Signal processing algorithms for simplified-hardware multistatic synthetic aperture radar

Abstract

This work represents contributions to the state of the art of radar tomography and SAR in three main areas. First, we develop a SAR measurement and image reconstruction framework for a downlooking, omnidirectional and multistatic SAR system. This framework, a significant generalization of previous work for downlooking monostatic SARs, models the SAR measurement process as a generalized integral transform of a reflectivity function. Secondly, based on this measurement and inversion framework, we resolve the novel question of SAR vehicle trajectory design for the purpose of optimal image quality in the presence of measurement noise. In addition, we introduce the concept of a random-location multistatic trajectory, and show that (for large enough problems) such a strategy is likely to approximate an upper performance bound; therefore a random strategy is likely to perform as well or better than any deterministic trajectory strategy. Finally, we expose connections between the Welti phase-coded radar sequences and the Discrete Wavelet transform. This connection directly yields very simple and natively high-speed implementations of pulse compression filters at the radar receiver. Furthermore, we generalize the idea of zero correlation zone sequences by developing methods to generate transmit waveforms which extend the perfect correlation width of such filter outputs without modifying the simplicity of the receive filters. This method provides a new system tradeoff between transmit energy and receiver hardware complexity and power.