The Research On Lifetime Of Wireless Sensor Networks

Recent advances in micro-electro-mechanical systems, digital electronics, and wireless communications have led to the emergence of wireless sensor networks (WSN), which consist of a large number of sensing devices each capable of sensing and processing and transmitting environmental information. As the sensor nodes are usually supported by batteries, energy economization is considered as one of the main design principles for WSNs. Extending energy efficiency and network lifetime are important objectives and challenges in wireless sensor networks. In this dissertation, four aspects relating to network-lifetime in wireless sensor networks are studied, which include Pareto optimal, mobile base station, connectivity and coverage and maximizing network lifetime in Large-scale Wireless Sensor Networks(LWSN). The corresponding mathematical models and algorithms are presented, and simulation results and analysis show their effectiveness. The main innovations in this dissertation are outlined as following:(1) An optimal Lifetime Algorithm Based on Pareto optimal in wireless sensor networks is presented. Targeting to solve the problem that the decline of the performance of system which is caused by the failure of nodes in sensor networks, the principle of Pareto optimality in economics is introduced and applied to prolong the network lifetime, which maximizes the nth minimum lifetime subject to the (n-1)th minimum lifetime being maximum,..., subject to the minimum lifetime being maximum, so as to maximize, to a large extent, the lifetime of every node in sensor networks. When there are sensors who use up theirs energy, the others would send their data to base station by the max-min routing strategy which could balance the energy consumption of nodes and prolong the network lifetime.(2) A method based on mobile base station and routing strategy for prolonging lifetime of wireless sensor networks is brought forward. By analyzing the influence that the positions of base station have upon network energy consumption and network lifetime, the best choice strategy of the positions of base station which is the barycenter of nodes in the network is obtained. In order to reduce the performance of the system which is caused by the failure of nodes, the linear programming is used to reduce the difference between sensors'energy and max-min sensors' lifetime. When lapses happend on sensors, the base station would refresh its position and send the new routing for sensors.(3) A dynamic sleeping scheduling algorithm is constructed based on the connectivity and coverage for prolonging the network lifetime in the light of the features of wireless sensor network. Aiming at the random distribution of sensors in scene, these nodes which are energetic and satisfactory for network connectivity and coverage, are selected as working nodes capable of collecting data in turn. Judging by the coverage redundancies of sensors by their sensing neighbor's coverage, and connectivity redundancies of sensors by the number of spanning tree rooting by base station. The linear programming is applied to balance the energy consumption of sensors so as to reduce the bad performance caused by the failure of sensors after turning off the redundant sensors. When lapse happens on certain sensor, the adjacent sleep nodes can be activated to replace the failure node and continue to maintain the normal work of the network.(4) A method for maximizing network lifetime in large-scale wireless sensor networks(LWSN) is presented. By analyzing influence that the amounts of nodes in network have upon communication links, it is concluded that the one-hop neighbor of base station would have the same lifetime when using linear programming max-min sensors' lifetime. Based on that, a graph coloring method is used to divide the node into uncorrelated and saturated subset for the sake of reducing the complexity of routing algorithm. And a clustering method is used to combine the colored subset properly in order to reduce the effect brought by the deployment of nodes randomly distributed. Then, the unions come into being by applying the strategy of linear programming for the routing of the sub-network is the solution of large-scale wireless sensor network.