Load balancing routing in multi-hop wireless networks

Multi-hop wireless networks have attracted significant interests in recent years because of the improved flexibility, mobility and low-cost compared to wired networks. Routing in multi-hop wireless networks has been heavily studied in the last decade and numerous routing protocols were proposed in literature. Most current wireless routing protocols are based on Shortest Path Routing (SPR) where packets are delivered along the shortest route from a source to a destination. Even for geographical localized routing protocols, such as the greedy routing, the packets usually follow the shortest paths between sources and destinations when the network is dense and uniformly distributed. Taking the shortest path can achieve smaller delay or traveled distance. However, under uniform communication shortest path routing suffers from uneven load distribution in the network, such as crowded center effect where the center nodes have much heavier load than the nodes in the periphery. We propose a novel routing method, called Circular Sailing Routing (CSR) seeking to distribute the traffic more evenly in the network. The proposed method first maps the network onto a sphere via a simple stereographic projection. Then the route decision is made by the spherical distance on the sphere instead of the Euclidean distance in the plane. By spreading the traffic across a virtual sphere, CSR can diminish the crowded center effect and increase the energy lifetime of the network. CSR can be easily implemented by using current position-based routing protocol without any major changes or additional overhead. The only modification is a simple mapping calculation of the position information. We theoretically prove that for a network the distance traveled by the packets using CSR is no more than a small constant factor of the minimum (the distance of the shortest path). We call this factor as Competitiveness Factor. We then extend CSR to a localized version, Localized CSR, by modifying the greedy routing without any additional communication overhead. Although most existing wireless systems and protocols are based on two-dimensional (2D) design, in reality, a variety of networks operate in three-dimensional space. In this dissertation, we also investigate how to design CSR load balancing routing for 3D networks. We describe two mapping methods for 3D CSR, i.e., wireless nodes in a 3D network are projected on a 3D or 4D sphere. We provide theoretical proofs of their competitiveness compared to SPR. For all proposed methods, we conduct extensive simulations to study their performance and compare them with global shortest path routing or greedy routing in 2D and 3D wireless networks. The simulation results demonstrate our routing protocols have much smaller maximum traffic load and the standard deviation of traffic load and the simulation results also confirm our theoretical bounds about the competitiveness factor.