Throughput capacity analysis of large multi-channel, multi-interface random wireless mesh networks

Abstract

This paper studies the asymptotic throughput capacity of a random infrastructure wireless mesh network (RndInfWMN). Generally an infrastructure wireless mesh network (InfWMN) comprises mesh clients, routers, gateways and has hierarchical structures. An InfWMN can be divided into two categories, which are arbitrary InfWMN (ArbInfWMN) and random InfWMN (RndInfWMN). In an ArbInfWMN, the locations of the WMRs are arbitrary while in an RndInfWMN, WMRs are distributed randomly. The latter is more interesting when randomly distributed WLANs are desired to be connected through a wired network of gateways. There are some analytical research on the asymptotic capacity of ArbInfWMN where the number of interfaces per-infrastructure node, m, is at the same order of the number of available channels for the network, c, i.e. cm=θ(1). In our previous research, we investigated the asymptotic throughput capacity of ArbInfWMNs for a more general case in which cm=O(1). However, to date, analytical facility has been limited by the absence of analysis in RndInfWMNs, especially for the general case in which cm=O(1). In this paper, we carry out an original analysis of the asymptotic per-client throughput capacity of multi-channel multi-interface RndInfWMNs for the case in which cm=O(1). Our analysis shows that by identifying cm in different scaling regimes, the asymptotic per-client throughput capacity of multi-channel multi-interface RndInfWMNs exhibits different bounds, depending on the ratio between c and m.