Optimization Research On Routing And Topology Control For Energy-harvesting Wireless Sensor Networks

The energy harvesting wireless sensor network, composed of many sensor nodes, is a distributed sensing system, where the sensor node is capable of harvesting energy, detecting data, processing data and transferring data. The traditional wireless sensor network takes the limited-capacity micro-batteries as the power source. Due to the complicated environment, it is hard to recharge or replace the battery, which restricts the wide application of the wireless sensor networks. Compared with the traditional wireless sensor network, the lifetime for the EHWSN has been improved greatly. By introducing the energy harvesting technologies and integrating an energy harvesting module into the sensor node, it converts the ambient environmental energy, such as wind, solar, etc, to electrical energy and stores it in the battery to supplement the node. And this ameliorates the node’s energy restriction problem. Under this situation, the energy harvesting wireless sensor network has attracted a lot of attention from academia and industry.Although the energy harvesting technology provides wide prospect for the wireless sensor network’s scalable application, how to utilize the harvested energy efficiently to serve the wireless sensor network faces numerous challenges. The traditional routing and topology control algorithms for the wireless sensor network mainly considers the nodes’ geographical information, real-time energy status, etc. However, these methods meet lots of problems while applied in the energy-harvesting wireless sensor network and the network performance has the space to be improved. Current research on wireless sensor network mainly includes the planar structure and hierarchical structure. Thus, this paper focuses on the routing and topology control algorithms for the planar and hierarchical energy harvesting wireless sensor network. The main objective is to develop efficient routing algorithms and network topology control algorithms for the EHWSN. The main contributions of this paper are as follows:First, for the planar network topology, this paper defines the energy-neutral operation node, the energy-neutral operation path and the energy-neutral operation minimum cost path according to the sensor node’s harvested energy and consumed energy in a future time slot on the basis of energy-neutral operation concept. And analyzes how to find the energy-neutral operation minimum cost path by adjusting the transmission power in a single-source single-sink wireless sensor network from the point of task-driven. Further, the data transmission scheme for multiple sources in a multi-source single-sink wireless sensor network is designed in order to make each node on the transmission paths achieve the energy-neutral operation. Then, this paper calculates the proposed algorithm’s performance in terms of path capacity at different confidence intervals from the perspective of probability theory. Last, the problem of how to obtain the network capacity based on the Ford-Fulkerson algorithm for the single-source and single-sink network is discussed. Also, this paper explores how to transform the multi-source single-sink network into the single-source single-sink network.Second, for the hierarchical topology network, the traditional Low Energy Adaptive Clustering Hierarchy(LEACH) protocol is improved to be suitable for the energy-harvesting wireless sensor network. Considering the uneven clustering characteristics of the LEACH protocol, the cluster node is determined according to the network nodes’ available energy, harvested energy and consumed energy in a future time slot, in order to optimize the clustering nodes in the network. The nodes’ energy conservation problem is mathematically formulated, and the nodes’ energy distribution problem is mapped into a max-min problem which has further been proved to be NP-hard theoretically. Then a polynomial-time algorithm has been proposed and the experiments have been completed to verify our proposed algorithm in terms of average node remaining energy, network lifetime etc.Third, a network Topology Control(TC) algorithm is proposed based on the ordinal potential game theory which is suitable for the planar energy harvesting wireless sensor networks. This paper aims at how to optimize the network topology by efficiently utilizing the harvested energy and adopt the Markov chain model to predict the energy harvesting profile, which is divided into different states. The nodes with strong harvesting capability “protect” the nodes which are less capable of collecting energy from the ambient environment in order to ameliorate the “overflow” problem, which might be incurred by the limited capacity of energy buffers equipped for storing the harvested energy. And through this way, this paper achieves the purpose of balancing the network data transferring load and preventing the node’s premature failure, which would lead to the network’s functional failure. Considering the sensor node’s transmission power and coverage comprehensively, the wireless sensor network’s node set, strategy set and utilization function according to the harvested energy and consumed energy are formulated mathematically, and the existence of a Nash Equilibrium(NE) for the game is proved theoretically. Then, a polynomial time algorithm is proposed to achieve the Nash equilibrium and extensive experiments have been completed to verify the proposed algorithms’ feasibility and effectiveness in terms of nodes’ energy consumption, the price of anarchy, etc.