Details

Autor(en) / Beteiligte

• AYUSHREE
• SANDEEP KUMAR ARORA

Titel

COMPARATIVE ANALYSIS OF AODV AND DSDV USING MACHINE LEARNING APPROACH IN MANET

Ist Teil von

• Journal of engineering science & technology, 2017-12, Vol.12 (12), p.3315-3328

Ort / Verlag

Taylor's University

Erscheinungsjahr

2017

Link zum Volltext

Quelle

EZB Electronic Journals Library

Beschreibungen/Notizen

• Mobile Ad-Hoc networks possess a dynamic structure which is characterized by the absence of central administrator. Due to such dynamic network, the possibilities of acquisition of optimal path diminish to a great extent and hence the durability of the optimal transmission of data packet becomes severe. Each and every node in MANET is battery powered up and mobile in nature, hence mobility becomes the prime reason of energy exhaustion in such network. The main objective of presented paper is to attain the most reliable path with least mobility for successful transmission of data packets. The algorithm used for attainment of optimal path is knowledge based learning algorithm which is implied over two routing protocols; AODV (Ad-Hoc On Demand Distance Vector Routing) and DSDV (Destination Sequence Distance Vector Routing). The performance evaluation is done by means of Relay Number which is inversely proportional to the mobility of node. AODV and DSDV are further employed over network systems with varying number of nodes, i.e., 12 and 24 nodes network system. The performance comparison is made on the basis of two performance parameters such as throughput and PDR (Packet Delivery Ratio). A proposition is made that analysis of PDR and throughput in knowledge based learning algorithm is better in comparison with other traditional techniques like Destination Sequence Distance Vector (DSDV). The simulation is performed over NS-2 network simulator, which enables the implementation of wired and wireless simulation.