Research On Cluster Algorithm And Energy-saving Strategy In MANET

The mobile Ad Hoc network (MANET) is a temporary network fixed with no basic station and centralized management. Nodes's ruleless moving makes routing arithmetic and QoS guarantee etc face new difficulty. It is important to select appropriate network topology architecture to exert capability of network according to network scale of appliance and expansibility, and appliance request of reliability and real time.Cluster in network gives an efficient way for these difficulties. Most existing clustering algorithms are based on network topology informtion. Some used the mobility information of the nodes. But none of them explorn and used group mobility information of the nodes, so all the existing clustering algorithms could not support mobile collaboration well.This paper presents a mobility-adaptive k-hop (MAKH) clustering algorithm, which forms variable-diameter clusters based on node mobility patterns. It provides support to the group movement by using fuzzy logic, which enables nodes with the same movement pattern to be divided into the identical cluster, thus cluster achieves the stability by maximum limit. Unlike existing clustering algorithms, the diameter of clusters is flexible and determined by the stability of nodes. This makes MAKH a truly adaptive, distributed and localized algorithm. Fristly, the performance of MAKH is examined from a theoretical perspective where both the time and message complexities are derived. We showed the overhead incurred by multihop clustering. Subsequently, this paper presents the empirical results of MAKH based on a series of extensive NS-2 simulations. In three performance parameters of the connectivity, the number of cluster and probability of the node joining in the cluster, the simulation results show that MAKH forms clusters which are more stable than those formed by Lowest-ID and Max Connectivity Clustering Algorithm in both Random Waypoint and Reference Point Group Mobility Model.Besides, we propose a power-saving mechanism based on the characteristics of cluster roles.Based on the IEEE802.11 and WLAN PC card's product brief, we build a power transition diagram. Then discuss the suitable energy conservation mechanism respectively for four kinds of roles:the clusterhead, the gateway, the middle node and the leaf node. Furthermore, various reference tables are built based on our simulation results. According to the numerical factors of these tables, suitable power-saving schemes can be obtained for each role under various environments.