Research On Deployment And Topology Control In Wireless Sensor Networks

Along with the expansion of human exploration field and space, wireless sensor networks (WSNs) came into being, which are the integration of micro-electro-mechanical systems (MEMS), wireless communications, embedded computing and distributed information processing technologies. WSNs consisting of a large number of micro-sensor nodes achieve information collection, processing and transmission by self-organization, so they have broad application prospects in many fields. In order to improve network performance and prolong the network lifetime, in this paper, the deployment and topology control techniques are researched according to the dense distribution and limited energy of sensor nodes in WSNs.Firstly, a distributed method based on genetic algorithm (GA) is proposed to optimize the design of WSNs consisting of multi-class, multi-level sensor nodes. The optimization model is divided by using the independence of heterogeneous sensor nodes, and the population is initialized and processed feasibly by priori knowledge, which accelerate the convergence of algorithm. The simulation experiment results verify the effectiveness and convergence of this method.Secondly, a deployment optimization method based on fish-swarm algorithm is proposed for WSNs consisting of many mobile and stationary sensor nodes. The movement of mobile nodes is analogized to the follow and prey action of the artificial fish with the network coverage as objective function, and the virtual force factor is introduced to direct the moving in the process of status updating for artificial fish. Simulation results show that this algorithm improves the network coverage and performance efficiently with the better ability on global searching and adapting.Thirdly, an adaptive clustering topology control algorithm based on node partition is proposed for WSNs with sensor nodes deployed randomly and densely. The load in clusters is balanced by node partition, and the cluster heads are elected by a probability optimized and adjusted by adaptive weight ratio of residual energy and communication distance of nodes. Simulation results show that this algorithm is effective in prolonging the stability period and lifetime, and improving the amount of data transmission for the whole network system.Finally, an energy efficient dynamic sensor nodes selection topology control algorithm based on grid partition is proposed for the sleep scheduling scheme in WSNs with sensor nodes deployed densely. In order to ensure the network coverage, grid partition is carried out for monitoring area. In order to save energy and balance energy consumption, the status of sensor node is determined by the ratio of residual energy and adjusted by the relative distance information. Comprehensive simulation results show that this algorithm prolongs network lifetime significantly taking into account the network coverage.