Adaptive reliability satisfaction in wireless sensor networks through controlling the number of active routing paths

• We propose a reliability estimation algorithm for WSNs called RE-WSNs using Ordered Binary Decision Diagrams (OBDD).
• The proposed RE-WSNs estimates network’s reliability in terms of the reliability of all individual links.
• We propose a novel algorithm called Adaptive Reliability Satisfaction-Multipath Routing (ARS-MR).
• The proposed ARS-MR adaptively sets the sufficient number of active paths aiming at satisfying a desired reliability degree.
• The proposed ARS-MR algorithm can be used in conjunction with any arbitrary multipath routing algorithm in WSNs.

The number of active paths in multipath routing scenarios (with concurrent data transmission over the established paths) affects the provided reliability degree as well as the imposed overhead of path management. Since the reliability of individual links varies over the time, adaptively setting the sufficient number of active paths turns out to be essential. In this paper, we first propose a Reliability Estimation for WSNs (RE-WSNs) algorithm, based on ordered binary decision diagram (OBDD) data structure, which gives the network reliability in terms of the reliability of all individual links. Second, we propose a novel algorithm called adaptive reliability satisfaction–multipath routing (ARS–MR) which adaptively sets the sufficient number of active paths, aiming at keeping the network reliability within a desired quantitative range and minimizing path management overhead. In activation/inactivation process it further takes into account energy efficiency considerations. The proposed ARS–MR algorithm can be used in conjunction with any arbitrary multipath algorithm in WSNs. Simulation results with NS-2 reveal that ARS–MR is quite successful in timely reacting to variations of links reliability. Indeed, it manages the number of active paths and keeps the reliability of the network satisfactory over the course of network lifetime.