Wavelet-based Identification of Specific Emitters

The Discrete Wavelet Transform (DWT) is a Time-Frequency Distribution (TFD) which offers the potential to efficiently characterize both transient and global intrapulse signal features in a highly compressed format. The overall objective of this three phase effort is to develop Wavelet analysis as a new approach to modulation on pulse (MOP) processing in support of electronic warfare (EW) systems identification of radar emitters. In the first phase, a study shall be conducted to assess the performance of the DWT to support emitter classification and identification. Its performance will be compared and contrasted with that of the current time and frequency domain approaches used in DoD programs. An extensive database of pulsed radar signatures will be used to support the analysis of the Wavelet features. We will draw upon a proven set of signal analysis tools developed under previous MOP study efforts to investigate performance for classification and identification. Applied research experience in the areas of multidimensional discriminant analysis, adaptive neural networks, and MOP coloration due to environmental (low SNR and multipath) and sensor distortion will directly support the performance evaluations.