Research On Optimization Of Energy Consumption Of WSN

Wireless Sensor Network (WSN) is a kind of wireless ad hoc network. It is a set of sensor technology, integrated circuit technology, wireless communication technology and micro electro mechanical system. As a new way of information access and process, wireless sensor network has an advantage of high accuracy, flexibility, reliability and convenience, compared with traditional sensors and wireless network. So it is more suitable for monitoring tasks of poor environment which is an unattended or difficult for human to reach.Wireless sensor network has many advantages on the field of deep-sea oil monitoring applications. However, the limitation of computing and storage capability of sensor nodes is an important factor, affecting the performance of wireless sensor network. How to complete monitoring tasks with minimum of energy consumption and try to extend lifecycle of the network is an important issue of the current field of wireless sensor network. In order to address energy consumption issues of wireless sensor network, this paper puts forward to optimizing the energy consumption of the network at both node level and network level and propose the corresponding improvement algorithm, which is also the main work and innovations of this paper.At the node perspective, this paper processes the data collected from the sensor nodes using the method of generating synopsis data and presents a dynamic weighted random sampling algorithm (DWRS) on time-based sliding window over data stream to extend lifecycle of the whole network. Under the premise of ensuring important data are not lost, DWRS manages to reduce the amount of data transmission and reduce energy consumption used for communication and extend lifecycle of the whole network.At the network perspective, this paper raises a multiple cluster-heads routing protocol (MCHRP) for wireless sensor networks from the perspective of load balancing on the basis of studying LEACH protocol. With MCHRP protocol, factors of node's energy, geographic location and number of elected as cluster-heads are all considered. For the first time, MCHRP protocol presents the concept of policy function to elect cluster-heads more reasonably and make the energy consumption more balanced to extend lifecycle by using multiple cluster-heads mechanism instead of only one cluster-head in one cluster.Simulation results show that, DWRS algorithm is effective in reducing data traffic at the node level and saving a lot of energy consumption. At the same time, MCHRP algorithm makes energy consumption more balanced at the network level. In a word, DWRS algorithm and MCHRP algorithm optimize energy consumption at different angles, achieving the goal of extending network life cycle and optimizing performance of the network.