Research On Three Dimensional Self-Deployment Algorithms In Mobile Sensor Networks

The deployment and coverage of wireless sensor networks (WSNs), as a key issue in the research of WSNs, is relatively difficult and limited because of many bottlenecks. Currently, the research of deployment algorithms mostly focuses on the two-dimensional centralized deployment algorithms. However, few attentions are paid to design the three-dimensional deployment algorithms, let alone the work on the three-dimensional distributed deployment algorithms. This leads to the fact that it is difficult to find a theoretical model which can match the deployment of WSNs in most application areas. For the current algorithms for nodes deployment in WSNs, is hardly sutiable to the exploratory probing into unknown spaces such as the ocean, atmosphere and cosmic space can not well achieved.In this thesis, we first expand the perceptual model for wireless sensor nodes from two-dimensional to three-dimensional, and then propose a three-dimensional self-deployment algorithm for mobile sensor networks in the three-dimensional space with obstacles, based on the three-dimensional perceptual model and mobility of nodes. In particular, this thesis theroretically analyzes the optimal coverage of the three-dimensional perceptual model for wireless sensor nodes in three-dimensional space and the corresponding calculation method for the largest effective coverage. Then, a three-dimensional nodes clustering algorithm is proposed by combining the traditional Lowest ID Clustering Algorithm and the three-dimensional perceptual model. Further, a virtual force algorithm for three-dimensional space based on the three-dimensional perceptual model is proposed by applying the idea of virtual force to the display of the mutual force between nodes and the force between nodes and obstacles. In addition, an approach for diffusion by nodes is found through the research on density control strategy. By doing so, we can synchronize the deployment through the density control on wireless sensor nodes and balance the density through clustering in different areas, so that spontaneous diffusion from high density area to low density area by wireless sensor nodes is enabled and the distributed self-deployment achieved.The simulation experiment on performance of this algorithm indicates that in a three dimensional space with obstacles, the initial random placing of nodes within a small area can rapidly diffuse and eventually cover the entire space evenly, reaching a relatively considerable coverage. This thesis provides some theories and methods for the self-deployment of WSNs nodes in the three-dimensional space.