Energy Efficient Adaptive Sensing Framework for WSN-Assisted IoT Applications

Wireless Sensor Network (WSN) based Internet of Things (IoT) solutions are extensively deployed to monitor environmental parameters and physical conditions in various domains. However, wireless sensor nodes have limited energy sources and in many cases are deployed in out-of-reach areas, which makes their substitution impossible. As a result, maximising network longevity requires efficient energy use. This paper suggests an adaptive sampling strategy based on the spatio-temporal correlation among sensor readings and the residual energy on a cluster-based network. We propose the implementation of an adaptive sampling framework which operates in cycles; each cycle is composed of two periods: learning period and adaptive sampling period. In the first period, all sensor nodes are actives and transmit sensed data to the base station, while in the second period, sets of disjoint sampling nodes are activated successively in each cluster to sense and send data to a selected cluster head. Further, a data reconstruction algorithm is implemented in the base station to retrieve non-sampled data. Extensive simulations were conducted to validate the efficacy of the proposed framework using synthetic data. Simulation findings show an energy saving up to 47% while maintaining a good data quality.