Open Access

ARTICLE

IoT-EMS: An Internet of Things Based Environment Monitoring System in Volunteer Computing Environment

Sourav Kumar Bhoi¹, Sanjaya Kumar Panda², Kalyan Kumar Jena¹, Kshira Sagar Sahoo³, N. Z. Jhanjhi^4,*, Mehedi Masud⁵, Sultan Aljahdali⁵

1 Department of Computer Science and Engineering, Parala Maharaja Engineering College (Govt.), Berhampur, 761003, India
2 Department of Computer Science and Engineering, National Institute of Technology, Warangal, 506004, India
3 Department of Computer Science and Engineering, SRM University, Amaravati, 522502, India
4 Department of Computer Science and Engineering, Taylor’s University, Subang Jaya, 47500, Malaysia
5 Department of Computer Science, College of Computers and Information Technology, Taif University, Taif, 21944, Saudi Arabia

* Corresponding Author: N. Z. Jhanjhi. Email:

Intelligent Automation & Soft Computing 2022, 32(3), 1493-1507. https://doi.org/10.32604/iasc.2022.022833

Received 20 August 2021; Accepted 22 September 2021; Issue published 09 December 2021

Abstract

Environment monitoring is an important area apart from environmental safety and pollution control. Such monitoring performed by the physical models of the atmosphere is unstable and inaccurate. Machine Learning (ML) techniques on the other hand are more robust in capturing the dynamics in the environment. In this paper, a novel approach is proposed to build a cost-effective standardized environment monitoring system (IoT-EMS) in volunteer computing environment. In volunteer computing, the volunteers (people) share their resources for distributed computing to perform a task (environment monitoring). The system is based on the Internet of Things and is controlled and accessed remotely through the Arduino platform (volunteer resource). In this system, the volunteers record the environment information from the surrounding through different sensors. Then the sensor readings are uploaded directly to a web server database, from where they can be viewed anytime and anywhere through a website. Analytics on the gathered time-series data is achieved through ML data modeling using R Language and RStudio IDE. Experimental results show that the system is able to accurately predict the trends in temperature, humidity, carbon monoxide level, and carbon dioxide. The prediction accuracy of different ML techniques such as MLP, k-NN, multiple regression, and SVM are also compared in different scenarios.

---

Keywords

---