Using linear system reliability to obtain theoretical understanding of wireless routing

摘要

Wireless multihop networks have a wide variety of applications, due to their rapid deployment times and minimal configuration requirements. Transmissions in wireless networks may require performance guarantees, which can be achieved by using advanced routing strategies. We examine one such performance metric, namely reliability, or packet delivery ratio in a failure-prone wireless network. We use control theoretic methods to obtain an understanding of routing in wireless multihop networks. In particular, we model ad hoc wireless networks as stochastic dynamical systems where, as a base case, a centralized controller pre-computes optimal paths to the destination. This technique can be used to obtain the highest achievable reliability for a given transmission. We compare this approach with the reliability achieved by some of the widely used routing techniques in multihop networks. We also propose extensions to the base case that can be more applicable to practical scenarios. Results show that our approach can be used to theoretically characterize reliability of end-to-end transmissions in wireless networks.