Research On Mobile Ad Hoc Network Routing And Correlative Security Technologies

Mobile Ad Hoc Network (MANET) is a multi-hop temporary autonomous system consisting of mobile terminals equipped with wireless transmitters and receivers. Communication in MANET relies on nodes’ cooperation and operates in a wireless multi-hop manner. MANET has the advantage of rapid constuction, flexibility and strong survivability, and is free from limits of fixed network infrastructures. MANET can be applied not only to military communications, but also for environmental monitoring, disaster relief, field scientific investigation, home networking and other civilian communications. But the open network architecture, shared open wireless medium, limited network capabilities and dynamic network topology, have made the traditional routing protocols for fixed network are no longer suitable. Designing routing protocols, which would meet the characteristics and requirements of MANET, is a core issue which must be addressed in the practical application of MANET.Existing MANET routing protocols are facing many problems in routing efficiency and scalability respect, and due to lack of effective security mechanisms, the routing protocols can not effectively respond to malicious behavior and unreliable quality of service in the network system. Focusing on routing efficiency and routing security issues in MANET, this dissertation makes comprehensive and intensive research. As to replenish and improve the blank and vacancy of recent research, efficient routing models and and effective security mechanisms were constructed. The specific contents are as follows:(1) The integration methods of structured P2P routing and MANET routing were fully studied. In view of symmetric structured P2P routing algorithm, a MANET routing model based on Kademlia, KDSR, was proposed. KDSR adopted DHT technology to map the physical nodes into the XOR-based overlay topology, and thus achieved P2P-based routing. Highly efficient route discovery and route maintenance methods were proposed, and unique route optimization strategies were proposed respectively. Simulation results show that compared to existing routing algorithms and protocols, KDSR has made a greater advantage in the routing performance, and acheives better scalability and robustness.(2) According to asymmetric structured P2P routing algorithm, a Terrace-based MANET routing model was proposed. The MANET system was projected into the Terrace topology and the MAENT Routing Tree (MRT) was defined. An efficient MRT construction algorithm was given, and its convergence was proved. The routing model adopted prefix matching based routing in MRT and dynamic source routing in physical network. Efficient DHT based routing optimization strategies and MRT topology repair procedures were designed. Simulation results show that, compared with the traditional routing algorithms, this routing model has the advantage of higher routing efficiency, stronger dynamic adaptability and better scalability.(3) Aiming at nodes’malicious behaviour, a distributed trust mechanism based on Bayesian framework was proposed. First, the trust model was constructed, and the trust calculation was fully analyzed with respect to accuracy and security of trust. The distributed trust management mechanism was studied in depth. Procedures for evidence collection and recommendation obtaining were designed. The trust evaluation system was presented and combined into MANET routing protocols. Simulation results show that this trust mechanism can effectively identify malicious nodes and dishonest recommenders, so that greatly enhance the overall network performance and ensure the network availability.(4) Aiming at node’s selfish behavior, the incentives mechanism for MANET based on game theory was proposed. First, a single-stage packet forwarding game model is established to analyze the "Prisoner’s Dilemma" in MANET. A Stern Tit for Tat Strategy (STFT) was designed to enforce node’s cooperation. Then, a repeated game model was constructed, its Nash equilibrium was solved, and the incentive consistency condition was obtained. An evolutionary game model was constructed to analyze the dynamic process of selfish nodes turning to cooperate, and the evolutionary stability of STFT had also been proven. Simulation results show that by setting reasonable penalty parameters, nodes’cooperations can be effectively motivated, so as to enhance the network availability and prolong network lifetime.The research work of this dissertation is supported by National Natural Science Foundation of China and The Science and Technology Plan of Shenyang. Relevant research results have been applied to the real prototype system and product development.