A Q-Learning-Based Fault-Tolerance Data Routing Scheme for IoT-Enabled WSNs

Wireless sensor network (WSN) is one of the essential ingredients of any Internet of Things (IoT)-based system. In WSNs, sensor nodes (SNs) are responsible for collecting real-time data from the monitoring environment. Energy is one of the most significant resources in SNs, which is used to sense and transmit data to the base station (BS). IoT-enabled WSN is a resource-constrained network. The collection of voluminous data from resource-constrained sensors creates several challenging issues within the network, such as poor network lifetime (NLT), message overhead (MOH), and data transmission delay. Furthermore, these sensors are vulnerable to fault due to deployment in harsh environments and natural calamities. It drastically reduces NLT as well as the overall performance of the network. Thus, IoT-enabled WSNs require an energy-efficient fault-tolerant data routing scheme to enhance network performance. This article proposes a novel mobile sink-based fault-tolerance scheme with Q-learning to enhance NLT and overall performance of the networks. A genetic algorithm-based optimal cluster head selection mechanism is also proposed to improve energy efficiency and balance the energy consumption across the network. Extensive simulations and testbed experiments are performed to prove the out-performance of the proposed scheme in terms of NLT, average packet loss ratio, average energy consumption, data transmission latency, MOH, and fairness index.