k-Connectivity Estimation from Local Neighborhood Information in Wireless Ad Hoc and Sensor Networks

Abstract

A robust wireless ad hoc and sensor network tolerates the failures of nodes without losing its connectivity. A network is k-connected if it remains connected after failures in any k-1 nodes. Finding the k value in a WSN provides useful information about its robustness and reliability. In this paper, we propose a distributed algorithm that provides more accurate estimations than the existing solutions by collecting the upper and lower bounds of local estimations in a single node and taking the average of selected bound. The comprehensive simulation results reveal that the proposed algorithm finds up to 10% more accurate estimations and up to 37% lower mean square error values with lower energy consumption than its closest competitor.