandwidth in the speckle component under the stationary platform and the Doppler shift and the broadening caused by the moving platform to obtain the speckle component under the moving platform; S400, calculating sea peak parameters of the sea area to be simulated, and generating a sea peak model using the sea peak parameters; S500, constructing full-dimensional local sea clutter of the sea area to be simulated under the high-speed moving platform by using the texture component, the speckle component under the moving platform and the sea peak model.

2. The method for simulating local sea clutter in all dimensions under a high-speed motion platform according to claim 1, characterized in that: S100 includes: S110, determining a sea area to be simulated and dividing it into a plurality of facets; S120, calculating the RCS of each surface element; S130, using the RCS of each face element and a random sequence obeying an inverse Gamma distribution, generating a texture component of the sea area to be simulated.

3. The full-dimensional local sea clutter simulation method under a high-speed motion platform according to claim 2 is characterized in that: S110 includes: A sea area to be simulated is determined, and the sea area to be simulated is segmented using a surface analysis method to obtain a sea surface with a predetermined sampling interval, wherein the sea surface is composed of a plurality of facets.

4. The method for simulating local sea clutter in all dimensions under a high-speed motion platform according to claim 2, characterized in that: S120 includes: S121, taking the rubbing angle of each facet as an input of a TCS model, calculating a normalized scattering coefficient through the TCS model, and calculating the RCS of each facet using the normalized scattering coefficient and the facet area; S121, using the RCS of each facet to calculate the RCS of each spatial resolution unit, and composing them into a time-varying RCS sequence; the time-varying RCS sequence reflects the space-time variation of the sea clutter texture trend.

5. The method for simulating local sea clutter in all dimensions under a high-speed motion platform according to claim 4, characterized in that: S130 includes: S131, fusing the order of the time-varying RCS sequence with a random sequence obeying an inverse Gamma distribution to obtain a texture component of a spatial resolution unit; S132, combining the texture components of all spatial resolution units into texture components of the sea area to be simulated.

6. The full-dimensional local sea clutter simulation method under a high-speed motion platform according to claim 1, characterized in that: S200 includes: S210, generating a speckle component sequence of the sea area to be simulated by calculating the instantaneous Doppler bandwidth of the speckle component; S220, calculating the instantaneous Doppler shift of the speckle component sequence.

7. The method for simulating local sea clutter in all dimensions under a high-speed motion platform according to claim 6, characterized in that: S210 includes: S211, calculating the average half-power Doppler bandwidth of the speckle component using the wind speed of the sea area to be simulated; S212, calculating the instantaneous Doppler bandwidth and the time-varying coefficient of the speckle component by using the average half-power Doppler bandwidth; S213, generating a low-pass Gaussian distribution random sequence using a first-order AR process; the low-pass Gaussian distribution random sequence includes a hysteresis coefficient; S214, replacing the hysteresis coefficient with the time-varying coefficient to obtain a low-pass Gaussian distribution random sequence whose instantaneous Doppler bandwidth converges to a theoretical value, and using the sequence as a speckle component sequence.

8. The method for simulating local sea clutter in all dimensions under a high-speed motion platform according to claim 6, characterized in that: S220 includes: S221, calculating an average Doppler shift of a speckle component sequence, where the average Doppler shift is expressed as a function of wind speed and wind direction angle; S222: Calculate an instantaneous Doppler shift using the average Doppler shift.

9. The method for simulating local sea clutter in all dimensions under a high-speed motion platform according to claim 1, characterized in that: 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.

10. The method for simulating local sea clutter in all dimensions under a high-speed motion platform according to claim 1, characterized in that: The sea peak 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 occurrence position of the sea spike.

Images