Spectrum sensing and vector signal analysis preprocessing based on compressed sampling

Abstract

This paper investigates the application of a compressed sampling (CS) algorithm as a spectrum sensing and signal analysis preprocessor for vector measurements of digital modulations. Compressed sampling is a paradigm which exploits sparsity, a feature common to several signals of interest, to allow the design of efficient data acquisition schemes. These need to be followed by more complex signal processing algorithms for accurate signal reconstruction. The discussion focuses on the application of a CS algorithm to spectrum sensing and modulation analysis in wireless communications. When the signals of interest occupy only a few among several possible bands, and do so only for short time bursts, feeding a vector signal analyser with the required preliminary information becomes increasingly important, but also more challenging. Results presented in the paper show that a CS algorithm can successfully extract such information from a record of signal samples, providing spectrum-blind sensing capabilities.