Research On Group Key Agreement Scheme Based On ECC And Clustering Algorithm For Manet

Mobile Ad Hoc network is a distributed self-organization, self-healing network. The network is characterized by dynamic topology, limited resources, multi-hop communications and fragile network of safety. In view of the characteristics above, to ensure security is the basic requirements for key management in the Mobile Ad hoc network, otherwise, we should minimize the amount of node computation. Paper proposes a cryptosystem base on elliptic curve group key agreement considering the premise of network security elements; at the same time, for the reason of the Ad Hoc network nodes with limited resources, paper propose an improved clustering algorithm.First of all, the research background, Ad Hoc Network Security and Research of group key agreement, group key management problems, elliptic curve cryptosystem, as well as the realization of the agreement GDH.3 process are introduced; an agreement based on Elliptic Curve Cryptography system GDH.3 is proposed, then paper describe the details of the program and steps to achieve; analyzes and verifies the improved performance of group key protocols through the simulation results.Secondly, the article bases on Distributed Efficient Clustering Approach propose an improved clustering algorithm which is applicable to Ad Hoc Networks. On the basis of the DECA algorithm the improved algorithm takes into account the energy consumption load-balancing factor of cluster head. The improved algorithm through the node energy consumption model calculates the cluster head energy and the average energy consumption. Use the energy consumption variance to describe the energy consumption load balance.Finally, the article introduce the simulation platform NS-2 in brief, simulate the process of the improved algorithm which bases on the energy consumption of the sub-cluster load balancing. The simulation consider the factors of energy consumption of cluster head load balance, the time of reconstruction in fixed period and the survival time of cluster head. The simulation results show, compared with the DECA algorithm the improved algorithm has greatly improved in load-balanced cluster head node and extending the network life time.