Research On The Coverage Of Autonomous And Controlled Mobile Sensor Networks

The mobile sensor network can detect the targets in the given zone.The sensor nodes are equipped with sensors such as vibration,sound,magnetic,infrared,and image sensors for real-time detection of human,wheeled vehicles,tracked vehicles and even low-altitude helicopters.Moreover,the sensor node transmits the target to the terminal through the network,and provides the target information.The sensor nodes can carry out dynamic reconnaissance in the case of ensuring network connectivity(nodes controlled move based on mobile model),can also be pre-arranged in the area,with node movement to ensure that the perception of the global coverage(nodes can plan move path autonomously).This paper focuses on the connectivity coverage of the controlled mobile sensor networks and the perceptual coverage of autonomous mobile sensor networks.The main work and innovations are as follows:(1)For the controlled mobile sensor network,this paper proposes the minimum transmission distance of the random waypoint model,which is based on the theory of asymptotic spatial distribution.Firstly,it is proved that the minimum transmission distance is equal to the length of the longest edge in the minimum spanning tree.It is further found that the minimum distance can be obtained by only computing the minimum value of the asymptotic spatial distribution.And then find the minimum of asymptotic spatial distribution is a constant depended on pause-time and velocity.Finally,the relationship between the minimum transmission distance and the number of nodes is confimed.(2)For the autonomous mobile sensor network,this paper proposes a new algorithm,which is based on particle balance and the classical virtual force algorithm.The algorithm introduces the density factor(defined as the ratio of the local density to the expected density)to speed up the process of the node layout,and uses both the oscillation check and the steady state check to ensure the convergence of the nodes.In order to improve the performance of the algorithm in the heterogeneous network,the concept of the neighbor area in the algorithm is extended from the twice of perceptive area to the largest joint range of sensing,which makes the node far away from the neighbor node followed by maintaining the minimum communication links within the neighbor node.(3)To address the coverage control problem of the autonomous mobile sensor network,this paper combines the merits of both the clustering-based method and the point-to-point mode,and proposes an adaptive coverage control algorithm based on mode decision.Each node independently chooses one mode according to the local environment information,including the local node density and residual energy level.If the local density is close to the expected density,the node selects the clustering mode.For the area that already have a reasonable density level,the internal nodes do not need to move too much to improve the coverage and the node uniformity,and their movement depends on the residual energy of the relative neighbor nodes.(4)For the autonomous mobile sensor network,this paper applies the Voronoi diagram in an energy-efficient manner to address the problem of the node layout,and proposes a new coverage control algorithm based on the optimal energy utility point.The utility of the node indicates whether the node can perceive its validity in the effective area during its life cycle,and is used to estimate the coverage of energy-efficient network which each node can get.The node moves from the current position to the different node location,which generates the different energy consumption.And the final node topology may affect the communication energy consumption.Therefore,the optimal energy utility point is determined by the utility gains.This algorithm searches the optimal energy utility point in a distributed manner until the overlap between the master regions and sensing regions in the local Voronoi diagram reaches a preset threshold.(5)This paper comprehensively compares three coverage control algorithms of the autonomous mobile sensor network,and makes three valuable conclusions.Firstly,none of method can get the best value of each metric in all scenarios.In practical application,the appropriate algorithm should be selected on the specific demand.Secondly,in general,the pattern decision-based algorithm is easier to achieve the performance of the network coverage.Thirdly,both the desired energy and the distance of movement in the pattern decision-based algorithm are almost constant in different densities of nodes.That means the energy consumption of this algorithm is not sensitive to the density of network.