Protocol Analysis And Modeling Of The Wireless Self-organizing Networks

Wireless self-organizing networks technology is one of networks developed from the 1970s. Due to its fast automatic self-configuring and without the infrastructure, it has obvious advantages in earthquake disaster prevention, weather monitoring, military intelligence fields, and will be widely applied in the future. As wireless self-organizing networks application be promoted, the problems of information security is becoming more and more prominent, such as network security architecture, security routing, safe access and key management and so on. How to carry on the wireless network protocols analysis and the model establishment are mainly researched in this thesis.In wireless self-organizing networks, the nodes with different scale require different real-time and reliability. In order to optimize the performance of the network, it is necessary to choose appropriate network topology and hierarchical management mode. Well, the realization of the hierarchical management relies on the clustering algorithm. Firstly, this thesis analyzes the advantages and disadvantages of several classical clustering algorithms. In which the weight-based clustering algorithms are welcome because of its better adaptability and flexibility. But they do not take into account the problem of non-uniform units when processing the weight factors. So, an improved adaptive on-demand weighted clustering algorithm(ImAOW) is proposed. In the algorithm, the performance of nodes are unitary changed into value without unit to totally evaluate the performance of nodes. Then, an example of clustering is presented, and the communication overhead of ImAOW clustering algorithm is analyzed. The result shows that, comparing with the similar typical algorithms, the ImAOW clustering algorithm has lower time complexity and message complexity.Secondly, the routing protocols in the wireless self-organizing networks and their security threats are introduced. And the AODV routing protocol and the blackhole attack are analyzed in detail. Then, an attack method based on malicious mobile is presented by the freedom movement characteristic of network nodes. We make some simulations with NS2, the result shows that this method has simple acts, considerable threats.Thirdly, a blocking attack method is proposed based on the security flaw of RTS/CTS handshake mechanism. The pair attackers conspire to forge malicious information, and make the communication among the nodes in the same cluster are blocked utilizing the fact that Custer-type structure is easily to lead to the false blocking problem. The blocking attack is a kind of Denial of Service attack. Moreover, a defense strategy against the proposed attack is presented. The simulations with NS2 show that the throughput falls down much when the attack happen. On the contrary, it goes up greatly when the defense strategy is implemented.Finally, the operation processes of IEEE 802.11 MAC protocol in wireless self-organizing networks are described as different state transitions by using Finite State Machine. Based on this description, a Finite State Machine model of MAC protocol in wireless self-organizing networks is established. Under this model, how attackers avoid successfully being detected and make efficiently attacks to MAC protocol is studied. Simulations with NS2 show that the average throughput of the network is decreased as attacks are strengthened, while the consumption of the route is increased gradually.