Fine-time resolution passive RF source imaging

With the electromagnetic spectrum more congested than ever, the task of passively monitoring the spectrum to detect, separate, identify and locate emitters is fraught with diffculty. Enabled by modern solid state RF technologies, Radar and communications systems are more agile, providing a foundation for low probability of intercept (LPI) emitters to exist for both, Radar and communications systems. One of the key characteristics is a broadband, spread-spectrum modulation scheme. Previous efforts in this area have focussed on improving signal to noise and signal to interference plus noise ratios (SNR and SINR) of LPI emitters using various time and frequency domain methods. This neglects an important aspect of electronic surveillance - the spatial domain using antenna arrays. Current literature on antenna array signal processing exploits the simplifications inherent in the frequency domain at the expense of limiting these techniques to narrowband signals and making them inadequate for broadband sources. This thesis addresses spatio-temporal antenna array signal processing for LPI emitters in order to improve spatial resolution for signal separation. One of the key tools throughout this thesis is the use of polynomial matrices and the polynomial eigenvalue decomposition that allow spatial super resolution techniques, such as the MUSIC algorithm, to be extended to the broadband domain. Results are demonstrated via simulations and statistical analysis throughout this thesis, concluding that while computationally expensive, the novel methods contained within this thesis provide an attractive solution to the processing of LPI emitters.