ALARM: An Adaptive Load-Aware Routing Metric for Hybrid Wireless Mesh Networks

Abstract — Hybrid Wireless Mesh Networks (WMN) are gen-erally employed to establish communication during disaster recovery operations. The Hybrid WMN network is formed in a spontaneous manner when wireless nodes belonging to different agencies are operated in the disaster area. These wireless nodes generally have heterogeneous configurations in terms of number of transceivers, computational power, battery resources and mobility pattern. Internet Protocol (IP) acts as the common platform for integrating these heterogeneous devices. Routing protocols are engaged to determine paths between sets of wireless nodes. A number of routing metrics have been developed to achieve performance gains in multi-radio, multi-hop WMNs. However, most of these metrics need access to external information like link quality statistics, channel numbers, etc on a regular basis. This frequent information exchange coupled with the incessant variation in mobility and traffic load conditions causes degraded performance of these metrics in Hybrid WMNs. In this paper, we present a routing metric named ALARM specifically designed for Hybrid WMNs. ALARM is computed using the number of packets queued per wireless interface. This computed value offers an accurate representation of the traffic load, link quality, interference and noise levels. With the help of extensive simulations, we show that our routing metric outperforms well-know routing metrics like ETT and WCETT under varying mobility and traffic load conditions in Hybrid WMNs. We further show the practicality of the metric through a prototype implementation and provide performance results obtained from a small-scale testbed deployment.1