Research On Distributed Storage And Query Strategy In 3D Wireless Sensor Networks

With the development of internet of things era, there will be more and more received data in wireless sensor networks, and deployment region of sensor nodes need to be expanded from 2D to 3D. However, data managing and using capacities of data storage and retrieval strategy are limited. So it is necessary to accelerate the processing data capacity and explore the efficient data storage and retrieval strategy in 3D space.In this paper, we consider the research progress of data storage and retrieval algorithms in wireless sensor networks and study the issues, and then the data storage and query algorithm in 3D wireless sensor networks are proposed. The algorithm can decrease the storage space of sensor data, reduce the overhead of transmissions and balance the network load.This paper first computes the boundary surface of sensors in the 3D deployment area and reconstructs the boundary surface by triangulation based on finite element analysis principle. Then the boundary surface is cut along the cutting edge boundary to obtain a curved opened surface. Second, we map curved opened surface to two-dimensional rectangular plane applying the Gaussian curvature. When the nodes on the non-boundary surface carry out the operation of data storage or retrieval, they will send the corresponding request to the nearest node on the boundary surface. For the nodes on the boundary surface, the improved double-ruling-based scheme can be carried out directly on the mapping plane. At last we use the improved double-ruling-based scheme in two-dimensional rectangular plane to achieve data storage and retrieval.Effectiveness and reliability of the improved algorithm are proved in the simulation platform of MATLAB. Simulation experiments are conducted to compare with the performance of the proposed algorithm, GHT algorithm and the original spherical double ruling algorithm in terms of data storage, communication cost and energy-load. The simulation results show that the improved algorithm has a better performance than GHT algorithm and the original spherical double ruling algorithm in the case of saving overhead and balancing load.