| WSN (Wireless Sensor Networks) is a newly emerging network technology with huge development potential. As a carrier or bridge of communicating between objective physical and subjective sensing worlds, WSN is a newly information acquiring and processing technology, and which is also a revolution of information sensing and collection. Data aggregation technology is an important content of WSN network support technology, and a foundation of WSN technology research. Data aggregation is a integrated technology, that the main task is collecting the physical information in the sensor nodes monitoring area, and transmitting the data to the upper nodes in the networks. We have researched the in-networking data aggregation technology with considering the spatial and temporal data characteristics, and proposed calculation models and aggregation solutions for the data aggregation system from data sensing layer to network protocol layer. The main innovation ideas as follows:(1) Based on the most common sensing mode of WSN sensor nodes, we analyze the correlation characteristics of the sensing data; propose a calculation model of sensing data correlation based on nodes spatial positions, which has low computation complexity and the parameters are configurable. The calculation model achieves rapid calculation of data correlation between the sensor nodes in different spatial positions and provides a correlation parameter for a variety of data processing methods, such as network protocol design, route planning, and data compression strategies and so on. The simulation results prove the accuracy and efficiency of the model based on Omni-directional Boolean sensing model. (2) Based on spatial position and temporal interval, we propose system frame and analysis method of aggregated data distortion by using the correlation model between sensing data. Then under the conditions of the maximum distortion which could be tolerated by the network, we derive two optimal network configure parameters, which are the optimal number of sensor nodes in a cluster and the optimal frequency of sensor nodes report to sink node. The sending layer distortion caused by sensor nodes hardware property and sensing mode difference is eliminated in the process of WSN data aggregation. The simulation results show the infection of distortion from correlation model and different network configuration parameters.(3) A systematic and new method of WSN data aggregation by using publish/subscribe mechanism is proposed. And based on the system, we propose data interaction process, information format define, and software frame. By using the publish/subscribe data aggregation method, massive data can be send to multi-users in a fast, accurate, and effective way, and data granularity of sensing data is enhanced, the transmission of useless data is reduced, the energy effectiveness is improved, the users’ acquirement of sensing data is more easily and intelligent in WSN.(4) Based on the optimal network configure parameters from aggregation data distortion analysis, we propose a wireless token-ring MAC protocol which is suitable for publish/subscribe mechanism. This protocol eliminates data collision, achieves in-timing scheduling adjustment and inflexible access control, and improves network service differentiation and energy efficiency. The simulation results prove the performances of energy efficiency, transmission delay and network throughput. |
