Exploiting the Wireless Broadcast Advantage in Coded Wireless Mesh Networks

点击翻译按钮获取中文摘要

This research project aims at exploring the potential of the wireless broadcast advantage in wireless mesh networks. Recent theoretical and experimental research has shown that the wireless broadcast advantage offers significant gains in wireless mesh networks. A key technique to exploit the wireless broadcast with manageable overhead is network coding, which focuses on forwarding the information contained in packets to the destinations instead of the unaltered packets themselves. We propose to study the wireless broadcast advantage with network coding in mesh networks and experimentally demonstrate the applicability of network coding in real world environments. To this end we extend the recently developed polymatroid broadcast function model for the wireless broadcast advantage to state of the art WiFi networks and study the interaction of medium access mechanisms and physical layer techniques with this model. Based on this model, we design a wireless mesh protocol that utilizes network coding to efficiently exploit the wireless broadcast advantage to improve various performance metrics. Although the state of the art network coding protocols for wireless networks are COPE and MORE, both developed at the MIT and both showing fundamental gains due to the broadcast nature of the wireless medium, they both leave room for major contributions with respect to the optimization of the medium access scheme, the bidirectional and multicast traffic, the interoperability with higher layer protocols, and with respect to adaptive physical layer models. In addition our proposed protocol is the core of a wireless mesh demonstrator and experiment setup with up to 25 nodes based on IEEE 802.11 consumer devices. The experiment is designed to demonstrate the benefit of network coding in wireless mesh networks with different traffic scenarios and to analyze the performance of network coding to exploit the wireless broadcast advantage in a real world scenario in terms of throughput, reliability, delay, protocol and coordination overhead, interaction with common network and transport layer protocols, and further metrics. We also aim at comparing the proposed approach to state of the art wireless mesh network solutions like IEEE 802.11s.