Amit Kumar Mishra

Radar Signal Classification Using Pca-Based Features

Principal component analysis (PCA) has been used in many applications ranging from social science to space science, for the purpose of data compression and feature extraction. Usage of PCA for synthetic aperture radar (SAR) image classification, though widely reported by remote-sensing researchers, has not been exploited much by automatic target recognition (ATR) community. In the present paper, PCA has been used in SAR-ATR using the MSTAR data base, and comparison has been made with the conventional conditional Gaussian model based Bayesian classifier (M.D. DeVore and J.A. OSullivan, 2002). The results have been compared based on percentage of correct classification, receiver operating characteristics (ROC), and performance with limited amount of training data. By all standards of comparison, the PCA based classifier was observed to outperform the conditional Gaussian model based Bayesian classifier (CGBC) or at the worst it performs at par. And given the computational and algorithmic simplicity of PCA based classifier, the new algorithm was concluded to be a highly prospective candidate for real time ATR systems

Bistatic SAR ATR using PCA-based features

Target recognition is desirable feature of any defence radar system. With the present revival of interest in bistatic and multi-static radar systems, the future radar systems are predicted to invariably have bistatic abilities. The present project aims at looking into prospects and limitations of bistatic target recognition and to develop efficient algorithm for the same. The work in this paper reports the development of a database of bistatic target signatures, and the application of principal component analysis (PCA) based classifiers on the same. Results are compared with the more basic conditional Gaussian model based Bayesian classifier.

Database generation of bistatic ground target signatures

Automatic target classification has been an area of active study due to many of its significances. With there advent of bistatic technology in radar, the research into target classification in bistatic scenario seems quite timely. But the study of any classification algorithm demands for a suitable database. Due to the unavailability of suitable field collected bistatic database, computer simulation of such a database is quite pertinent. The present work discusses the approach towards the generation of a similar database, using FEKO as the electromagnetic simulation tool. Various special needs of such a project and the answers of FEKO to those needs, have been discussed. The results conclusively show the suitability of FEKO for the project.

Generation of SAR Image for Real-life Objects using General Purpose EM Simulators

Abstract In the applications related to airborne radars, simulation has always played an important role. This is mainly because of the two fold reason of the unavailability of desired data and the difficulty associated with the collection of data under controlled environment. A simple example will be regarding the collection of pure multipolar radar data. Even after phenomenal development in the field of radar hardware design and signal processing, till now the collection of pure multipolar data is a challenge for the radar system designers. Till very recently, the power of computer simulation of radar signal return was available to a very selected few. This was because of the heavy cost associated with some of the main line electro magnetic (EM) simulators for radar signal simulation, and secondly because many such EM simulators are for restricted marketting. However, because of the fast progress made in the field of EM simulation, many of the current generic EM simulators can be used to simulate radar returns from realistic targets. The current article expounds the steps towards generating a synthetic aperture radar (SAR) image database of ground targets, using a generic EM simulator. It also demonstrates by the help of some example images, the quality of the SAR image generated using a general purpose EM simulator.

Monostatic Radar Signatures of Significant Classes of Ground Targets, in the Time and Frequency-Domain

The problem of characterizing the radar returns from complex man-made localized targets using airborne radar platforms has been and currently is a topic of tremendous interest. This paper presents results of modeling four generic types of ground targets, and analyzes the monostatic radar cross section (RCS) both in the frequency and the time domain. The recorded results reinforce the conclusion that in realistic situations return signals are complex enough to include higher order terms, other than the shifted in time and frequency replica of the transmitted signal. A mechanism is shown on how to estimate these higher order terms. The existence of these terms allows one to see how signals are structured in the time domain. This in turn may successfully lead in the development of suitable matched filters for the class of targets studied in this paper