Reliability Modeling and Monte Carlo-Based Simulation for Optimal Wireless Sensor Networks Lifetime Assessment

Abstract

Regarding the wide area of wireless sensor networks (WSNs) applications during the recent years, the research challenges such as lifetime optimization, reliability, maintainability, and resilience have become very significant. The non-disjoint set covers (NDSC)-based coverage control approach with its capability of sensors to be scheduled in one or more covers for one or more monitoring seasons has brought out a better performance in terms of the lifetime optimization. Also, it yields a promising indicator regarding the WSNs’ reliability and resilience. This paper addresses the WSNs’ reliability assessment via a NDSC approach applied for WSNs’ lifetime optimization. It specifies the mathematical model for the WSNs’ lifetime considering the energy reserve and the failure hazard. Then, it introduces a dynamic simulation method based on NDSC, using the Monte Carlo method for the WSNs’ reliability assessment. For a WSN with m sensors participating in q NDSC scheduled for a given number of sensing periods, our method could estimate the reliable and predict the WSNs’ lifetime, considering the expected failure hazards on a given time horizon corresponding to the number of monitoring seasons. We have considered that a WSN with a number of NDSC equal to q has the capability to perform the coverage task while q is greater than zero and reserve covers are available to be activated if the current cover fails. An experimental study by simulation using C programming language allows explaining the failure probability effects. For a given instance, the maximum lifetime decreases to 18.74% when the failure probability increases from 0.0001 to 0.1. The investigations have shown the faulty covers recovery capability enabling to design resilient strategies using the NDSC in future works.