A hybrid classifier approach to multivariate sensor data for climate smart agriculture cyber-physical systems

Abstract

In this paper, we propose a novel climate-smart Agriculture Cyber-Physical System (ACPS) for precision farming. The primary motive of the ACPS is to perform real-time fault location tracking in the agricultural field using multivariate sensor data. The computing model in the ACPS uses a novel hybrid classification approach which combines two classifiers for the location estimation of the sensor node. The novelty of the proposed method lies in predicting the locations that need more irrigation, soil nutrients or immediate human intervention using the sensor data. We also derive the computational complexity of the proposed method. The location accuracy improves reasonably as compared to the current-state-of-the-art methods.