一种基于数据预处理和卡尔曼滤波的 温室监测数据融合算法

温室具有空间大、无线传感器节点易受到干扰等特点,节点采集的数据波动性较大且易出现丢失现象.为了提高温室监测无线传感网的可靠性和数据融合的精度,提出了一种基于数据预处理和卡尔曼滤波的无线传感器网络数据融合算法.经过对各传感器数据进行预处理和卡尔曼滤波估计,再将数据发送到簇头节点进行基于状态补偿策略的加权数据融合.通过对温室湿度数据进行仿真,结果表明:数据预处理能明显减小数据波动,大幅减少网络数据传输量和能耗,提高抗干扰能力.另外,针对温室无线传感器网络容易出现丢包的现象,基于状态补偿策略的加权数据融合算法可以明显提高在数据丢包情况下的融合精度.

A Data Fusion Algorithm on Data Preprocessing and Kalman Filter for Greenhouse Environment Monitor

The greenhouse has the larger space,and the wireless nodes are vulnerable to the interference from the environment. The data collected by nodes are more volatile and many interference factors easily lead to packet loss. In order to enhance the reliability of wireless sensor neworks for the greenhouse monitoring,and improve the preci-sion of data fusion,a data fusion algorithm on data preprocessing and Kalman filter is proposed. Firstly,the data pre-processing method and Kalman filter is utilized to decrease the influence of abnormal data,then these data are sent to the cluster head and fused on the weighted data fusion algorithm with the state compensation strategy. The simula-tion is conducted on the greenhouse humidity, which shows that data preprocessing can significantly reduce data fluctuations, the amount of data transmission and the network energy consumption while improve the anti-interference ability of wirless networks. In addition,the weighted data fusion algorithm based on state compensation strategy can also significantly improve the fusion accuracy in the case of packet loss.