Energy-efficient broadcasting in wireless ad-hoc sensor networks
Date of Issue2014
School of Computer Engineering
Emerging Research Lab
In dense wireless sensor networks (WSNs), broadcasting can cause high energy consumption and unnecessary rebroadcast. To overcome these problems, a variable preamble length-based broadcasting scheme is proposed. The proposed scheme uses a small preamble of variable size before the data packet is broadcast to mitigate the problem of excessive energy consumption due to large preambles without compromising the reachability. The preamble size varies depending on the network density. An existing probability-based broadcasting scheme is implemented and compared with the proposed schemes. A detailed analytical study is also done for the proposed variable preamble length-based broadcasting scheme and an existing probability-based broadcasting scheme for WSNs. The simulation results show that the proposed scheme conserves up to 19% of energy without compromising the reachability and throughput. The schemes are also implemented on real nodes, and the results compared with simulation results. To further improve the performance and conserve energy of sensor nodes, we have proposed a coverage-based adaptive preamble broadcasting protocol, which exploits the advantages of existing coverage-based as well as counter-based schemes. It dynamically adjusts the preamble length of a rebroadcast packet according to the node density in the host’s neighbourhood without any prior knowledge about the neighbouring hosts. It uses the extra coverage to infer the node density and the duplicate packet counter, regardless of the number of neighbours in the network. The proposed scheme can be integrated with any asynchronous preamble sampling MAC layer protocol. A Dynamic Probabilistic Rebroadcast with Coverage Area and Neighbour Confirmation (DPCN) and probability-based broadcasting schemes are implemented and compared with the proposed coverage-based adaptive preamble broadcasting scheme. The proposed scheme conserves up to 16% of energy against the existing broadcasting schemes. A detailed study on the effect of distance estimation is also performed and presented in this thesis. The thesis also presents a comprehensive study of the effect of adaptive preambles in dense and sparse area networks. It has been successfully demonstrated that the proposed broadcasting schemes achieve higher energy-efficiency than other broadcasting schemes. This thesis also shows that the increase in energy-efficiency is realized without sacrificing the saved rebroadcast and reachability.
DRNTU::Engineering::Computer science and engineering::Computer systems organization::Computer-communication networks