PKF: A communication cost reduction schema based on Kalman filter and data prediction for Wireless Sensor Networks

The implementation of a wireless sensor mote as a system on chip (SoC) enables the energy consumption to be decreased substantially by using dedicated hardware accelerators. This work combines an efficient prediction model with a Kalman filter (PKF) to reduce the communication cost in Wireless Sensor Networks (WSNs) with a guaranteed data quality. The hardware accelerator requires fewer resources than previous approaches, while achieving higher energy reductions. Exhaustive experimental results based on datasets from a real WSN application confirm the advantages of the proposed mechanism.