The Recognition Algorithm And Research Of Bottleneck Node In Sensor Network

As a new information access and process technology, wireless sensor networks integrate of sensor technology, micro-electromechanical systems technology, embedded computing technology, distributed information processing technology and network communication technology. It can collaboratively monitor, sensor and collect the information of environment or monitoring object within the area of network coverage in real-time and process the information to obtain detailed and accurate data to transmit to the users who need such information.In practical applications, the sensor nodes are generally powered by non-renewable battery, so how to efficiently utilize the limited energy is the greatest concern of wireless sensor networks. The bottleneck nodes in wireless sensor network refer to the isolated nodes that connect two or more regional due to random deployment. Because these nodes are in a special position, the inter-regional transmission of data must go through these nodes, so that their life is significantly less than other nodes. Once these nodes die, the network will be divided into several unconnected branch, resulting in the network is not working properly, or even the end of the network lifetime, so study this type of node has great significance.On the basis of the research work at home and abroad, this thesis studies the bottleneck nodes in wireless sensor networks and their impact on network performance, the research focuses on the identification algorithm of the bottleneck node. Firstly, this thesis makes a brief introduction of the sensor network-related technical background and basic knowledge. Then give the definition and characteristics of the bottleneck node, and analyze the impact of bottleneck node on the network performance by simulation. On this basis, propose several bottleneck node identification algorithms, including the quasi-bottleneck node algorithm, the bottleneck node identification algorithm, the bottleneck coefficient method and the bottleneck node local detection algorithm, and make a detailed description and analysis of these algorithms. The simulation result shows that the bottleneck node local detection algorithm has a high degree of accuracy on the detection of bottleneck node in the network. Finally, this thesis describes three solutions for the problem of the bottleneck node:data aggregation, additional support node and to improve the clustering routing protocol, and verify that they can better avoid the bottleneck node consume energy excessive and weaken the adverse effects of the bottleneck node on the network to extend the lifetime of the network by do the simulation.