An Energy Balanced Routing Protocol Based On Data Prediction In WSN

The wireless sensor networks integrate microelectronic technology, wireless communication technology, embedded computing technology, modern network and distributed information processing technology and so on, which have become an import issue in wireless communication field. Wireless sensor networks can coordinate the real-time monitoring, sensing and collecting information of the entire network coverage area in different environment and monitoring objects, and then processing the information and sending through the way of wireless communication, and transmitting data to sink based on self-organizing multi-hop mode. Wireless sensor network has its own characteristics. First, the energy of node is limited and difficult to maintain and has a huge number. Second, the network is self-organization and is based on data centric routing and has to aggregate data locally. The node energy consumption is the main issue in front of us. Therefore, designing and developing efficient communication and routing protocol has becomes the most import research topics in the field of wireless sensor networks.On the basis of studying the LEACH protocol, this paper proposes two new network communication protocols:one is DACP protocol that a data transfer protocol which is based on data prediction and aggregation in wireless networks, the other is SEBRP protocol that a clustering routing protocol based on sink node and energy balance.The Energy consumption of LEACH protocol is large and part of the network nodes will be dead earlier. So we propose an energy-efficient data aggregation transfer protocol based on clustering and data prediction called DACP. Firstly, sensor nodes send messages to sink node, and then sink node divides entire networks into many clusters and elects cluster head nodes for each cluster. In the prediction phase sensor member nodes receive prediction data and compare it with sensed data to decide whether send it or not. In the data aggregation phase cluster head nodes aggregate sensed data received from cluster member nodes and its data to decide to send it to sink node or not according to receiving predicted data. Simulation experiment shows that it is effective to reduce data transmission and improve the efficiency of data aggregation by data prediction.The energy consumption is not balance and many nodes will dead earlier. So we propose the MEBRP protocol. All nodes can consume energy balanced as much as possible to prolong the life of nodes and improve working performance of the network. Sink node send the information of itself position and its neighbor node to every node, and then clusters can choose multiple cluster nodes for the next hop dynamically after the entire network has clustered. We can create many disjoint paths to the next cluster and we can split a data packet into multiple packets and send them separately. Therefore, the energy consumption of nodes will be shared by multiple paths and reduce the consumption of clusters as much as possible to maintain the energy balance of the entire network.We adopt the MATLAB software and simulations results demonstrate that the proposed protocol effectively improve the survival time of the entire networks. DACP protocol improves45%performance and MEBRP protocol improves35%performance.