Data Collection using Artificial Intelligence on Wireless Sensor Networks

Abstract

Data collection is a challenging task in wireless sensor networks. In applications of wireless sensor networks, data collection is a wise choice due to the constraints in communication bandwidth and energy budget. This paper focus on efficient data collection with error bounds in wireless sensor networks. The key idea of our data collection approach is to divide a sensor network into clusters, find out local data correlations on each cluster head, and perform global data collection on the sink node based on parameters uploaded by cluster heads. Specifically, we propose a local estimation model to approximate readings of sensor nodes in subsets, and prove rated error-bounds of data collection using local estimation model method. In the process of data collection method, we formulate the problem of selecting the minimum subset of sensor nodes into a minimum dominating set problem which is known to be NP-hard, and greedy heuristic algorithm to find an approximate solution. We use another monitoring algorithm to adjust the composition of node subsets according to changes of sensor readings. Artificial Networks (ANNs) has been used for missing field data recovery. An architecture involve hash technique has been used to remove duplicate data values from a dataset. Finally we demonstrate that data collection remarkably reduces communication cost of data collection with guaranteed error bounds.
Keywords: Data collection, Sensor nodes, artificial neural networks