Evaluating the effect of quality of service mechanisms in power-constrained wireless mesh networks

Abstract

Wireless Mesh Network (WMN) is a collection of wireless nodes which can dynamically communicate with one another in multi-hop manner. This network has received considerable attention as a means to connectivity in community and commercial entities. The easy deployment and self-management characteristics of WMN, makes it a good choice for rural areas. However, in most developing countries, electricity is scarce or unreliable in the rural areas. A candidate solution to the lack of electricity supply in these areas is the use of solar/battery-powered nodes. Significant efforts have gone into optimization of Quality of Service (QoS) provisioning in WMN; hence, a lot of QoS mechanisms on OSI layer have been proposed. It is, however, not clear how different QoS mechanisms on OSI layer affect the node lifetime and the energy cost per bit of a battery-powered WMN nodes. Different protocols at different layers have varying effects on the energy efficiency of the battery-powered WMN nodes, when those protocols are subjected to various transmission power levels and payload sizes. The goal of this study is to evaluate how different existing QoS mechanisms affect the operational lifetime of battery-powered WMN nodes. This goal was achieved by evaluating how connection (TCP) and connectionless (UDP) Transport Layer protocols together with Reactive (AODV) and Proactive (OLSR) Routing protocols influence the lifetime of battery-powered nodes when subjected to different transmission power levels and payload sizes. The evaluation was carried out using NS-2 simulation and a fourteen nodes indoor testbed. The overall results of both the simulation and the testbed experiments show that for a TCP-based scenarios, TCP with OLSR at maximum transmission power level and maximum payload size outperform others in terms of packet delivery ratio, average throughput and average energy cost per bit, while TCP with AODV at minimum transmission power level and maximum payload size outperform others in terms of node lifetime. And for the UDP-based scenarios, UDP with AODV at maximum transmission power level and maximum payload size outperform others in terms of packet delivery ratio, average throughput, node lifetime and average energy cost per bit, while TCP with AODV at minimum transmission power level and maximum payload size outperform others in terms of node lifetime. The results of this study also reveal that simulation results only give a rough estimate of the real world network performance. Hence, whenever it is feasible, validating a simulation result using testbed is highly recommended in order to have clear and better understanding of the protocol performances.