Energy-efficient strategies for deployment and resource allocation in wireless sensor networks

A Wireless Sensor Network (WSN) comprises of a collection of sensor nodes that perform sensing and communication tasks and possibly a set of additional nodes (called micro-servers) which receive and aggregate local sensor data and send it to a central base station. Two figures, of merit for assessing the quality and efficacy of a deployed WSN are Quality of Monitoring (QoM) and Monitoring Lifetime (MoL). The QoM is strongly influenced by the density of sensors inside the monitoring region whereas the MoL is directly related to the initial energy level assigned to the nodes. Chief among design problems of WSNs are optimal sensor node deployment and energy resource allocation.; In chapter 2, considering a flat network architecture comprised of randomly-deployed sensor nodes, we determine the energy and node densities that result in allocating the minimum total resources while meeting constraints on the QoM and MoL figures of merit. After showing that the optimal solution is a strong function of the routing scheme used to send a packet of data from sensors to the base station, we propose and analyze a novel data routing scheme for such a network.; In chapter 3, we investigate the same problem for a two-tier WSN comprising of randomly-deployed sensors and micro-servers at its lower and higher levels. The sensor density is preset in order to achieve a minimum level of QoM. The micro-server and energy resource densities are, however, to be determined.; In chapter 4, we investigate lifetime-aware strategies for detailed placement and energy allocation of infrastructure in the second level of a two-tier WSN. Various forms of the following problem are formulated: Assign positions and initial energy levels to the micro-servers, and concurrently, assign groups of sensors to each micro-server so as maximize MoL of the two-tier WSN subject to a total energy budget. Simulation results show that the optimal design and deployment of a two-tier network can increase the WSN lifetime by one order of magnitude compared to a flat WSN with the same total initial energy and quality of monitoring.