Travel Route Planning Of Mobile Sink In Wireless Sensor Networks

In recent years, wireless sensor network with mobile sink(WSN-MS) emerges by introducing mobile sink into wireless sensor network(WSN). The features of WSN-MS mainly include direct data transmission and periodic data collection. The direct data transmission means that sensor nodes transmit data to the mobile sink by single hop. Each round of periodic data collection contains the following steps: the mobile sink starts from a fixed position, and then, accesses all sensor nodes, and at last, returns to the fixed position to prepare for the next round of data collection. Under the comprehensive effect of the two modes, not only the life time of WSN and the energy consumption of sensor nodes, but also the endurance ability of mobile sink, the energy consumption rate between sensor nodes, and the data collection delay per round should be considered. In this thesis, we solve these emerging problems by studying the route planning problem of mobile sink. The main contents of this thesis are as follows.First, the shortest travel route planning for WSN-MS is studied. It is the main method to increase endurance ability of mobile sink by reducing travel route length. However, as previous studies setting the mobile sink to collect data at the exact location of sensor nodes, and not considering the spatial characteristics of wireless transmission range of sensor nodes will lead the programmed travel route longer. In this paper, we will consider the spatial characteristics of wireless transmission range of sensor nodes to plan the shortest travel route. In the study, sensor nodes are regarded as disk regions, thus, the shortest travel route can be achieved by making the mobile sink traverse all disk regions. In this thesis, the shortest path planning problem is formulated as a mixed integer nonlinear programming, and the enumeration and cut algorithm(ECA) and the decomposition algorithm(DA) are proposed. The ECA is a global optimal algorithm, which can obtain the optimal solution, and the DA is a local optimal algorithm, which can obtain the approximate optimal solution. Extensive experiments show that: the ECA can plan out the shortest travel route, however, it is only available in small scale networks due to its high complexity; the DA can plan the approximate shortest route, and it is feasible in large scale network for the low complexity.Second, the energy balance travel route planning for WSN-MS is studied. The study of the shortest travel route planning for WSN-MS suggests that there is tradeoff between energy consumption of sensor nodes and the route length of mobile sink. However, pre-vious studies often ignore this problem. In this thesis, we will consider the relationship between the energy consumption of sensor nodes and the travel route planning of mobile sink, and then try to plan an energy balance travel route, which makes all the sensor nodes have the same energy consumption rate. We expect that sensor nodes with less energy can reserve more residual energy after each round of data collection, thus we formulate the energy balance travel route planning problem as a max-min optimization problem. Due to the strong relationship between the energy consumption and communication radius of sensor nodes, we transform the max-min optimization problem as anchor position adjusting problem and communication radius adjusting problem, and then propose anchor position adjusting algorithm and communication radius adjusting algorithm, respectively. A large number of experiments show that: the travel route of mobile sink will be energy balance after few iterations of data collection; the energy consumption rate of all sensor nodes are basically the same when the travel route becomes energy balance.At last, the minimum delivery latency travel route planning for WSN-MS is studied. From the schema of periodic data collection, we can see that: the delivery latency is large for the mobile sink introduced, and it is not effectively by simply reducing the length of travel route. In this thesis, we introduce the motion data collection mechanism to further reduce the delivery latency, which makes the mobile sink moving and collecting data concurrently. In this thesis, we formulate the minimum delivery latency travel route planning problem as an integer programming problem not only considering the travel route planning of mobile sink but also taking the access schedule of sensor nodes into account, and then propose a substitution algorithm. A large number of experiments show that: the proposed algorithm can effectively calculate the approximate minimum delivery latency travel route of mobile sink; the algorithm can also return the optimal access schedule for sensor node.