Author: Maryam Vala

Publisher:

Published: 2014

Total Pages:

ISBN-13:

"Ultra Wideband (UWB) is a rapidly growing technology for digital wireless communications. UWB utilizes low power, ultra short, pulses and is specifically suited for short-range, high-rate indoor wireless communications as well as fine localization applications. The attractive properties of UWB are a direct consequence of its very wide bandwidth that also implies an extremely high Nyquist sampling rate, so digital processing of UWB signals requires the use of fast and efficient, and therefore, expensive, Analog to Digital Converters. UWB systems are commonly used in indoor environments in which propagation is characterized by long but sparse multipath channels. Multi-path channels cause inter symbol interference (ISI) and introduce distortion to the received signal. To counter the effects of the channel, the channel impulse response must be accurately estimated at the receiver. There are two main estimation approaches for UWB channels: Compressed Sensing (CS)-based and Adaptive Filter (AF)-based. The former combine sampling and compression into a single linear measurement process that operates at sub-Nyquist rate by capitalizingon the sparsity of UWB signals. The latter, uses Least-Mean-Square filtering principles to obtain low complexity channel estimation methods. The problem we consider in this thesisis the estimation of a long sparse multi-path channel in an UWB system. We focus on systems that use Pulse Position Modulation (PPM) that is one of the most commonly used modulation schemes in UWB systems. We review CS and AF based methods and then we propose new AF-type methods specifically for the estimation of sparse multi-path channels with PPM inputs. The main idea behind the proposed methods is to estimate the long channel in sections in order to reduce the computational cost and improve the estimation performance. Finally, we present simulation results showing the superior performance of the proposed algorithms." --