Multi-sensor data fusion for low power transmission of wireless sensor network in a greenhouse
The Wireless Sensor Network (WSN) is widely used for the acquisition of distributed greenhouse micro-climate data, such as temperature, humidity, CO2 concentration and solar radiation. In this paper the communication reliability of a WSN was evaluated based on experimental data within the greenhouse. In order to reduce redundant and invalid data to achieve a low power transmission, this paper proposes two kinds of multi-sensor data fusion scheme depending on the characteristics of the parameters. For parameters that change slowly (such as temperature, humidity, and CO2), a time fusion algorithm which can detect the time consistency of data is used to reduce the amount of data transmitted. The spatial fusion based on the supporting vector is used to give the final decision variable for the control system. For rapidly-changing parameters such as solar radiation, this paper proposes the Dual Prediction Scheme (DPS) based on an adaptive model selection to reduce the amount of data transmitted, with a spatial fusion algorithm based on GIA (Grey Incidence Analysis) for the aggregator. The data fusion schemes were implemented and tested in a greenhouse at the Tongji University Jiading Campus. The implementation results showed that the data fusion schemes can guarantee the sensor precision, and also significantly reduce the amount of data transmitted, which means prolonging the lifetime of the network.
Wei, R., Xu, L. and Wang, X. (2017). Multi-sensor data fusion for low power transmission of wireless sensor network in a greenhouse. Acta Hortic. 1170, 201-208
time fusion, dual prediction scheme, grey incidence analysis