Research On Mobile Data Collection Technology Based On Compressive Sensing In Wireless Sensor Networks

A main task of wireless sensor network is that sensor nodes sense and collect data collaboratively, and transmit the data to the Sink node for calculation. However, due to the limitations of hardware technology, the sensor node has only a limited amount of energy, communication bandwidth and storage capacity. How to achieve efficient data collection in resource-constrained, data-intensive wireless sensor network is one of the important issues for researchers. In this paper, according to the characteristics of wireless sensor network, the mobile data collection problem in wireless sensor network based on compressed sensing is studied in depth.For energy consumption problem in the process of data collection, a model of mobile data collection is established and a single-hop data collection mechanism based on a mobile data collector is proposed. All the sensor nodes in the network can be covered by planning data collection positions for the collector, which guarantees mobile data collector’s one time collection can collect sensory data in the whole network. In the process of selecting polling-nodes, three heuristic polling-nodes selection algorithms based on the minimum set covering, the neighbor set and the shortest distance are proposed, respectively. The shortest path is solved through ant colony algorithm by utilizing the polling-nodes, and the best collection route of the mobile data collector is obtained. Further, using optimization ability of genetic algorithm, the route planning algorithm based on genetic algorithm and ant colony algorithm is proposed for mobile data collection. Simulation results show that the proposed algorithms can shorten the distance of collection path for mobile data collector and is helpful to prolong the network life cycle.In response to data collection problems in large-scale wireless sensor network, to further improve the efficiency of data collection, a limited multi-hop data forwarding model is presented, and a multi-hop mobile data collection scheme is proposed. By selecting the polling-nodes, data fusion is executed locally and the data is transferred to the mobile data collector. When some sensor nodes belong to these polling-nodes, data collection delay is reduced and energy is saved through limiting the number of hops. Considering the node storage capacity, the limited resources, and the characteristic of high temporal and spatial correlation among the transmission data transmitted, the compressed sensing technology is introduced to reduce the collected data stream, thereby the power consumption is reduced and the network lifetime is prolonged. According to the energy difference during signal reconstruction between adjacent phases, by achieving the sparsity gradually through variable step size, an adaptive matching pursuit algorithm is proposed based on variable step size. A series of simulation results verify the effectiveness of the proposed algorithms.