Determining the end-to-end throughput capacity in multi-hop networks

Abstract

In this paper, we present a methodology to analytically compute the throughput capacity, or the maximum end-to-end throughput of a given source and destination pair in a multi-hop wireless network. The end-to-end throughput capacity is computed by considering the interference due to neighboring nodes, as well as various modes of hidden node interference. Knowing the throughput capacity is important because it facilitates the design of routing policy, admission control for realtime traffic, as well as load control for wireless networks. We model location-dependent neighboring interference and we use a contention graph to represent these interference relationships. Based on the contention graph, we formulate the individual link capacity as a set of fixed point equations. The end-to-end throughput capacity can then be determined once these link capacities are obtained. To illustrate the utility of our proposed methodology, we present two important applications: (a) route optimization to determine the path with the maximum end-to-end throughput capacity and, (b) optimal offered load control for a given path so that the maximum end-to-end capacity can be achieved. Extensive simulations are carried out to verify and validate the proposed analytical methodology.