Reducing the Variability in Routing Decisions in Wireless Mesh Networks

Wireless mesh networks are multi-hop wireless networks which are useful in situations where there is little or no network infrastructure. Recently, those networks have been proposed for scenarios of Smart Grids and Internet of Things. The variability in routing metrics, which causes frequent changes in path choice, is a challenge in these networks. In this article, we propose a method, called Hypothesis Testing Estimator (HTE), for estimating the frame delivery probability of a link based on the concept of hypothesis testing. In this method, the estimate is fixed until the link behavior in the recent past significantly deviates from the expectation. The method was evaluated using simulations whose results show a significant reduction in the variability of the link quality estimates.