Radar Signal Processing

I am currently involved in two research projects in the area of radar systems:

■   Space-Time Adaptive Processing for OTHR

Over the horizon radar has the potential to detect and identify targets of interest at extremely large ranges. In cooperation with Defense Research and Development Canada, I am working on developing practical space-time adaptive processing techniques for OTHR. This includes receive processing using, for example, our recent contribution in fast fully adaptive processing. Given that the system being developed by DRDC will include waveform diversity, we plan to include both transmit and receive adaptive processing.

■   Signal Processing for Concealed Weapon Detection

With several researchers in various Canadian and US universities, I am working on signal processing techniques for the detection of concealed weapons using a noise radar. Noise radars are inherently broadband and can provide the discrimination required to detect and identify weapons concealed under, for example, clothing. This effort is also supported by DRDC.

Projects Completed at UofT

■   Waveform Space-Time Adaptive Processing

In airborne radar systems, space-time adaptive processing (STAP) has been proved to be essential to distinguish small, slowly moving strategic targets from background interference (both unintentional ground clutter and intentional jamming). Recently we have shown that to extend STAP to distributed apertures, arrays of apertures widely distributed in space, one needs to augment the spatial and temporal dimensions with a waveform dimension. We:

■   developed an interference model and investigating adaptive processing for frequency diverse adaptive processing

■   developing an understanding of the interaction between wave shapes and interference statistics with the goal of adapting waveform design to minimize interference

■   develop adaptive processing algorithms specifically for the waveform-space-time case

■   Robust processing for the knowledge based STAP (KB-STAP) concept

This project made robust the hybrid algorithm (developed in 1999 by Adve, Wicks and Hale) for adaptive processing in non-homogeneous interference scenarios. The hybrid approach, combining statistical and non-statistical adaptive processing, has proved extremely effective in countering acting effects of non-homogeneities on space-time adaptive processing (STAP). We are now continuing the development of the theoretical foundations of this algorithm, especially the role of an emphasis parameter that determines the performance of the system.

 (Sponsored by AFRL/SNRT )

■ Application development for WIPL-DP

In this three year project (2002-2004) we developed and implemented applications for an extremely powerful electromagnetic analysis tool. Under the auspices of the Common High Performance Computing Software Support Initiative (CHSSI), the US DoD has sponsored the parallelization of WIPL-D . A parallel version of WIPL-D (designated WIPL-DP) has the potential to enable many applications of interest, such as detection of buried targets, foliage penetration and  synthetic aperture radar. This project involved investigating and developing models of changing radar scenes, moving platforms and targets to be incorporated into a frequency domain analysis technique. An important aspect of this effort was the modeling of the non-linear interaction between the target and its environment.

(Sponsored by Black River Systems Co. Inc. , the primary contractor on the CHSSI effort)