Enhanced target detection using stereoscopic imaging radar

An earlier correspondence reported experiments which suggested that the visibility of a target in clutter could be improved through stereoscopic viewing of high resolution radar images. Here we provide a more thorough discussion on the application of stereo for improving radar detection and recognition. Experiments are reported which confirm and extend the earlier reported results. An example of the use of stereo in a practical system is provided which demonstrates the potential for acquisition of high quality radar stereograms     

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     

On radar detection and direction finding using sparse arrays

The potential performance of radar systems employing sparse arrays is investigated. Different sparse array geometries are evaluated with respect to detection and estimation performance. We present appropriate tools for analytical computation of the detection and estimation performance of the likelihood ratio test (LRT) and the maximum likelihood estimator (MLE), respectively. Particular attention is paid to the effect of ambiguity errors on system performance. The results show that sparse arrays have several advantages over nonsparse arrays, but that ambiguities deteriorate performance when sidelobe interference is present. Some ways to mitigate the ambiguity effects on system performance are briefly discussed.     

Target detection and tracking with HF radar using reciprocal SARtechniques

Sustained research and development at Memorial University of Newfoundland has led to an operational High Frequency Ground Wave Radar (HF-GWR) system for coastal surveillance. This radar system has demonstrated over-the-horizon detection of targets such as vessels, ice hazards and low-flying aircraft, and performed ocean parameter measurements over a large area. The industrial developers of offshore hydrocarbon reserves in ice infested regions have an urgent requirement for the long range detection and tracking of icebergs from their production platforms. However, due to space restrictions, a rig- or ship-based system can only accommodate a compact antenna array. The uniform trajectory and low velocity of icebergs is ideal for Reciprocal Synthetic Aperture Radar (RSAR) processing with long target dwell times. The proven ice detection capability of HF-GWR systems, coupled with the compact antennas suggested by the RSAR technique, can be used to develop a rig- or ship-based all-weather surveillance device for ice hazards. It is also anticipated that the results of this research will allow the use of shorter antenna arrays for many other applications. Preliminary results using real data from the operational HF-GWR system are presented     

UWB radar detection of targets in foliage using alpha-stableclutter models

We address the problem of detection of targets obscured by a forest canopy using an ultrawideband (UWB) radar. The forest clutter observed in the radar imagery is a highly impulsive random process that is more accurately modeled with the recently proposed class of alpha-stable processes as compared with Gaussian, Weibull, and K-distribution models. With this more accurate model, segmentation is performed on the imagery into forest and clear regions. Further, a region-adaptive symmetric alpha stable (SαS) constant false-alarm rate (CFAR) detector is introduced and its performance is compared with the Weibull and Gaussian CFAR detectors. The results on real data show that the SαS CFAR performs better than the Weibull and Gaussian CFAR detectors in detecting obscured targets     

Improving intrusion detection radar

The first monostatic microwave intrusion detection sensor with range cutoff was introduced in 1984. This range cutoff circuit as used in the Model 375 and 385 has proven very effective in preventing nuisance alarms beyond a user-defined range. The Intrepid Digital Transceiver introduced in this paper builds upon this proven technology with the addition of a unique digital signal processing routine that measures range to the intruder. Intrepid Digital Transceiver alternately transmits pulses at two discrete frequencies within K-Band. When an intruder enters the detection zone, the Doppler response at each frequency is digitally recorded. The difference between the two frequencies is controlled so that the phase angle between the two Doppler responses can be used to determine the unambiguous range to the target. As a byproduct of the process targets approaching the transceiver can be distinguished from those moving away from the transceiver. Range information is used to enhance the signal processing. The amount of signal integration is increased with range in proportion to the width of the detection pattern so as to optimize the signal to noise ratio (SNR).     

Novel radar target detection

Herein, a novel method of radar target detection based on 2-dimensional (2-D) fractal dimension is proposed. The proposed approach exploits both range information and azimuth information to estimate fractal dimension. Moreover, the approach can increase the number of data points. The above two merits result in the fractal dimension estimated by this method are more accurate and robust than the previous method. The detection performance is also better than the previous, which only makes use of 1-dimensional (1-D) information to estimate fractal dimension. Theoretical analysis and experimental results show that the proposed method performs well in a strong clutter background. The proposed method is also validated by real-life radar data, and the better result has been achieved.     

基于内嵌物理机理神经网络的热传导方程的正问题及逆问题求解

建立了一种基于内嵌物理机理神经网络(PINN)的热传导方程的正问题及逆问题求解方法。该方法利用自动微分技术将一维热传导方程嵌入到深度网络的损失函数中,通过以损失函数最小为目标来优化深度网络,求解一维热传导方程以及对方程中的未知导热系数进行辨识。随后,分析了基于PINN求解正问题的收敛精度以及参数辨识的鲁棒性,并得出以下结论:在给定网络结构的情况下,基于PINN求解一维热传导方程的收敛误差在样本点数较少时主要由采样误差主导,而当样本点数较多时,收敛误差由优化误差主导;由于损失函数中包含了方程相关的正则化项,以及采用了自动微分技术,因此,基于PINN的参数辨识方法噪声标签数据具有较强的鲁棒性。     

Universal equations for radar target detection

Two equations express detection probability and detectability factor for all Swerling targets and for partially correlated intermediate cases.     

Incoherent radar detection in compound-Gaussian clutter

The detection of incoherent pulse trains in compound-Gaussian disturbance with known spectral density is dealt with here. Two alternative approaches are investigated, The first, assuming perfect knowledge of the signal fluctuation law and implementing the Neyman-Pearson test on the observed waveform, turns out to be not applicable to the radar problem. The second, instead, relying on the generalized likelihood ratio optimization strategy, leads to a canonical detector, whose structure is independent of the clutter amplitude probability density function. Interestingly, this detector turns out to be constant false-alarm rate in the sense that threshold setting does not require any knowledge as to the clutter distribution, Moreover, since such a processor is not implementable in real situations, we also present an FFT-based (fast Fourier transform) suboptimum structure. Finally, we give closed-form formulas for the detection performance of both receivers, showing that both of them largely outperform the square-law detector, especially in the presence of very spiky clutter     

UWB radar for human being detection

UWB radar for detection and positioning of human beings in complex environment has been developed and manufactured. Novelty of the radar lies in its large operational bandwidth (11.7GHz at -10dB level) combined with high time stability. Detection of respiratory movement of a person in laboratory conditions has been demonstrated. Based on experimental results human being radar return has been analysed in the frequency band from 1 GHz to 12 GHz. Novel principle of human being detection is considered and verified experimentally.     

Multiband coherent radar detection against compound-Gaussianclutter

A Multiband GLRT-LQ (Generalized Likelihood Ratio Test-Linear Quadratic), MBGLRT-LQ, detector is derived for the coherent radar target detection against a compound-Gaussian clutter background. This scheme is an extension to the multiband case of the Asymptotically Optimum Detector (AOD), also derived under the name of GLRT-LQ in. The proposed multiband version of the algorithm shows two main advantages with respect to the original single-band algorithm. 1) For the adaptive implementation, it requires a much smaller area of homogeneous clutter echoes to estimate the covariance matrix of the interference; 2) it provides an optimum processing of the radar echoes when the radar operates in frequency agility, as electronic counter-countermeasure (ECCM) strategy. A closed form performance analysis is provided for the MBGLRT-LQ detector, which is used to compare it with the single-band version. An application to live recorded data is also presented to validate the obtained results     

Target detection with synthetic aperture radar

Target detection with synthetic aperture radar (SAR) is considered. We derive generalized likelihood ratio (GLR) detection algorithms that may be used with SAR images that are obtained with coherent subtraction or have Gaussian distributions. We analytically compare the performance of (1) a single pixel detector, (2) a detector using complete knowledge of the target signature information and known orientation information, (3) a detector using incomplete knowledge of the target signature information and known orientation information (4) a detector using unknown target signature information and known orientation information, and (5) a detector using unknown target signature information and unknown orientation information     

Low altitude target model for radar simulation

A simulation model of antenna array signals from a low altitude target is considered. Explicit expressions for statistical and spectral characteristics of the scattered signals are obtained in the Kirchhoff approximation. The model permits to evaluate the efficiency of different techniques of low altitude target tracking. It benefits the comprehension of the influence of multipath propagation on tracking radars     

CFAR detection and estimation for STAP radar

The algorithm presented here provides both a constant false-alarm rate (CFAR) detection and a maximum likelihood (ML) Doppler-bearing estimator of a target in a background of unknown Gaussian noise. A target is detected, and its parameters estimated within each range gate by evaluating a statistical test for each Doppler-angle cell and by selecting the cell with maximum output and finally comparing it with a threshold. Its CFAR performance is analyzed by the use of the sample matrix inversion (SMI) method and is evaluated in the cases of a fully adaptive space-time adaptive processing (STAP) and two partially adaptive STAPs. The performances of these criteria show that the probability of detection is a function only of the sample size K used to estimate the covariance matrix and a generalized signal-to-noise ratio. The choice of the number K is a tradeoff between performance and computational complexity. The performance curves demonstrate that the finer the resolution is, the poorer the detection capability. That means that one can trade off the accuracy of ML estimation with the performance of the CFAR detection criterion     

Spatial adaptive subspace detection in OTH radar

The work presented here addresses the problem of target detection against spatially structured interference composed of jamming plus noise, where for practical reasons, the received target wavefront may also deviate from the traditional plane wave model. This detection problem arises in over-the-horizon (OTH) radar systems where spatially distributed targets often compete for detection against directional interference that is spread over the entire range-Doppler search space. Conventional detection processing schemes are compared with a recently proposed adaptive subspace detector (ASD) that takes both the spatial structure of the interference and the possibility of target wavefront distortions into account. Experimental array data recorded by the Jindalee sky-wave and Iluka surface-wave OTH radar systems, located in central and northern Australia respectively, is used to evaluate detection performance.