Protocols for energy efficient and QoS aware communication in cluster-based wireless sensor networks.

摘要

Wireless sensor networks (WSNs) pose an enormous potential in providing diverse services to a broad range of applications including security, health care, environment, energy, military, and infrastructure monitoring by allowing intelligent monitoring of the physical environments. Such capabilities make WSNs immensely useful for data collection, analysis and possible actions. The pragmatic realization of WSN applications depends on many challenges such as energy efficient operation and support for application-specific QoS. Inspired by these challenges, this dissertation investigates techniques for energy conservation and QoS support in cluster-based WSNs. In this regard, four main contributions are made towards designing such techniques.;Firstly, a novel energy efficient cluster formation scheme is designed that uses a multi-criterion optimization technique. The novelty of this technique lies in employing multiple individual metrics for cluster formation that are critical for balanced energy dissipation of the system. By using multiple parameters that are of global and local significance, a fine control is achieved on the energy consumption behaviour of sensor nodes. In addition, the technique relies on simple design that allows sensor nodes to form clusters using local decisions only. Moreover, the decisions made by sensor nodes are based on information that is easily collected by only listening to the CH advertisement messages. Extensive simulations are performed to evaluate the performance of the proposed technique. Simulation results demonstrate that the proposed method achieves significant energy savings as compared to other popular protocols and significantly extends the network lifetime.;Secondly, at the MAC level an adaptive reservation-based protocol is designed that allows flexible assignment of bandwidth to the sensor nodes based on application demands. The reservation is achieved by maintaining a priority index for each application level. Nodes use the application priority index to request resources in the next scheduled access period. The cluster head uses the requests to generate the transmission schedule. The interactions from higher to lower layers are used to exploit application-specific characteristics of sensor networks while simultaneously maintaining energy efficiency.;Thirdly, a simple three-tier architecture is introduced that allows further energy savings by employing relay nodes. Relay nodes act as forwarding agents to transport the data collected by the cluster heads. Two novel algorithms are designed for selection of relay nodes thus providing a viable and energy efficient topology for cluster-based WSNs. Benefits in terms of energy gains and network lifetime are demonstrated by simulation results.;Finally, an analytical model of the reservation-based MAC protocol is developed using Network Calculus. Empirical relationships are derived to estimate the worst-case bounds for backlog and delay. The model is extended to include data with different priorities.