Machine learning based multi scale parallel K-means++ clustering for cloud assisted internet of things
- PDF / 2,441,664 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 57 Downloads / 201 Views
Machine learning based multi scale parallel K-means++ clustering for cloud assisted internet of things S. K. Mydhili 1 & S. Periyanayagi 2 & S. Baskar 3 & P. Mohamed Shakeel 4 & P. R. Hariharan 5 Received: 29 May 2019 / Accepted: 2 August 2019 # Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract Cloud assisted Internet of Things (CIoT) is the technology initiated towards the deployment of virtualization in wireless sensor networks (WSN). In the Recent times, parallel clustering routing schemes is the significant area of research to optimize energy and scalability problem in WSN on CIoT. In clustering schemes, Data collection, Aggregation and communication is the important activity which has been optimized through various conventional approaches which are not accounted to generate balanced clusters with optimum energy consumption and closer to corresponding cluster heads. In this research the novel Multi Scale Parallel K-means++ (MSPK++) clustering algorithm with balanced clustering has been proposed and improved further by applying machine learning techniques suitable for WSN on CIoT environment. The algorithm convergence has been proved globally and locally whereas the simulations are experimentally validated for the proposed algorithm in comparison with state of art algorithms in an acceptable complexity. Keywords Machine learning . CIoT . WSN . Multi scale parallel K-means++,parallel clustering
1 Introduction Wireless Senor Networks (WSNs) is the collection of independent nodes with limited sensing, which helps to monitor the environment conditions and the collected information is transmitted to the base station for making the effective communication system [1] in CIoT environment. The sensor network has several thousands of nodes that utilizes radio transceiver methodologies while implementing the communication process which has been influenced by several applications This article is part of the Topical Collection: Special Issue on AI-based Future Intelligent Internet of Things Guest Editors: Kelvin K.L. Wong, Quan Zou, and Pourya Shamsolmoali * S. Baskar [email protected] 1
Department of Computer Science and Engineering, KGiSL Institute of Technology, Coimbatore, India
2
Department of Electronics and Communication Engineering, Ramco Institute of Technology, Rajapalayam, Tamilnadu, India
3
Department of Electronics and Communication Engineering, Karpagam Academy of Higher Education, Coimbatore, India
4
Faculty of Information and Communication Technology, Universiti Teknikal Malaysia Melaka, Melaka, Malaysia
5
Junior Technical Assistant, Department of Scientific and Industrial Research, New Delhi, India
such as industrial monitoring, military applications, ecological observation, health monitoring, commercial applications and so on [2]. Due to various responsibilities of wireless communication process, the CIoT technology has been recently innovated though large number of sensor nodes with reasonable cost as well as compact as shown in the Fig. 1. Even though the innovated
Data Loading...
