Energy optimization in sensor networks

Recent advances in wireless communications and computing technology are enabling the emergence of low-cost devices that incorporate sensing, processing, and communication functionalities. A large number of these devices are deployed to create a sensor network for both monitoring and control purposes. Sensor networks are currently an active research area mainly due to the potential of their applications. However, the operation of large scale sensor networks still requires solutions to numerous technical challenges that stem primarily from the constraints imposed by simple sensor devices. Among these challenges, the power constraint is the most critical one, since it involves not only reducing the energy consumption of a single sensor but also maximizing the lifetime of an entire network. The network lifetime can be maximized only by incorporating energy awareness into every stage of sensor network design and operation, thus empowering the system with the ability to make dynamic tradeoffs among energy consumption, system performance, and operational fidelity.;Optimizing the energy usage is a critical challenge for wireless sensor networks (WSNs). The requirements of energy optimization schemes are as follows. (1) Low individual energy consumption: Sensor nodes can use up their limited energy supply, carrying out computations and transmission. In typical WSNs, nodes play a dual role as both data sender and data router. Malfunctioning of some sensor nodes due to power failure can cause significant topological changes and may require rerouting of packets and network reorganization. Therefore, reducing the energy consumption of each sensor node is critical for WSNs. (2) Balanced energy usage: While minimizing the energy consumption of individual sensor nodes is important, the energy status of the entire network should also be of the same order. If certain nodes have much higher workload than others, these nodes will drain off their energy rapidly and adversely impact the overall system lifetime. The workload of sensors should be balanced in order to achieve longer system lifetime. (3) Low computation and communication overhead: The resource limitations imposed by sensor hardware call for simple protocols that require minimal processing and a small memory footprint. The extra computation and communication introduced by the energy optimization schemes must also be kept low. Otherwise, energy required to perform the optimization schemes may outweigh the benefits.;This thesis concentrates on the energy optimization issues in wireless sensor networks. We study the power consumption characteristics of typical sensor platforms, and propose energy optimization schemes in network and application level. We design distributed algorithms that reduce the amount of data traffic and unnecessary overhearing waste in WSNs, and further propose load balancing mechanisms that alleviate the unbalanced energy usage and prolong the effective system lifetime.;At the network level, Adaptive Aggregation Tree ( AAT) is proposed to dynamically transform the routing tree, using easily-obtained overheard information, to improve the aggregation efficiency. The local adaptivity of AAT achieves significant energy reduction, compared to the shortest-path tree where aggregation occurs opportunistically. We also propose Neighborhood-Aware Density Control (NADC ), which exploits the overheard information to reduce the unnecessary overhearing waste along routing paths. In NADC, nodes observe their neighborhood and adapt their participation in the multihop routing topology. By reducing the node density near the routing paths, the overhearing waste can be reduced, and the extremely unbalanced energy usage among sensor nodes is also alleviated, which results in a longer system lifetime. The unbalanced energy usage problem is further addressed at the application level, where we propose Zone-Repartitioning (Z-R) for load balancing in data-centric storage systems. Z-R reduces the workload of certain hot-spots by distributing their communication load to other nodes when the event frequency of certain areas is much higher than the others.