Research On Data Aggregation In Event-driven Based Wireless Sensor Networks

Wireless sensor networks are made up of many cheap sensor nodes deployed in the area. Those sensor nodes communicate with each other which form a self-organization network system, and collaborate to collect and process the information of the environment or the perceived object, then send these information to the user. However, the energy that the sensor nodes own is quite limited. To decrease the energy consumption, it is better to make local process first instead of sending the data immediately after gathering the data. Data aggregation can reduce the data transmission between the nodes, and thus save the energy and extend the life of the network. In this study, data aggregation in event-driven based wireless sensor network is discussed.In this paper, the feature of the event-driven based wireless sensor networks is analyzed, as well as the traditional data aggregation algorithms for this kind of networks. Through the analysis of these algorithms, a new algorithm named TBCNS is designed on the foundation of the CNS algorithm. First discuss the advantages of this new algorithm, and then introduce the detail of the implementation of the TBCNS algorithm. TBCNS algorithm can decrease the time in the election of the aggregation node effectively. Use the specific data chrematistics of the event-driven based networks, ACS (aggregation in the center source) algorithm is proposed. In this algorithm, the node that collect the biggest feature data is elected as the aggregation node, and build the aggregation tree by only twice broadcasts.At last, two new algorithms are compared with traditional algorithms, and use MATLAB to simulate the algorithms. The results show that these two proposed algorithms have great advantages in energy consumption and time delay.