An energy-efficient data prediction and processing approach for the internet of things and sensing based applications

The Internet of Things (IoT) is a vision in which billions of smart objects are linked together. In the IoT, "things" are expected to become active and enabled to interact and communicate among themselves and with the environment by exchanging data and information sensed about the environment. In this future interconnected world, multiple sensors join the internet dynamically and use it to exchange information all over the world in semantically interoperable ways. Therefore, huge amounts of data are generated and transmitted over the network. Thus, these applications require massive storage, huge computation power to enable real-time processing, and high-speed network. In this paper, we propose a data prediction and processing approach aiming to reduce the size of data collected and transmitted over the network while guaranteeing data integrity. This approach is dedicated to devices/sensors with low energy and computing resources. Our proposed technique is composed of two stages: on-node prediction model and in-network aggregation algorithm. The first stage uses the Lagrange interpolation polynomial model to reduce the amount of data generated by sensor nodes while, the second stage uses a statistical test, i.e. Kolmogorov-Smirnov, and aims to reduce the redundancy between data generated by neighbouring nodes. Simulation on real sensed data reveals that the proposed approach significantly reduces the amount of data generated and transmitted over the network thus, conserving sensors' energies and extending the network lifetime.