Low-Complexity Signal Processing Algorithms for Wireles Sensor Networks

Recently, wireless sensor networks (WSNs) have attracted a great deal of research interest because of their unique features that allow a wide range of applications in the areas of military, environment, health and home. One of the most important constraints on WSNs is the low power consumption requirement as sensor nodes carry limited, generally irreplaceable, power sources. Therefore, low complexity and high energy efficiency are the most important design characteristics for WSNs. In this thesis, we focus on the development of low complexity signal processing algorithms for the physical layer and cross layer designs for WSNs. For the physical layer design, low-complexity set-membership (SM) channel estimation algorithms for WSNs are investigated. Two matrix-based SM algorithms are developed for the estimation of the complex matrix channel parameters. The main goal is to reduce the computational complexity significantly as compared with existing channel estimators and extend the lifetime of the WSN by reducing its power consumption. For the cross layer design, strategies to jointly design linear receivers and the power allocation parameters for WSNs via an alternating optimization approach are proposed. We firstly consider a two-hop wireless sensor network with multiple relay nodes. Two design criteria are considered: the first one minimizes the mean-square error (MMSE) and the second one maximizes the sum-rate (MSR) of the wireless sensor network. Then, in order to increase the applicability of our investigation, we develop joint strategies for general multihop WSNs. They can be considered as an extension of the strategies proposed for the two-hop WSNs and more complex mathematical derivations are presented. The major advantage is that they are applicable to general multihop WSNs which can provide larger coverage than the two-hop WSNs.