A multi-scale second-order autoregressive recursive filter approach for the sea ice concentration analysis

To effectively extract multi-scale information from observation data and improve computational efficiency,a multi-scale second-order autoregressive recursive filter(MSRF)method is designed.The second-order autoregressive filter used in this study has been attempted to replace the traditional first-order recursive filter used in spatial multi-scale recursive filter(SMRF)method.The experimental results indicate that the MSRF scheme successfully extracts various scale information resolved by observations.Moreover,compared with the SMRF scheme,the MSRF scheme improves computational accuracy and efficiency to some extent.The MSRF scheme can not only propagate to a longer distance without the attenuation of innovation,but also reduce the mean absolute deviation between the reconstructed sea ice concentration results and observations reduced by about 3.2%compared to the SMRF scheme.On the other hand,compared with traditional first-order recursive filters using in the SMRF scheme that multiple filters are executed,the MSRF scheme only needs to perform two filter processes in one iteration,greatly improving filtering efficiency.In the two-dimensional experiment of sea ice concentration,the calculation time of the MSRF scheme is only 1/7 of that of SMRF scheme.This means that the MSRF scheme can achieve better performance with less computational cost,which is of great significance for further application in real-time ocean or sea ice data assimilation systems in the future.

Persymmetric adaptive polarimetric detection of subspace range-spread targets in compound Gaussian sea clutter

This paper focuses on the adaptive detection of range and Doppler dual-spread targets in non-homogeneous and nonGaussian sea clutter.The sea clutter from two polarimetric channels is modeled as a compound-Gaussian model with different parameters,and the target is modeled as a subspace rangespread target model.The persymmetric structure is used to model the clutter covariance matrix,in order to reduce the reliance on secondary data of the designed detectors.Three adaptive polarimetric persymmetric detectors are designed based on the generalized likelihood ratio test(GLRT),Rao test,and Wald test.All the proposed detectors have constant falsealarm rate property with respect to the clutter texture,the speckle covariance matrix.Experimental results on simulated and measured data show that three adaptive detectors outperform the competitors in different clutter environments,and the proposed GLRT detector has the best detection performance under different parameters.

Study of scattering from time-varying Gerstners sea surface using second-order small slope approximation

Backscattered fields from one-dimensional time-varying Gerstners sea surface are calculated utilising the secondorder small slope approximation. It is well known that spectral properties of the backscattered echoes relate to the velocity of the small elementary scatterers on sea surface profiles. Therefore, modeling Doppler spectra from the ocean requires an accurate description of the sea surface motion. The profile of nonlinear Gerstners sea surface shows verticalskewness of sea waves, it is sharper at the crest and flatter at the trough than linear waves, and its maximum slope position is closer to the crest than to the trough. Furthermore, the horizontal component of the small elementary scatterers orbit velocity on the sea surface, which yields noticeable influence on Doppler spectra, can be obtained conveniently by Gerstners sea surface model. In this study the characteristics of Doppler spectra of backscattered fields from time-varying Gerstners sea surface are investigated and the dependences of the Doppler frequency and the Doppler bandwidth on the parameters, such as the wind speed, the radar frequency, the incident angle, etc. are discussed. It is shown that the Doppler bandwidth of microwave scattered fields from Gerstners sea surface is considerably broadened. For the case of high frequency backscattered fields, the values of the higher-order spectrum peaks are larger than those obtained by linear sea surface.

Study on Dim Target Detection and Discrimination from Sea Clutter

Dim target detection from sea clutter is one of the difficult topics in ocean remote sensing application. By aiming at the shortcoming of false alarms when using track before detect （TBD） based on dynamic programming, a new discrimination method called statistics of direction histogram （SDH） is proposed, which is based on different features of trajectories between the true target and false one. Moreover, a new series of discrimination schemes of SDH and Local Extreme Value method （LEV） are studied and applied to simulate the actually measured radar data. The results show that the given discrimination is effective to reduce false alarms during dim targets detection.

Modified GIT model for predicting wind-speed behavior of low-grazing-angle radar sea clutter

A modified GIT model for describing the variational trend of mean clutter reflectivity as a function of wind speed is proposed. It uses two slope adjustment factors and two critical wind-speed factors to define and adjust the increasing slope of reflectivity with respect to wind speed. In addition, it uses a constant factor to compensate the overall amplitude of clutter reflectivity. The performance of the modified GIT model has been verified on the basis of the L-band low-grazing-angle radar sea clutter data. The results are in good agreement with the experimental data, indicating that the model is more effective in predicting the wind-speed behavior of clutter reflectivity than the conventional GIT model, especially for lower and higher wind speeds. We believe that the proposed model can provide deeper insights into the relationship between radar sea clutter reflectivity and sea state conditions.

Estimation of sea clutter inherent Doppler spectrum from shipborne S-band radar sea echo

Measurement of shipborne radar sea echo is instrumental in collecting the sea clutter data in open sea areas.However,the ship movement would introduce an extra Doppler component into the spectrum of the sea clutter,so the sea clutter inherent spectrum must be estimated prior to investigating the sea clutter Doppler characteristics from the shipborne radar sea echo.In this paper we show some results about a shipborne sea clutter measurement experiment that was conducted in the South China Sea in a period between 2017 and 2018;abundant clutter data have been collected by using a shipborne S-band clutter measurement radar.To obtain the sea clutter inherent Doppler spectrum from these data,an estimation method,based on the mapping relationship between the shipborne clutter spectrum and the inherent clutter spectrum,is proposed.This method is validated by shipborne clutter data sets under the same measuring conditions except for the ship speed.Using this method,the characteristics of the Doppler spectrum lineshapes in the South China Sea are calculated and analyzed according to different sea states,wave directions,and radar resolutions,which can be instrumental in designing the radar target detection algorithms.

Fractal-based weak target detection within sea clutter

The target on the sea surface is complex and difficult to detect due to the interference of backscattered returns from the sea surface illuminated by the radar pulse. Detrended fluctuation analysis （DFA） has been used successfully to extract the time-domain Hurst exponent of sea-clutter series. Since the frequency of the sea clutter mainly concentrates around Doppler center so that we consider to extract frequency-do- main fractal characterization and then detect a weak target within sea clutter by using the difference of frequency-domain fractal characterization. The generalized detrended fluctuation analysis （GDFA） is more flexible than traditional DFA owing to its smoothing action for the clutters. In this paper, we apply the GDFA to evaluate the generalized Hurst exponent of sea-clutter series in the frequency domain. The difference of generalized Hurst exponents between different sea-clutter range bins would be used to determine whether the target exists. Moreover, some simulations with the real IPIX radar data have also been demonstrated in order to suooort this conclusion.

A statistical distribution of quad-pol X-band sea clutter time series acquired at a grazing angle

Although the complex Wishart distribution has been widely used to analyze the statistic properties of quad-pol SAR spatial data, the applicability of this complex distribution to the time series of sea clutter is rarely discussed.The measured data of the quad-pol X-band marine radar demonstrate that the time series of the sea echoes are also satisfied the circular Gaussian distributions if the low intensity signals, which are mainly dominated by a radar noise, in the shadow regions of the large-scale waves are removed. On the basis of this fact, the probability density functions（PDFs） of the intensity as well as the phase, the real and the imaginary parts of the sea echoes obtained by the marine radar have been derived, and the theoretical models are all expressed in closed forms. In order to validate the theoretical results, the PDFs are compared with the experimental data collected by the Mc Master IPIX radar. And the comparisons show that the PDF models are in good agreement with the experimental data.

Bayesian track-before-detect algorithm for nonstationary sea clutter

Radar detection of small targets in sea clutter is a particularly demanding task because of the nonstationary characteristic of sea clutter.The track-before-detect(TBD)filter is an effective way to increase the signal-to-clutter ratio(SCR),thus improving the detection performance of small targets in sea clutter.To cope with the nonstationary characteristic of sea clutter,an easily-implemented Bayesian TBD filter with adaptive detection threshold is proposed and a new parameter estimation method is devised which is integrated into the detection process.The detection threshold is set according to the parameter estimation result under the framework of information theory.For detection of closely spaced targets,those within the same range cell as the one under test are treated as contribution to sea clutter,and a successive elimination method is adopted to detect them.Simulation results prove the effectiveness of the proposed algorithm in detecting small targets in nonstationary sea clutter,especially closely spaced ones.

Effective Sea Clutter Spectrum Extraction Method for HFSWR in Adverse Conditions

An effective approach in solving the sea clutter spectrum extraction problem is studied in the paper.Different from the conventional signal to noise ratio（SNR）method based on Doppler frequency or range domain information,a method is developed to characterize the differences between the sea echo and those interferences are by signal to interference plus noise ratio（SINR）which jointly utilizing the range,Doppler frequency and azimuth domain information.Furthermore,these differences can be adaptable to adverse conditions by forming the necessary boundaries and constraints in searching of the maximum SINR,which greatly promotes the extraction of sea clutter spectrum.The real high frequency surface wave radar（HFSWR）data demonstrate that the proposed method is less influenced by those interferences and can effectively extract the sea clutter spectrum even under the adverse conditions.Furthermore,it has been shown as an effective method for ship detection and sea state remote sensing of HFSWR.

Effect of Changes in Sea-Surface State on Statistical Characteristics of Sea Clutter with X-Band Radar

We have made observations of X-band radar sea clutter from the sea surface and sea-surface state in the Uraga Suido Traffic Route, which is used by ships entering and leaving Tokyo Bay, and the nearby Daini Kaiho Sea Fortress. We estimated the distributions of reflected amplitudes due to sea clutter using models that assume Weibull, Log-Weibull, Log-normal, and K-distributions. We then compared the results of estimating these distributions with sea-surface state data to investigate the effects of changes in the sea-surface state on the statistical characteristics of sea clutter. As a result, we showed that observed sub-ranges not containing a target conformed better to the Weibull distribution regardless of Significant Wave Height (SWH). Further, sub-ranges conforming to the Log-Weibull or Log-normal distribution in areas contained a target when the SWH was large, and as SWH decreases, sub-ranges conforming to a Log-normal. We also showed that for observed sub-ranges not containing a target, the shape parameter, c, of both Weibull and Log-Weibull distribution correlated with SWH. The correlation between wave period and shape parameters of Weibull and Log-Weibull distribution showed a weak correlation.

Distributed Sea Clutter Denoising Algorithm Based on Variational Mode Decomposition

In order to improve the detection accuracy of chaotic small signal prediction models under the background of sea clutter,a distributed sea clutter denoising algorithm is proposed,on the basis of variational modal decomposition(VMD).The sea clutter signal is decomposed into variational modal functions(VMF)with different center bandwidths by means of VMD.By analyzing the autocorrelation characteristics of the deco mposed signal,we perform instantaneous half-period(IHP)and wavelet threshold denoising processing on the high-frequency and low-frequency components respectively,and regain the sea clutter signals.Based on LSSVM sea clutter prediction model,this research compares and analyzes the denoising effects of VMD.Experi ment results show that,the RMSE after denoising is reduced by two orders of magnitude,approximating 0.00034,with an apparently better denoising effect,compared with the root mean square error(RMSE)of the prediction before denoising.

The multifractal spectrum of a sea clutter using a random walk model

The radar echo signal of a sea clutter is nonlinear, nonstationary and time varying. A multifractal measure analysis can describe the local singularity of a physics system. The random walk model of a sea clutter scattering is analysed to disclose the intrinsic physical characteristics and laws of the sea clutter. Stochastic differential equations are given for the physical quality of the sea clutter. A diffusion process model is established using It o formula. The singularity of the random walk model is tested by a multifractal spectroscopy, and the accuracy of this model is proven by the multifractal spectroscopy of a real-life IPIX radar data set. Thus, the random walk model is effective for describing the dynamics mechanism of the sea clutter.

First order sea clutter cross section for bistatic shipborne HFSWR

This paper studies the development on the first order sea clutter cross section for bistatic high frequency surface wave radar (HFSWR). Based on the received first order electric field expression, a closed-form of cross sections is derived to account for the case of receiving antenna array being mounted on the shipborne platform. The uniform linear motion and sway motion components are assumed to be responsible for the observed differences in comparison with the bistatic fixed antenna case. Correspondingly, simulations are conducted to study the sea clutter spectral characteristics for these two cases versus different system parameters and sea state conditions. It is shown numerically that the forward motion component will spread the Bragg lines severely and the influence triggered by the sway motion can be explained as the Bessel function modulation of the ordinary sea clutter spectra. The obtained results have important implications in the application of shipborne HFSWR technology to ocean remote sensing and target detection.

Reflectivity Model of Low Grazing Angle Radar Sea Clutter

The commonly used reflectivity models of radar sea clutter are summarized. Among these models, the adjusted Barton model and the adjusted Morchin model are compared. From the analysis result, the γ-p reflectivity model is presented for low grazing angle radar sea clutter by the adjustment of the original Barton reflectivity model. The model takes into account radar frequency, grazing angle, sea condition, and polarization property. The influences of these factors on the proposed model are analyzed. The model absorbs the merits from commonly used reflectivity models for sea clutter. It introduces several researchers' opinions, and extends them. And it accounts for the reflectivity at arbitrary radar frequency from VHF to X-band, arbitrary low grazing angle, arbitrary sea condition and different polarization property. One of the main results is the proposed γ-p reflectivity model can reflect the influence of polarization on sea clutter reflectivity to some extent. The proposed γ-p reflectivity model of low-angle radar-sea clutter is validated by comparing the simulated and statistically experimental data.

基于残差单元与注意力门的非对称编解码海杂波抑制网络

针对非均匀海杂波环境下弱小目标检测困难的问题,本文基于复值残差单元和注意力门机制,提出一种用于海杂波抑制的非对称编解码网络(Asymmetric Encoder-Decoder Network,AED-Net).该网络以雷达回波经匹配滤波后得到的复值信号为输入,利用复值残差单元取代常规卷积单元进行弱小目标和海杂波特征的提取,增强网络特征提取能力的同时避免特征信息退化.然后采用注意力门模块将编码路径各模块提取的特征信息分别送入到解码路径对应的模块.最终输出海杂波抑制后的复值信号.由于各注意力门的输入和输出维度可根据网络结构自主选择,该网络设计是一种非对称编解码结构.与典型对称编解码网络UNet相比,复值残差单元与注意力门的引入显著降低了特征信息的冗余度,增强特征信息的提取与传递,提升了海杂波抑制性能.与此同时,复值残差单元的参数规模远小于卷积单元,而注意力门的引入也有效减少解码路径单元的数量,整个网络的参数规模显著减小.基于海杂波实测数据的实验结果表明,与典型复值UNet(Complex Value-UNet,CV-UNet)网络相比,AED-Net的输出信杂比平均提升9 dB,有效工作的最低信杂比降低了3 dB,模型参数量和计算量分别减少57.8%、50%.

基于WGAN-GP-CNN的海面小目标检测

针对传统基于统计理论的海面小目标检测方法在复杂海面环境中性能不高的问题,该文提出了一种改进的检测方法。首先通过分析海杂波和目标回波的特征,将检测问题转化为特征空间的分类任务。鉴于海面小目标样本数量有限,存在样本不平衡的问题,该文引入了一种基于梯度惩罚的沃瑟斯坦生成对抗网络(Wasserstein Generative Adversarial Network with Gradient Penalty,WGAN-GP)来增强目标数据,从而在数量上平衡目标样本与海杂波样本。同时,对原始WGAN-GP网络的损失函数进行了改进,引入相位损失以确保生成数据能够反映真实数据的相位信息。基于这些数据,进一步提取了生成目标和海杂波的高维特征,并将其送入卷积神经网络(Convolutional Neural Network,CNN)进行训练。为了应对高维特征空间中虚警概率难以控制的问题,对CNN算法进行了改进,通过设置Softmax分类器的阈值,实现了虚警概率可控。最后,借助公开的IPIX雷达数据集进行实验验证,所提的WGAN-GP-CNN检测器在积累时间为1.024 s,虚警概率为0.001时,平均检测概率达到0.8683,具有良好的检测效果。

基于无监督图互信息最大化的海面小目标异常检测

受到复杂海洋环境的影响,雷达对海面慢速小目标难以实现高性能检测。对于这类目标,传统的基于能量的统计检测方法存在着严重的性能损失。针对这一问题,该文提出了基于互信息最大化框架下的海面小目标检测方法,实现海杂波背景下的无监督目标异常检测任务。首先,考虑到高分辨雷达回波不满足传统神经网络对样本独立同分布的假设,该文从图的角度重新建模数据,利用回波的空时相关特性来构建图拓扑结构。该文提出相对最大节点度并联合7个已有特征作为节点的初始表示向量。接下来,采用图注意力网络作为互信息最大化框架中的编码器学习节点表示向量。最后,使用异常检测算法进行目标检测,并实现虚警可控。经实测数据验证,使用快速凸包学习算法时,相比三特征检测器,所提检测器性能提升了9.2%;相比时频三特征检测器,性能提升了7.9%。当网络输出更高维的表示向量时,使用孤立森林算法的检测器的性能提升了27.4%。

大掠射角对海雷达导引头实测回波特性分析

大掠射角下,对海雷达导引头观测到的海杂波强度激增,舰船目标检测困难。本文结合全极化雷达导引头挂飞实测数据,分析了回波幅度和相位特性。首先,采用典型分布对回波幅度分布进行拟合,并利用拟合优度对拟合效果作出评估;然后,分析了平均相位变化量。结果表明:在大掠射角条件下,雷达导引头各极化通道海杂波幅度服从K分布;海杂波与舰船回波幅度在同极化通道服从K分布,在交叉极化通道服从对数正态分布;交叉极化通道回波信杂比高于同极化通道;海杂波与舰船回波相位平均变化量小于海杂波。

基于改进DeeplabV 3+的HFSWR电离层杂波及海杂波自动识别

针对高频地波雷达(HFSWR)回波谱中背景噪声复杂、杂波占比较小且电离层杂波形态位置各异,难以自动识别的问题,以DeeplabV3+深度学习算法为基础架构并加以改进,提出一种HFSWR电离层杂波及海杂波自动识别方法。选用轻量级MobileNetV2作为主干特征网络,加入通道注意力机制模块SENet,实现对杂波标签的侧重学习,优化训练集中各类标签的损失权重;采用模型预训练迁移法对网络进行预训练,解决样本空间过小的问题。实测数据集上的实验结果表明,本方法可以实现HFSWR电离层杂波及海杂波的自动识别。相比原DeeplabV3+算法,杂波识别结果更为准确和精细,海杂波识别结果的平均交并比(mIoU)和准确率(ACC)分别提升2.9%和5.1%;电离层杂波识别结果的mIoU和ACC分别提升3.0%和4.9%。