| Their features of rapid establishment and self-organization have rendered wireless ad hoc networks to be identified as an indispensable component to support ubiquitous communications. However, the viable deployment of such networks faces many challenges resulting from some innate unfavorable features such as the constrained resources, for example, finite energy supply and limited bandwidth. In this dissertation we investigate the cross-layer design of resource-aware protocols that conserve energy and support quality-of-service (QoS) in resource-constrained wireless ad hoc networks.;We first study the issue of energy conservation in heterogenous wireless ad hoc networks, where most nodes are powered by batteries with small capacity, while some others are more powerful, for example, powered by batteries with large capacity. We propose a cross-layer designed device-energy-load aware relaying framework, called DELAR, to capitalize such powerful nodes to conserve energy. Different from the previous work that addresses the energy conservation issue at one layer, DELAR is a joint design of scheduling, power control and routing that can conserve energy thus prolong the network lifetime. We further propose a resource-aware movement mechanism to utilize such powerful nodes for data relaying. Such a resource-aware movement mechanism provides a new way to conserve energy in wireless ad hoc networks.;We then study the issue of how to support QoS in wireless ad hoc networks. We propose a scheme called Courtesy Piggybacking to utilize system dynamics including time-varying channel condition and stochastic traffic characteristics, to support differentiated services in such networks. The basic idea of Courtesy Piggybacking is to let high priority traffic help the low priority traffic by sharing the unused residual bandwidth. As a result, the scheme improves the overall performance for different prioritized services.;Finally, we propose an energy efficient, reliable and scalable data dissemination scheme for wireless sensor networks. Based on supply chain management methodology, for each sensing task, the sensor field is conceptually partitioned into several functional areas, and then different routing schemes are applied in different areas. This scheme addresses energy-efficiency, reliability, and scalability at the same time. |
