Research And Design On Prediction-Based Algorithm For Event Detection

Wireless Sensor Network is composed of series of sensor nodes deployed in the concerned region. Most of the nodes are limited by power and processing ability. The nodes form a self-organizing network by communicating with each other. And the nodes transmit the data to the sink node or base station. Event detection is being paid more and more attention as an important application of Wireless Sensor Networks. The event detection algorithm should suit for the realistic scenarios and should solve some problems, such as low-power and high fault tolerance. At the same time, the routing algorithm also should think over the same problems. In the paper, we propose a prediction-based event detection algorithm and a corresponding routing algorithm.Firstly, we proposed a prediction-base event region detection algorithm. In the paper, we consider that the event has the spatial correlation and time correlation. However, the nodes error doesn’t have the correlation. We use the improved gray model as the prediction model. The sensor nodes collect data periodically. And then we use a pre-set filter to filter out the useful data. At last we input the data to the prediction algorithm. When an event occurs, the nodes around the nodes in the event region will also show the data trend due to the spatial correlation. We consider only the nodes whose half of neighbor nodes has the same data trend with them are normal. When the nodes are determined to be the normal nodes and they show the data trends when an event occurs, we can get the event region.Meanwhile, we also studied the link stability routing algorithm suitable for Wireless Sensor Networks. We use the transmission power of the sensor nodes as the evaluation criterion. The link stability model considers the neighbor nodes of a sensor node as a study of the direct link stability and the neighbor of the neighbor nodes as a study of the indirect link stability. Finally, the link stability model is applied to the multicast tree establishment and maintenance. The routing model takes full advantage of the proposed event region detection algorithm, when the algorithm detects an error node, the node is masked to look for a line with the best link stability. In order to verify the performance of the proposed event detection algorithm, we use MATLAB and NS2to complete the data simulation.First.we build the background of the event detection, i.e. a large number of sensor nodes are deployed randomly in a set area, and these sensor nodes have a certain failure rate, which set the percentage of faulty nodes can test the fault tolerance of the algorithm. And then event detection algorithmstarts running in the predetermined background,after the detection, comparing the detection result with the true situation, we can find the event detection accuracy and false positive rate, which confirms the higher performance of the proposed detection algorithms.And we use the node received packet rate and the amounts of the nodes received the packet to verify the applicability of the routing algorithm.