DFA Based QoS-Aware Clustering Approach for Future Prediction and Performance Improvement in MANET

摘要

The tremendous usage of mobile nodes in wireless communication medium makes energy efficiencya fundamental requirement of Mobile Adhoc Network. Energy plays a vital role in prolonging the network lifetime. DFA based QOS Aware Clustering (DFA-QAC) for the Energy-aware Optimized Link State Routingprotocol (E-OLSR) is used. This algorithm takes into account the mobility, residual energy of the nodes andgives major improvements regarding the number of elected cluster heads. The objective is to elect a reasonablenumber of cluster heads which regulates the network traffic and preserving bandwidth. The proposed schemeDFA-QAC uses modified weighted clustering technique which helps to decide the mode of each node. It canable to select the nodes having highest residual energy as a cluster head and able to switch between states.DFA based representation helps to determine the pattern of switching between various states and can able topredict average energy required for cluster formation. Reclustering mechanism helps in electing the new clusterhead, when a node has entered or it has just left its cluster. The results of the proposed scheme E-OLSR performbetter than the traditional AODV protocol. The performance metrics such as energy consumption. Packetdelivery ratio and throughput has been analyzed in the proposed scheme.