CIF:Medium:Collaborative Research:Low Resolution Sampling with Generalized Thresholds

University Of California-Irvine, Irvine CA

Investigators

Abstract

CIF: Medium: Collaborative Research: Low-Resolution Sampling with Generalized Thresholds Jian Li, Lee Swindlehurst, and Mojtaba Soltanalian Abstract Quantization of signals of interest is a necessary first step in digital signal processing applications. When signals across a wide frequency band are of interest, a fundamental tradeoff between sampling rate, amplitude quantization precision, cost, and power consumption is encountered. The investigators study low resolution sampling techniques with general thresholds, which are affordable, technically feasible, easy to apply, energy-efficient, and consistent with technological trends. The enormous gains in capacity and spectral efficiency, for example, that could be provided by a successful millimeter wave (mm-wave) massive multiple-input multiple output implementation could have a revolutionary effect on the performance of wireless systems nearly everywhere we use them: at home, at work, at school, commuting via public transportation or by plane, shopping, at restaurants, recreational venues, sporting events, and so on. Besides consumer applications, there are many military- and security-related scenarios where our systems could be used. This project involves advancing fundamental knowledge in developing dynamic energy-efficient and cost-effective sampling techniques and applies engineering principles to address the critical needs of several important and related applications. Specifically, this project involves addressing significant open questions, including deterministic identifiability, performance bounds, and impact of thresholding pattern on spectrum sensing and array processing, radio frequency interference mitigation, and mm-wave communications to gain fundamental insights into the novel paradigm of low resolution sampling with general thresholds, devising novel signal processing algorithms, including effective and efficient sparse signal recovery techniques and parametric maximum likelihood methods for enhanced performance, and evaluating and demonstrating the performance using measured data. This project also involves preparing students for engineering in the 21st century through the incorporation of practical design and problem-solving techniques into both the education curriculum.