Machine learning based MAC protocol design for pipeline leakage detection in smart city project. (4th July 2021)

• Record Type: Journal Article
• Authors: Sakya, Gayatri; Dalela, Chhaya; Singh, Laxman; Jain, Anuj
• Is Part Of: Journal of discrete mathematical sciences & cryptography. Volume 24:Number 5(2021)
• Page Start: 1283
• Page End: 1292
• Publication Date: 2021-07-04
• Subjects: 68Txx (Artificial intelligence) -- 68M10 (Network design and communication); Intelligent MWSN (Mission Critical WSN) -- IoT -- ML -- Smart city project -- Pipeline leakage detection; Computer science -- Mathematics -- Periodicals; Cryptography -- Periodicals; Computer science -- Mathematics; Cryptography; Periodicals; 004.0151
• DOI: 10.1080/09720529.2021.1932915
• Languages: English
• ISSNs: 0972-0529

Abstract

Abstract: The integration of WSN with IoT is growing exponentially for implementing real time applications in today's world. Also an Intelligent WSN design is required for implementing these applications. This paper discusses a MWSN based system for detection and alert of the pipeline leakage. In this method all the nodes are not allowed to communicate with the central server directly. Whenever a node receives a data of some irregularities in the system, the node sends it in multihop fashion to the nearest cluster node, which is the only node that is allowed to communicate with the central server. In the remote areas with no infrastructure of internet connection, this system can used to transmit the data to the server, thus the system is cost effective and efficient. As the number of sensor nodes increases and time bound data is required at the IoT devices and cloud for implementing smart applications. In the first phase the paper discussed the deployment strategy of wireless sensor nodes on pipe. In Phase 2, the paper discussed the ML based MAC protocol design. Here ML algorithm is discussed in the paper for energy efficient information transmission node to node. The duty cycle will be tuned based on ML models as discussed. The decision tree classifier will be finally chosen to tune the duty cycle of wireless sensor node. In the final phase the IoT application and website designing is discussed to collect the data and send it to cloud for observing the status of leakage in each zone and sending the alerts to area manager. The work is novel in terms of intelligent model development for pipeline leakage detection and has can be modified for number of mission critical scenarios.