Sunday, March 14, 2010

Clustering Algorithm for Tactical Self-organized Network


Ad Hoc network is a collection of wireless mobile nodes dynamically forming a temporary network without use of any existing network infrastructure or centralized administration;and the mobile nodes in the Ad Hoc network communicate with each other using multi-hops wireless links.As a new wireless communication technology,it is because it has such characters as without using any infrastructure,being able to be set rapidly up and powerful ant-destructibility,that the digital war field always firstly selects it as its communication network.With-out question,Ad Hoc network has very important application network.At present,many countries have widely used Ad Hoc technology in military communication.But PLA doesn't use this new technology because the research in our country is still at a beginning level and there exists some questions that are not solved.The main conclusions are as follow in the paper:First of all,the paper firstly introduces the basic concepts,the main characteristics, application fields and the network structure of the Ad Hoc network;analyze the characteristics,functions and main factors influencing its properties of the tactical Ad Hoc network;and analyze and research the function demand for the Ad Hoc network construction under the tactics background.On the base of the current command relationship and the flow way of information,the architecture for three-level self-organizing network is proposed.The architecture has features of simple structure,high reliability and good applicability.Secondly,on the base of analyzing and comparing of Several Typical Clustering Algorithms,detailed analyzing specifically of AOW(Automatic On-demand Weighted algorithm),and improving of weights of AOW,the Paper brings forward CCA(Command-based Clustering Algorithms).This algorithm uses the tree structure knowledge.It provides a platform for the army using Ad Hoc to command.Final,using the well-known network simulation software OPNET,the paper analyzes and compares the performance of several typical clustering algorithms.As the result shows, CCA can be used well in the tactical Ad Hoc network.