Compressive sensing using USRP

Abstract

Wireless communication is rapidly growing form of communication worldwide, ranging from voice communication in cell phone, data transfer including high-speed video transfer, sensor networks etc. However, with the increasing usage, capacity limitations have also increased. The project aims at bridging an existing signal processing technique, via compressive sensing to estimate the channel impulse response and to show the improved results. This project has been implemented on GNU (software platform) and USRP (hardware platform). In signal processing, the common goal is to reconstruct a signal from a series of measurements. In digital imaging, compression becomes a necessary step due to increased samples leading to high Nyquist rate, inorder to store or transmit a signal. Hence, compressive sensing provides an effective way for reconstruction from limited measurements. In wireless system, signal undergoes reflection, diffraction or scattering when transmitted and at receiver end, we get an attenuated and delayed version of the original signal.