文献类型： 外文期刊

作者： Zhang, Xiaheng ¹ ; Shu, Kunliang ² ; Rajkumar, S. ³ ; Sivakumar, V. ⁴ ;

作者机构： 1.Northwest Univ Polit Sci & Law, Business Sch, Xian 710122, Shaanxi, Peoples R China

2.Jilin Acad Agr Sci, Inst Agr Econ & Informat, Changchun 130033, Jilin, Peoples R China

3.Vellore Inst Technol, Sch Comp Sci & Engn, Vellore, Tamil Nadu, India

4.Vel Tech Rangarajan Dr Sagunthala R&D Inst Sci &, Sch Comp, Dept Comp Sci & Engn, Morai, Tamil Nadu, India

5.Qinghai Nationalities Univ, Sch Econ & Management, Xining 810007, Qinghai, Peoples R China

关键词： AI; Environmental monitoring system; Wireless sensor network; Internet of things

期刊名称：ENVIRONMENTAL IMPACT ASSESSMENT REVIEW （ 影响因子：7.9； 五年影响因子：6.8 ）

ISSN： 0195-9255

年卷期： 2021 年 86 卷

页码：

收录情况： SCI

摘要： Environmental monitoring, modeling, and managing allow a better understanding of major processing and techniques for managing environmental changes. The pollution level has risen over time due to many factors such as a rise in population and the use of the vehicle, industrialization, and urbanization that have a direct impact on people's health. Hence, in this paper, Artificial intelligence assisted Semantic Internet of Things (AISIoT) has been proposed using a wireless sensor network (WSN) for the environmental monitoring system and the real economy. The Artificial Intelligence technique can very effectively analyze data and make precise decisions on the provision of services in different types. This study provides a mathematical framework for the analysis of interdependent aspects of the WSN protocol for communication and design of signal processing. The Internet of Things (IoT) based framework comprises the complete information system from the sensor level to data management about the environment. The experimental results show that the proposed method provides an effective way to analyze the long-term monitoring of environmental data. The proposed AI-SIoT method using the WSN method enhances accuracy(95.6%), performance(98.7%) increase efficiency (93.7%) with reliability (97.4%) when compared to other existing methods.