Source and Channel Coding in Wireless Sensor Networks

Student thesis: Doctoral Thesis

Abstract

Wireless Sensor Networks (WSNs) consist of small nodes equipped with sensors (temperature, light, humidity, etc.), a small scale processor and a limited range radio frequency communication unit. Usually these nodes are powered by batteries and/or by scavenging energy from the environment. WSNs have a wide range of applications including industrial, medical, environmental and weather monitoring, among others. The nodes in the WSN sense their environment, encapsulate the measurements into packets and forward them to a collecting Base Station. Due to the limitation of the transmission power and stored energy of the nodes, they forward the packets in a multi-hop manner. For some applications, deploying Distributed Source Coding (DSC) can significantly improve the lifetime of the network by exploiting correlations between the measurements from adjacent nodes. To maximize the network lifetime using DSC, routing paths and compression rates have to be optimized together. The first part of this thesis formulates the optimization problem and presents a developed scheme for optimizing routing paths and compression rates, which seeks a balance between per node energy consumption and total network energy consumption for maximizing the network lifetime.

Communications over a noisy channel are prone to noise and interference that usually cause errors in the received packets. Plain retransmission of erroneous packets requires high amounts of energy from the nodes and introduces high latency in the network. The second part of the thesis presents Error Correcting Code (ECC) approaches for WSNs with low processing consumption at the sensor nodes and high immunity to channel noise. The ECC approach that is put forward in this dissertation requires a low complexity encoding and decoding process at the sensor nodes with iterative decoding processes running at the Base Station, where enough resources are available.
Date of Award22 Jun 2012
Original languageEnglish
SupervisorKris Steenhaut (Promotor), Ann Nowe (Co-promotor), Leo Van Biesen (Jury), Jacques Tiberghien (Jury), Rik Pintelon (Jury), Peter Schelkens (Jury), Marnix Goossens (Jury), Ingrid Moerman (Jury), Jean-Michel Dricot (Jury) & Dario Di Zenobio (Jury)

Keywords

  • WSN

Cite this

'