THE TRUSTEES OF COLUMBIA UNIVERSITY IN THE CITY OF NEW YORK INC
The objective of this research is to investigate novel architectures for the realization of ultra-low-power, integrated, pulse-based, ultra-wideband radar and imaging sensors. The approach is to exploit compressive sensing for high-speed baseband processing to simultaneously achieve high dynamic range while consuming ultra-low power. Ultra-wideband, pulse-based radar and imaging sensors require high-speed baseband processing. Compressive-sensing-based architectures take advantage of the presence of inherent structure in the received radar signal, specifically its time-domain sparsity, to significantly reduce the required sampling rate (and hence, power consumption) while preserving the information contained in the signal. The intellectual merit of the proposed research is to devise and investigate the feasibility of compressive-sensing-based ultra-wideband radar and imaging architectures through theoretical and simulation-based studies of the robustness of compressive sensing to practical issues such as noise, interference, presence of insufficient sparsity, and circuit imperfections such as clock jitter. These studies will be linked to measurements performed with prototype radars constructed with off-the-shelf components. The broader impact of the proposed research is in its ability to transform the current vision of ambient intelligence. The low-power and low-cost nature of silicon-based ultra-wideband radar and imaging sensors utilizing compressive sensing potentially enables ubiquitous deployment of massive sensor networks, which can have a dramatic impact on the quality of day-to-day life. The proposed research also bridges the disciplines of integrated analog- and digital-circuit design, signal processing and applied mathematics. This proposal will support the education and training of a graduate student researcher, who will develop skills in all these fields.