Efficient distributed lifetime optimization algorithm for sensor networks Jacques Bahi a , Mohammed Haddad b,c , Mourad Hakem a,⇑ , Hamamache Kheddouci b,c a DISC Laboratory, Institut FEMTO-ST, UMR CNRS, Université de Franche-Comté, France b LIRIS Laboratory, UMR CNRS, Université de Lyon, F-69003, France c Université Claude Bernard Lyon 1, F-69622, France article info Article history: Received 12 November 2012 Received in revised form 1 June 2013 Accepted 20 November 2013 Available online 28 November 2013 Keywords: Lifetime optimization Sensor networks Reliability Self-stabilizing algorithm Weibull distribution abstract One of the main design challenges in Wireless Sensor Networks (WSN) is to prolong the system lifetime, while achieving acceptable quality of service for applications. In WSN, each sensor node is battery powered and it is not convenient to recharge or replace the bat- teries in many cases, especially in remote and hostile environments. Due to the limited capabilities of sensor nodes, it is usually desirable that a WSN should be deployed with high density and thus redundancy can be exploited to increase the network’s lifetime. In this paper, we introduce an efficient lifetime optimization and self-stabilizing algorithm to enhance the lifetime of wireless sensor networks especially when the reliabilities of sen- sor nodes are expected to decrease due to use and wear-out effects. Our algorithm seeks to build resiliency by maintaining a necessary set of working nodes and replacing failed ones when needed. We provide some theoretical and simulation results, that fully demonstrate the usefulness of the proposed algorithm. Ó 2013 Elsevier B.V. All rights reserved. 1. Introduction Recent years have witnessed significant advances in wireless sensor networks which emerge as one of the most promising technologies for the 21st century [2]. In fact, they present huge potential in several domains ranging from health care applications to military applications. However, the advent of Wireless Sensor Networks (WSN) and its conceptual constraints, have posed a number of re- search challenges to the networking and distributed com- putation communities. A problem that has received renewed interest is lifetime optimization. Indeed, as the scale is expanding, node failures are more likely to occur and can have an adverse effect on the application. In addi- tion, as sensor nodes have limited resources in terms of memory, energy and computational power, they suffer from increased susceptibility to wear-out effect [23,25,26]. In WSN, if the wear-out failures are not taken into con- sideration during the execution of the involved application, some nodes may age much faster than the others and be- come the reliability bottleneck for the network, thus signif- icantly reducing the system’s service life. In the literature, this problem has been formulated and studied in various ways. For instance, prior work [14,30,36,37] in lifetime reliability assumes node’s failure rates to be independent of their usage times. While this assumption can be ac- cepted for memoryless soft failures, it is obviously inaccu- rate for the wear-out-related fail-silent (a faulty node does not produce any output) and fail-stop (no node recovery) failures, because the sensor node’s lifetime reliability will gradually decrease over time. Since sensor nodes have limited battery life and with- out being able to replace batteries, especially in remote and hostile environments, it is desirable that a WSN should be deployed with high density and thus redundancy can be 1570-8705/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.adhoc.2013.11.010 ⇑ Corresponding author. E-mail addresses: Jacques.Bahi@univ-fcomte.fr (J. Bahi), Mohammed. Haddad@univ-lyon1.fr (M. Haddad), Mourad.Hakem@univ-fcomte.fr (M. Hakem), Hamamache.Kheddouci@univ-lyon1.fr (H. Kheddouci). Ad Hoc Networks 16 (2014) 1–12 Contents lists available at ScienceDirect Ad Hoc Networks journal homepage: www.elsevier.com/locate/adhoc