A Nearest Neighbors based Data Filter for Fog Computing in IoT Smart Agriculture

Abstract

In smart agriculture, the Internet of Things (IoT) makes it possible to analyze and manage agricultural yield to increase productivity, reduce wasted resources, and decrease irrigation costs. In IoT systems, if data management is entirely performed in the cloud, the system may not work correctly due to connectivity problems, which is common in some remote regions where the agribusiness thrives. A fog computing solution enables the IoT system to process data faster and deal with intermittent connectivity. However, a high number of packets sent from the fog to the cloud can cause link congestion with mostly useless data traffic. Dealing with fog data filtering is a challenge because it requires knowing which data is essential to send to the cloud. This paper proposes an approach to collect and store data in a smart agriculture environment and two different methods filtering data in the fog. We designed an experiment for each filtering method, using a real dataset containing temperature and humidity values. In both experiments, the fog filters the data using the k-Nearest-Neighbors (kNN) algorithm, which classifies data into categories according to their value ranges. In the first experiment, the fog classifies the data and generates an output of the number of data categories. In the second experiment, data is classified and also compressed based on the previously obtained categories using the runlength encoding (RLE) technique to preserve the data time series nature. Our results show that data filtering reduces the amount of data sent by the fog to the cloud.