Energy-saving Strategies For Wireless Sensor Networks

With the technology advances in networking and electro-mechanical systems, it is now practical and relatively inexpensive to deploy a group of networked robots for autonomous sensing and data collection in a broad area. These robots sense the environment with onboard sensors and share sensory information through an ad-hoc wireless network. The vehicle group cooperatively performs tasks as a wireless sensor network, which adapts its configuration and coordination behaviors in response to the sensed environment.Wireless sensor networks may find myriad of civilian and military applications, such as battlefield surveillance, search and recovery operations, manipulation in hazardous environments, environment monitoring, target tracking and remote sensing.Generally, sensors in wireless sensor networks battery-powered, their energy is severely limited. But the replacement of the battery is difficult or even impossible and the survival time of wireless sensor networks have to last several months or even several years. Therefore, saving battery energy as far as possible under the premise of quality of service is the core issue and the focus.The objective of this paper is to develop and experimentally verify methodologies for reducing energy-cost of wireless sensor networks. The proposed project investigates several fundamental problems associated with wireless sensor networks and their energy-cost.Based on graph theory and robot kinematics, a dynamic model for the distributed cooperation and multi-hop communication of wireless sensor networks has been proposed. Delaunay triangulation and Voronoi diagram are introduced to model the geometrical relationship between neighboring mobile robots, where the motion of a robot is only related to its immediate one-hop neighbors and its environment. The distributed dynamic model provide a paradigm for the development of a variety of sensor network tasks, such as self-deployment for maximizing coverage area, data routing for querying the network, information sharing between neighboring sensors.Three kinds of self-deployment algorithms for the connectivity and coverage of sensor networks have been developed. The first method improves the self-deployment of sensor networks by means of PSO algorithm. Satisfied performances in terms of coverage uniformity, coverage speed and link reliability have been obtained due to optimized capacity of PSO for multiple targets. In the second method, virtual potential field method is combined with the particle swarm optimization algorithm develop distributed autonomous deployment algorithms for wireless sensor networks. In virtual potential field method, attractive and repulsive forces control the self-deployment process of mobile sensor network together. The values of their coefficients are relative to spreading out effect of the networks. The coefficients are optimized by means of improved particle swarm optimization algorithm. With the optimization results, virtual forces are calculated and sensors are deployed. The simulation results show that this method makes sensors deployment fast and with high coverage percentage. The third method applies the market competition law to connectivity and coverage of wireless sensor networks. In this algorithm, the sensor nodes in the network were seen as enterprises in economic activities, the interested areas were seen as resources, network configurations were seen as market competitions. Take advantage of this algorithm, the amount of computation, mobile distance and information complexity of every sensor can be reduced. The efficiency of energy-saving has been boost indirectly in these projects. Experimental results show that this method is effective.In order to solve multi-hop forwarding problems in wireless sensor networks, a necessary condition about reducing energy consume by means of relay communications has been deduced from Friis free-space equation, and an energy efficient unicast routing algorithm has been proposed. This algorithm adopt the following steps: the interested network has been prune first, only the relay sensors which can reduce path loss remained, the pruned network is seen as a graph, conferred to each hop a value according to its energy consumption and find the least cost path. Simulation results indicate the algorithm achieves satisfactory balance between saving energy and complexity, and show the superiority and potential of pruning in wireless sensor networks.An energy-saving strategy based on directional antenna has been proposed to reduce energy consumption of RF module in wireless sensor networks. Taking advantage of directional antennas with high gains and low side-lobes, in this method, the energy consumption from path loss, collisions and overhearing have been reduced greatly and the efficiency of energy can be improved greatly.Based on Delaunay triangulation and Voronoi diagram, an information fusions method has been proposed to reduce the energy-cost of data transmission in wireless sensor networks. With the energy consumption model proposed in this paper, the energy-saving reasons of this method have been discussed. Simulation results show information fusions in wireless sensor networks are of good effect of energy-saving.Finally, we summarize our achievements and limitations, and then make an expectation on future research of wireless sensor networks'energy-saving strategies at the last part of the thesis.