Title: Wireless sensor system for chemical and biological parameters monitoring and classification

Authors: Naser Zaeri; Rusul R. Qasim

Addresses: Faculty of Computer Studies, Arab Open University, Ardiya, 92400, Kuwait ' Faculty of Computer Studies, Arab Open University, Ardiya, 92400, Kuwait

Abstract: Wireless sensor systems provide powerful structures for monitoring and analysing data in complicated situations over extended periods of time. In this paper, we offer a smart integrated system composed of a number of buoys that are dispersed in the sea and are equipped with suites of sensors that can measure a variety of vital environmental, biological and chemical parameters. The system is a comprehensive and interactive multi-parameter monitoring structure with two-way communication through a wireless mobile network that implements machine learning to detect and categorise the various parameters. In pursuing so, four contemporary algorithms are applied and compared, namely: logistic regression, naïve Bayes, multilayer perceptron, and support vector machines. The experimental findings reveal that the support vector machine performs remarkably well, with best accurate classification rate of 86% with the corresponding weighted average precision, recall, F-score, and MMC of 0.86, 0.86, 0.858, and 0.84, respectively.

Keywords: chemical and biological monitoring; parameter classification; machine learning; sensors; wireless network.

DOI: 10.1504/IJSNET.2022.123605

International Journal of Sensor Networks, 2022 Vol.39 No.2, pp.125 - 135

Received: 03 Oct 2021
Accepted: 05 Dec 2021

Published online: 29 Jun 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article