Research On Several Key Technologies Of Wireless Sensor Networks

Wireless sensor network is a self-organizing network which is composed of a large number of low-cost, low-power tiny sensor nodes. These sensors can be used for monitoring, sensing, collecting the environment information in overlay area and transferring the processed data to users. WSN has many features, including flexible deployment, good scalability, high reliability and low-cost. It can be used in a wide range of potential applications, such as national security, industry control, medical health, environmental monitoring and commerce applications, etc. Both academia and industry are very interested in it.Wireless sensor networks are often used in harsh environments, and sensor nodes have limited and irreplaceable battery power, how to efficiently use network energy is a question of the first importance for routing protocols in WSNs. This paper will conduct mainly research on the maximum life communication model of multiple sinks, routing algorithm, fault diagnosis technology. Finally, the energy-efficiency prototype design of wireless sensor network will be proposed.(1) A multiple sinks communication model of maximum lifetime for WSNs.Generally, wireless sensor networks haven’t a fixed routing algorithm, and different optimal routing is designed to different application. It is one of key issue to how to improve energy-efficiency of data transmission from source code to sink. For the situation of multiple sinks in wireless sensor networks, the routing protocols based on optimization technology is poposed. Lifetime maximization is an important optimization problem specific to WSNs since they operate with limited energy resources. According to the energy constraints and data traffic conservation of wireless sensor networks, the problem is formulated as a mixed-integer program (MIP) problem by maximizing the time at which the first node runs out of energy. In this paper, relying on the optimization of a distributed algorithm, the MIP can be resolved. The performance of communication model in multiple inks environment is verified by simulation. Compared with WSTP and MD-MLR, the result shows that this model can maximize the network lifetime obviously.(2) A multi-path routing protocol based on improved ACO algorithm for WSNs.Wireless sensor networks consisting of nodes with limited power are deployed to gather useful information from the field. It is critical to collect the information and transfer it from source node to sink in an energy-efficient manner. Due to the special demand of energy control and balancing the energy of nodes in wireless sensor networks, energy-balance routing based on ant colony system was proposed. In this algorithm, probabilistic decision rule and fitness function are constructed with the energy of nodes, and the optimal path is found by calculating the probability and fitness value. Simulation results show that the algorithm based on wireless sensor network routing protocols can prolong the life cycle of wireless sensor network, improve the network performance.(3) Fault diagnosis based on multiple-kernel wavelet RVM Algorithm for WSNs.A faulty sensor node may produce incorrect data and transmit them to the others in WSNs. This would consume the limited energy and bandwidth of WSNs. Furthermore, the sink may make inappropriate decision when it receives the incorrect data. A multiple kernel wavelet relevance vector machine algorithm is firstly presented to fault diagnosis for wireless sensor network in the paper, and a new diagnosis algorithm is proposed. The values of the features and the corresponding state types of nodes are used to train multiple-kernel wavelet relevance vector machine classifier. At the same time, we also compare diagnosis results among MKWRVM, RBF RVM and SVM. It is indicated that the diagnosis results of the multiple-kernel wavelet relevance vector machine model for wireless sensor networks are better than those of RVM and SVM.(4) A prototype system based on energy-efficient algorithm for WSNs.An energy-efficient system prototype has been proposal based on researching algorithm above for wireless sensor networks. Its hardware is based on MSP4301611processor and CC1101RF chipset. In software, we optimize some protocols in order to save energy, such as MAC, routing protocol, etc. Experiment shows that this prototype can reduce energy consumption of networks and sensor nodes, and prolong the life cycle of wireless sensor networks.