Open Access

ARTICLE

An Optimal Cluster Head and Gateway Node Selection with Fault Tolerance

P. Rahul^*, B. Kaarthick

Coimbatore Institute of Engineering and Technology, Coimbatore, Thondamuthur, 641109, Tamil Nadu, India

* Corresponding Author: P. Rahul. Email:

Intelligent Automation & Soft Computing 2023, 35(2), 1595-1609. https://doi.org/10.32604/iasc.2023.025762

Received 03 December 2021; Accepted 28 February 2022; Issue published 19 July 2022

Abstract

In Mobile Ad Hoc Networks (MANET), Quality of Service (QoS) is an important factor that must be analysed for the showing the better performance. The Node Quality-based Clustering Algorithm using Fuzzy-Fruit Fly Optimization for Cluster Head and Gateway Selection (NQCAFFFOCHGS) has the best network performance because it uses the Improved Weighted Clustering Algorithm (IWCA) to cluster the network and the FFO algorithm, which uses fuzzy-based network metrics to select the best CH and entryway. However, the major drawback of the fuzzy system was to appropriately select the membership functions. Also, the network metrics related to the path or link connectivity were not considered to effectively choose the CH and gateway. When learning fuzzy sets, this algorithm employs a new Continuous Action-set Learning Automata (CALA) approach that correctly modifies and chooses the fuzzy membership functions. Despite the fact that it extends the network’s lifespan, it does not assist in the detection of defective nodes in the routing route. Because of this, a new Fault Tolerance (NQCAEFFFOCHGS-FT) mechanism based on the Distributed Connectivity Restoration (DCR) mechanism is proposed, which allows the network to self-heal as a consequence of the algorithm’s self-healing capacity. Because of the way this method is designed, node failures may be utilised to rebuild the network topology via the use of cascaded node moves. Founded on the fractional network information and topologic overhead related with each node, the DCR is suggested as an alternative to the DCR. When compared to the NQCAFFFOCHGS algorithm, the recreation results display that the proposed NQCAEFFFOCHGS-FT algorithm improves network performance in terms of end-to-end delay, energy consumption, Packet Loss Ratio (PLR), Normalized Routing Overhead (NRO), and Balanced Load Index (BLI).

---

Keywords

---