Loss-aware linear network coding for wireless networks

Recently, there has been a growing interest in employing network coding to improve the performance of wireless networks. Some authors adopt XOR network coding methods, e.g. COPE and BEND in multi-hop wireless networks, which can increases the network throughput obviously in loss-free or light lossy network. However we argue that in heavily lossy wireless environments, a better use of network coding is to provide higher throughput. Instead of XOR network coding, we adopt linear coding method to adapt the lossy wireless networks. In this paper, we present a detailed exploration of the linear network coding and propose our linear coding algorithms that can lead to higher throughput gains in wireless scenarios, and we employ double-decoding method to decode the transmitted packets. By means of accurate analysis, rigorous simulations, and detailed implementation on a simulation platform, the simulation result demonstrates that our algorithms can obtain higher throughput gain in lossy wireless networks.