Back propagation neural network-based energy efficient routing protocols for mobile ad-hoc networks. (12th December 2011)

• Record Type: Journal Article
• Authors: Karia, Deepak Chatrabhuj; Lande, Bhaurao K.; Daruwala, Rohin D.
• Is Part Of: International journal of intelligent engineering informatics. Volume 1: Number 3/4 (2011)
• Page Start: 246
• Page End: 260
• Publication Date: 2011-12-12
• Subjects: routing protocols -- ad-hoc on-demand multipath distance vector protocol -- AOMDV -- energy efficient ad-hoc on-demand multipath distance vector -- EE-AOMDV -- optimal ad-hoc on-demand multipath distance vector -- OP-AOMDV -- hop count -- residual energy -- back propagation neural network; Artificial intelligence -- Engineering applications -- Periodicals; Engineering -- Computer programs -- Periodicals; Knowledge management -- Periodicals; 620.0028563
• Languages: English
• ISSNs: 1758-8715

Abstract

The focus of this paper is to take into account several factors in wireless networks routing design, the rationale of which are the routing protocols that are designed only based on one criterion, e.g., shortest path or residual energy. The paper proposes two energy efficient routing schemes which consider residual energy, shortest path and network lifetime. These protocols, i.e., energy efficient (EE-AOMDV) and optimal (OP-AOMDV), extend the node-disjoint version of ad-hoc on-demand multipath distance vector protocol (AOMDV). The first EE-AOMDV protocol considers the hop count of the route and residual energy of the node and determines energy efficient route with the help of back propagation neural network, where we define a routing metric as the ratio of residual energy over initial energy and which is exchanged over the routing packets. Also, the second OP-AOMDV protocol considers multiple criterions to select an efficient shortest path route for data transfer. Our simulation results show that energy can be efficiently utilised with the proposed protocols as compared to original AOMDV protocol.