Research On Coverage In Wireless Multimedia Sensor Networks Based On Voronoi Diagram

With the development of microcomputer controling technique and novel sensor technology,we produce highly intelligent sensor node that can perceive the environmental changing,process data and communicate information with other nodes. Especially, with the rapid development of micro-motor technology,original sensor model that had fixing direction and had limited perceptual angle gradually develop into novel sensor model that can rotate and perceive omnidirectional environment. Such sensors greatly make up for the original sensor type and numerous novel nodes are used in road monitoring,environmental monitoring and health car. For the coverage problem of Wireless Multimedia Sensor Networks(WMSNs), the paper design reasonable control algorithm and complete network coverage task.Firstly, for researching on coverage of non-rotatable nodes, we apply deterministic deployment strategy to deploying nodes by voronoi diagram. We often use this model into real-time monitoring network. Because voronoi diagram has great characteristics of division of regional. So, this paper presents a coverage strategy based on voronoi diagram in WMSNs. We can find the coordinates of the new nodes by voronoi graph and initialize a orientation angle for new nodes. Then, we can calculate the centroid coordinates of its node and adjust the direction of nodes.Secondly, the node which is rotatable model often is used to some specific monitoring environments that don't have requirements for real-time networks.But original covering control strategies cannot be directly applied to this model.So, we have to consider the characteristics of rotatable node to achieve the design of algorithms. First of all, we can deploy a small number of nodes of fan-shape model in the target area. Then, nodes start to turn around and its monitoring range become omnidirectional model. We can achieve the voronoi diagram according to coordinate values of nodes. Then, we need to look for new coordinate values according to voronoi graph and deploy this sensing model.Above all, we have to delete certain nodes that its overlapping is large. Inaddition, we need to adjust node coordinate values through virtual force slightly.At last, the simulation test for algorithm is done in MATLAB. Comparing with other original algorithms, we verify the feasibility of algorithm in this paper.In the last chapter of this thesis, we summarize the research content and elaborate new research fields.