Low-Overhead Opportunistic Routing for Wireless Ad Hoc and Sensor Networks in a Fading Environment

Abstract

The development of miniaturized radio and sensing technologies have enabled the deployment of large quantities of wireless sensors capable of forming multi-hop networks. Emerging applications of this technology such as surveillance and disaster monitoring have throughput and efficiency requirements not met by current routing algorithms. These requirements are also shared by ad-hoc networks.

Early routing protocols for these wireless networks were based on algorithms designed for wired networks. Geographic routing (routing based on position), was proposed. These algorithms perform poorly since they do not account for the fading and interference effects of wireless channels. Recent protocols that have attempted to account for the wireless channel focus on single-hop situations and are not readily extensible to multi-hop networks.

In this thesis we present a framework for routing based on a distributed routing decision and provide several example protocols. This framework provides a cross-layer design where the routing decision is decided through silent negotiation between candidate relays. We investigate the performance and parameters of this framework. We then present an example protocol using this framework which provides low-overhead opportunistic routing using cooperative diversity. This protocol uses the intrinsic characteristics of the wireless channel to achieve diversity while still maintaining relatively low overhead. An adaptation of the protocol for heterogeneous networks equipped with multiple antennas has also been discussed and evaluated through simulations. We also investigate another protocol based on this framework using the product of the instantaneous packet reception rate and the marginal progress towards the destination as a routing metric, offering enhanced throughput.