Research On Adaptive Clustering Algorithm And Mobile Hidden Station Problem For MANET

Mobile Ad Hoc Network (MANET) is a set of mobile nodes who communicate with each other without the help of fixed infrastructure or central management. In this condition, hierarchical structure based on culsters is necessary to estabilish a large-scale muti-hop wireless network with higher scalabilty. Certain nodes in MANET are choosen as clusterheads who are responsible for the maintenance of clusters and rational allocation of resources. However, nodes in MANET keep the process of association and dissociation to clusters due to their dynamic nature. Collision can be caused if more than one nodes acess channel to transmit data packets.Therefore, it is important to improve clustering algorithm and remove hidden and exposed nodes in a cluster to obtain a better network performance,such as lower clustering-related maintenance costs, longer cluster stable time and better nodes'fairness.This paper analyze the advantages and disadvantages of the existing MANET clustering algorithms. On the basis of comprehensive analysis of future demands of MANET, propose three new network factors and prove their rationality by MATLAB simulations. Then we chose a adaptive compromise algorithm called AWCA to make a thorough research on the combination of stabilty and fairness of network. In the end, this paper give a new thought to reduce the probability communication collisions between cluster-heads and member nodes called Improved RTS-CTS Algorithm. This novel algorithm can improve throughput of network and lower collision rate of network communication.However there are not many choices on MANET simulation software. Matlab is a powerful network simulation software. So we use Matlab to built simple simulation environments for the simulation experiments to compare novel algorithms with traditional algorithms. Experiments'results show that new algorithms achieve much better performance on network features.