Space-time autoregressive filtering for matched subspace STAP

Practical space-time adaptive processing (STAP) implementations rely on reduced-dimension processing, using techniques such as principle components or partially adaptive filters. The dimension reduction not only decreases the computational load, it also reduces the sample support required for estimating the interference statistics. This results because the clutter covariance is implicitly assumed to possess a certain (nonparametric) structure. We demonstrate how imposing a parametric structure on the clutter and jamming can lead to a further reduction in both computation and secondary sample support. Our approach, referred to as space-time autoregressive (STAR) filtering, is applied in two steps: first, a structured subspace orthogonal to that in which the clutter and interference reside is found, and second, a detector matched to this subspace is used to determine whether or not a target is present. Using a realistic simulated data set for circular array STAP, we demonstrate that this approach achieves significantly lower signal-to-interference plus noise ratio (SINR) loss with a computational load that is less than that required by other popular approaches. The STAR algorithm also yields excellent performance with very small secondary sample support, a feature that is particularly attractive for applications involving nonstationary clutter.     

Multiple target detection using modified high order correlations

This work is concerned with the problem of multiple target track detection in heavy clutter. Using the “modified high order correlation” (HOC) process and a track scoring mechanism a new method is developed to perform data association and track identification in the presence of heavy clutter. Using this new scheme any number of very close, crossing or splitting target tracks can be resolved without increasing the computational complexity of the algorithm. The applicability of the method for continuous detection of target tracks that can originate and terminate at any scan is also demonstrated, In addition, the operating characteristics as a function of the clutter density are also provided. Simulation results on all the cases are presented     

Dim target detection using high order correlation method

This work presents a method for clutter rejection and dim target track detection from infrared (IR) satellite data using neural networks. A high-order correlation method which recursively computes the spatio-temporal cross-correlations between data of several consecutive scans is developed. The implementation of this scheme using a connectionist network is presented. Several important properties of the high-order correlation method which indicate that the resultant filtered images capture all the target information are established. The simulation results obtained with this approach show at least 93% clutter rejection. Further improvement in the clutter rejection rate is achieved by modifying the high-order correlation method to incorporate the target motion dynamics. The implementation of this modified high-order correlation using a high-order neural network architecture is demonstrated. The simulation results indicate at least 97% clutter rejection rate for this method. A comparison is also made between the methods developed here and the conventional frequency domain three-dimensional (3-D) filtering scheme, and the simulation results are provided     

Improving performance in pulse radar detection using neuralnetworks

A new approach using a multilayered feed forward neural network for pulse compression is presented. The 13 element Barker code was used as the signal code. In training this network, the extended Kalman filtering (EKF)-based learning algorithm which has faster convergence speed than the conventional backpropagation (BP) algorithm was used. This approach has yielded output peak signal to sidelobe ratios which are much superior to those obtained with the BP algorithm. Further, for use of this neural network for real time processing, parallel implementation of the EKF-based learning algorithm is indispensable. Therefore, parallel implementation has also been developed     

利用神经网络的海杂波幅度分布参数估计方法

海杂波是制约对海雷达探测性能的主要因素之一,掌握其特性,具有十分重要的意义。经典海杂波统计模型在参数估计方法上以传统统计学理论为基础,在样本数较少的情况下,估计结果往往较差,导致建模准确度下降。此外,在复杂非均匀探测背景下,难以实现海杂波模型参数的准确实时估计。针对该问题,文章将深度神经网络模型引入海杂波参数估计领域,通过构建合理的模型,使其具备海杂波幅度分布模型的高精度参数估计能力。该方法采用直方图统计的方法进行数据预处理,合理划分输入数据标签的分组区间,构建数据集训练神经网络,并利用测试数据得到神经网络估计结果。仿真数据和X波段IPIX雷达实测数据验证结果表明,与传统数理统计估计方法相比,该算法明显提升了海杂波统计模型参数估计精度。     

Sensor registration using neural networks

One of the major problems in multiple sensor surveillance systems is inadequate sensor registration. We propose a new approach to sensor registration based on layered neural networks. The nonparametric nature of this approach enables many different kinds of sensor biases to be solved. As part of the implementation we develop some modifications to the common network training algorithm to tackle the inherent randomness in all components of the training set     

基于点集合并的修正Hough变换TBD算法

在低信噪比条件下,基于Hough变换的检测前跟踪算法是进行强杂波背景下目标航迹检测的一种手段。本文针对Hough变换后一个目标产生多条可能航迹以及航迹内可能存在杂波点的问题,提出了一种基于能量最大点和点集合并的修正Hough变换检测前跟踪算法。该算法利用量测点时序、能量信息及目标速度先验信息对Hough变换后点迹进行关联和剔除,能够有效的对目标原始航迹进行回溯。针对高斯噪声背景下的飞行目标,仿真结果表明该算法能够对微弱目标进行有效检测,在目标数目、杂波密度、信噪比发生变化的条件下仍能保持较高的检测概率。     

Adaptive Radar Detection in Doubly Nonstationary Autoregressive Doppler Spread Clutter

The problem of adaptive radar detection in clutter which is nonstationary both in slow and fast time is addressed. Nonstationarity within a coherent processing interval (CPI) often precludes target detection because of the masking induced by Doppler spreading of the clutter. Across range bins (i.e., fast time), nonstationarity severely limits the amount of training data available to estimate the noise covariance matrix required for adaptive detection. Such difficult clutter conditions are not uncommon in complex multipath propagation conditions where path lengths can change abruptly in dynamic scenarios. To mitigate nonstationary Doppler spread clutter, an approximation to the generalized likelihood ratio test (GLRT) detector is presented wherein the CPI from the hypothesized target range is used for both clutter estimation and target detection. To overcome the lack of training data, a modified time-varying autoregressive (TVAR) model is assumed for the clutter return. In particular, maximum likelihood (ML) estimates of the TVAR parameters, computed from a single snapshot of data, are used in a GLRT for detecting stationary targets in possibly abruptly nonstationary clutter. The GLRT is compared with three alternative methods including a conceptually simpler ad hoc approach based on extrapolation of quasi-stationary data segments. Detection performance is assessed using simulated targets in both synthetically-generated and real radar clutter. Results suggest the proposed GLRT with TVAR clutter modeling can provide between 5–8 dB improvement in signal-to-clutter plus noise ratio (SCNR) when compared with the conventional methods.     

STOCHASTIC NEURAL NETWORK AND ITS APPLICATION TO MULTI－MANEUVERING TARGET TRACKING

ＳＴＯＣＨＡＳＴＩＣＮＥＵＲＡＬＮＥＴＷＯＲＫＡＮＤＩＴＳＡＰＰＬＩＣＡＴＩＯＮＴＯＭＵＬＴＩ－ＭＡＮＥＵＶＥＲＩＮＧＴＡＲＧＥＴＴＲＡＣＫＩＮＧＪｉｎｇＺｈｏｎｇｌｉａｎｇ；ＤａｉＧｕａｎｚｈｏｎｇ；ＴｏｎｇＭｉｎｇａｎ；ＺｈｏｕＨｏｎｇｒｅｎ（Ｄ...     

Stap using knowledge-aided covariance estimation and the fracta algorithm

In the airborne space-time adaptive processing (STAP) setting, a priori information via knowledge-aided covariance estimation (KACE) is employed in order to reduce the required sample support for application to heterogeneous clutter scenarios. The enhanced FRACTA (FRACTA.E) algorithm with KACE as well as Doppler-sensitive adaptive coherence estimation (DS-ACE) is applied to the KASSPER I & II data sets where it is shown via simulation that near-clairvoyant detection performance is maintained with as little as 1/3 of the normally required number of training data samples. The KASSPER I & II data sets are simulated high-fidelity heterogeneous clutter scenarios which possess several groups of dense targets. KACE provides a priori information about the clutter covariance matrix by exploiting approximately known operating parameters about the radar platform such as pulse repetition frequency (PRF), crab angle, and platform velocity. In addition, the DS-ACE detector is presented which provides greater robustness for low sample support by mitigating false alarms from undernulled clutter near the clutter ridge while maintaining sufficient sensitivity away from the clutter ridge to enable effective target detection performance     

一种基于修正Hough变换的航迹起始算法

为了在重杂波区内检测出运动的目标,提出了一种修正的Hough变换算法用于初始航迹的建立。与传统的算法相比,修正算法充分利用目标的运动学信息,选取了更为合适的变换参量,仅利用较少拍数的量测就可以完成起始,能够在检测概率较高的环境下具有良好的起始性能。计算机仿真结果表明,算法能够克服传统算法的多拍扫描,大大缩短实际的计算量和起始时间,实现对航迹起始的在线改进。     

修正的概率数据互联算法

阐明了概率数据互联(PDA)算法能很好地解决密集环境下的目标跟踪问题,在该算法基础上,人们又提出了联合概率数据互联(JPDA)算法和一些基于 PDA 的修正算法。在概率数据互联算法中,有一个很重要的参数就是杂波数密度(或波门内虚假量测期望数)。然而在许多实际情况中,这个参数是很难获取的。针对这一问题,文中提出了一种修正的概率数据互联算法,该算法通过实时地调整这一参数来获得对目标较为准确的估计结果。最后,给出了算法的仿真分析。     

密集杂波环境下的数据关联快速算法

基于联合概率数据互联（JPDA）的思想,提出了一种新的数据关联快速算法（Fast Al-gorithm for Data Association,简称FAFDA算法）.该方法不需象在最优JPDA算法中那样生成所有可能的联合互联假设,因而具有计算量小,易于工程实现的特点。仿真结果表明,与最优JPDA算法相比,FAFDA算法的跟踪性能令人满意,并且在密集杂波环境下可实时、有效地跟踪100批次以上的目标。     

基于IMF能量分布重构的目标检测技术

为提高海杂波中慢速目标的检测性能,提出了一种基于IMF能量分布重构的目标检测技术。该算法对原始信号尖峰区域经经验模态分解后得到的固有模态函数进行分段数据重构,计算前端IMF分量与后端IMF分量的能量比,并将其输入非参量检测器中进行目标检测。研究表明,相比于海杂波单元,目标单元尖峰区域有更小的前后端IMF分量能量比,适用于慢速目标的检测。     

Neural solution to the multitarget tracking data associationproblem

The problem of tracking multiple targets in the presence of clutter is addressed. The joint probabilistic data association (JPDA) algorithm has been previously reported to be suitable for this problem in that it makes few assumptions and can handle many targets as long as the clutter density is not very high. However, the complexity of this algorithm increases rapidly with the number of targets and returns. An approximation of the JPDA that uses an analog computational network to solve the data association problem is suggested. The problem is viewed as that of optimizing a suitably chosen energy function. Simple neural-network structures for the approximate minimization of such functions have been proposed by other researchers. The analog network used offers a significant degree of parallelism and thus can compute the association probabilities more rapidly. Computer simulations indicate the ability of the algorithm to track many targets simultaneously in the presence of moderately dense clutter     

Classification of radar targets using synthetic neural networks

Radar target classification performance of neural networks is evaluated. Time-domain and frequency-domain target features are considered. The sensitivity of the neural network algorithm to changes in network topology and training noise level is examined. The problem of classifying radar targets at unknown aspect angles is considered. The performance of the neural network algorithms is compared with that of decision-theoretic classifiers. Neural networks can be effectively used as radar target classification algorithms with an expected performance within 10 dB (worst case) of the optimum classifier     

EM-ML algorithm for track initialization using possibly noninformative data

Initializing and maintaining a track for a low observable (LO) (low SNR, low target detection probability and high false alarm rate) target can be very challenging because of the low information content of measurements. In addition, in some scenarios, target-originated measurements might not be present in many consecutive scans because of mispointing, target maneuvers, or erroneous preprocessing. That is, one might have a set of noninformative scans that could result in poor track initialization and maintenance. In this paper an algorithm based on the expectation-maximization (EM) algorithm combined with maximum likelihood (ML) estimation is presented for tracking slowly maneuvering targets in heavy clutter and possibly noninformative scans. The adaptive sliding-window EM-ML approach, which operates in batch mode, tries to reject or weight down noninformative scans using the Q-function in the M-step of the EM algorithm. It is shown that target features in the form of, for example, amplitude information (AI), can also be used to improve the estimates. In addition, performance bounds based on the supplemented EM (SEM) technique are also presented. The effectiveness of new algorithm is first demonstrated on a 78-frame long wave infrared (LWIR) data sequence consisting of an Fl Mirage fighter jet in heavy clutter. Previously, this scenario has been used as a benchmark for evaluating the performance of other track initialization algorithms. The new EM-ML estimator confirms the track by frame 20 while the ML-PDA (maximum likelihood estimator combined with probabilistic data association) algorithm, the IMM-MHT (interacting multiple model estimator combined with multiple hypothesis tracking) and the EVIM-PDA estimator previously required 28, 38, and 39 frames, respectively. The benefits of the new algorithm in terms of accuracy, early detection, and computational load are illustrated using simulated scenarios as well.     

空间非合作目标惯性参数的Adaline网络辨识方法

空间在轨操作中,航天器在对空间非合作目标的抓捕行动常常导致航天器本体的姿态和空间轨迹发生变化。为克服空间非合作目标对航天器本体动力学、运动学的影响,使控制系统做出精准及时的姿控策略调整,确保航天器正常在轨工作和轨迹姿态的高精度,需对抓捕的非合作目标的惯性参数进行辨识。针对传统辨识方法依赖广义逆求解导致的辨识过程运算量大,且数值容易产生剧烈振荡,造成辨识结果不稳定等不足,采用基于归一化最小均方（NLMS）准则的Adaline神经网络方法进行空间非合作目标惯性参数的辨识。首先,基于动量守恒理论建立抓捕后的航天器-机械臂-空间非合作目标系统模型;然后将辨识方程的系数矩阵作为网络的输入和输出,空间非合作目标的惯性参数作为神经网络的训练权重,基于迭代步长可变的NLMS准则实现对目标惯量参数的快速、准确辨识;最后,在构造的ADAMS/MATLAB联合仿真平台上进行了验证。仿真结果表明,基于NLMS准则的Adaline神经网络是一种快速、准确辨识目标惯量参数的有效方法。     

Detection of small objects in clutter using a GA-RBF neural network

Detection of small objects in a radar or satellite image is an important problem with many applications. Due to a recent discovery that sea clutter, the electromagnetic wave backscatter from a sea surface, is chaotic rather than purely random, computational intelligence techniques such as neural networks have been applied to reconstruct the chaotic dynamic of sea clutter. The reconstructed sea clutter dynamical system which usually takes the form of a nonlinear predictor does not only provide a model of the sea scattering phenomenon, but it can also be used to detect the existence of small targets such as fishing boats and small fragments of icebergs by observing abrupt changes in the prediction error. We applied a genetic algorithm (GA) to obtain an optimal reconstruction of sea clutter dynamic based on a radial basis function (RBF) neural network. This GA-RBF uses a hybrid approach that employes a GA to search for the optimum values of the following RBF parameters: centers, variance, and number of hidden nodes, and uses the least square method to determine the weights. It is shown here that if the functional form of an unknown nonlinear dynamical system can be represented exactly using an RBF net (i.e., no approximation error), this GA-RBF approach can reconstruct the exact dynamic from its time series measurements. In addition to the improved accuracy in modeling sea clutter dynamic, the GA-RBF is also shown to enhance the detectability of small objects embedded in the sea. Using real-life radar data that are collected in the east coast of Canada by two different radar systems: a ground-based radar and a satellite equipped with synthetic aperture radar (SAR), we show that the GA-RBF network is a reliable detector for small surface targets in various sea conditions and is practical for real-life search and rescue, navigation, and surveillance applications     

基于多特征图像增强深度卷积神经网络的航天用电子元器件分类算法

为了实现航天用电子元器件的全自动及非接触识别,并减少由照明系统造成的图像亮度不均、偏色等问题对检测结果的影响,通过结合局部、区域和总体三个层次特征提升物体检测精度,提出了一种基于多特征图像增强深度卷积神经网络(MFIE-DCNN)的航天用电子元器件分类算法。MFIE-DCNN算法包含多特征学习和深度学习,其学习过程类似于人类视觉系统,能够对形状、方向和颜色特征进行深度挖掘,突出元器件边界信息,抑制背景杂波干扰。实验结果表明,该算法能够区分电路板板载元器件的种类,检测准确度优于传统算法。对比基于稀疏自动编码器的深度神经网络,检测结果提高了近20%。