Research On Data Aggregation Algorithm For Wireless Sensor Networks

The wireless sensor network, which is integration of sensor techniques, MEMS techniques and wireless communication techniques, is an innovative technique of information acquisition and processing. It can sense, collect and process information of monitored object in the covered place. Wireless sensor network has a lot of unique features such as random deployment, self-organization, and adaptation and so on. These features make it have a wide application future and high application value in military, medical, industrial, environmental monitoring and other commercial area; it has become one of the active research branches of computer science. A few achievements have been acquired, but on some key techniques, there are also a lot of problems in need of resolution. This thesis makes research on how to use aggregation technique to save energy in wireless sensor network.First of all, we describe the concept, characteristics and technical challenges of data aggregation for Wireless Sensor Networks,and make an analysis of the current principles, performance of the typical data aggregation algorithms in this thesis. And then two kinds of data aggregation algorithms are proposed, which is An Energy-Aware Adaptive Data Fusion Routing Algorithm for Wireless Sensor Networks(EAAF)and A Novel Framework for Miscellaneous Data Aggregation in Wireless Sensor Networks(MDA).EAAF is one kind of the aggregation algorithms combined with routing. The closer is the distance between the two nodes provided that they are within the communication range, the more is the correlation among sensed data. If the data fusion is earlier operated, the performance of the algorithm will be better. On the basis of the above characteristics, the cost of fusion cost and transmission are two key elements to the minimum cost spanning tree. When the energy gain of data fusion is less than zero, the nodes will receive data by the shortest path tree directly forwarded to the sink node in order to avoid introducing more energy consumption. EAAF not only optimizes over both the transmission and fusion costs, but also adaptively adjusts fusion decisions for sensor nodes as to whether fusion shall be performed. EAAF achieves near optimal solutions under various networking conditions. The miscellaneous data aggregation is an algorithm based on data timeliness. The applied quality of service is the basic requirements for Wireless Sensor Networks; the same environmental parameters have different QoS requirements for Wireless Sensor Networks at different times. According to the applied requirement, the miscellaneous data aggregation scheme is proposed. The theoretical foundation of MDA is based on ADDA and AIDA algorithms. Through the semantic analysis of the timeliness of data, data packets are divided into real-time data packets (RT) and non real-time data packets (NRT) on the basis of timeless by MDA in Application layer. NRT packets wait at a queue until the number of accumulated packets is equal to the maximum aggregation limit. On the other hand, RT packets, which are sensitive to the aggregation latency, are transmitted as soon as possible without any waiting at the queue. In AIDA layer, the aggregation control unit can intercept the current network conditions; The data aggregation degree and aggregation time are decided by the aggregation control unit.