A New Fusion Structure Model on Greenhouse Environment Data and a New Fusion Algorithm of Sunlight

Abstract

In order to meet the requirements of data fusion for greenhouse monitoring system based on wireless sensor network. A kind of multi-sensor data fusion structure model about spatial and temporal fusion is proposed. According to the greenhouse environment parameters such as temperature, humidity, CO2 change slowly, while the parameter of sunlight changes rapidly, this paper proposes that it respectively uses temporal fusion model and dual prediction strategy fusion to fuse these two kinds of parameters to reduce the communication effort in WSN. The sensor nodes are vulnerable to be obscured by the greenhouse structure or crop. It has large impact on sunlight but little impact on other environment parameters. So this paper proposes a new spatial fusion algorithm based on gray incidence analysis to fuse sunlight, and it use power average operator based on support function to fuse other parameters. Algorithms were tested in the greenhouse of Tongji University Jiading Campus. Experimental results obtained demonstrate the efficiency and versatility of the proposed temporal fusion framework in improving the communication savings, the new spatial fusion of sunlight provides an effective way to avoid the fusion effect being affected by obscured nodes' data to acquire more accurate greenhouse sunlight intensity.