一种改进的多无线传感器数据分批估计自适应加权融合算法

为解决温室大棚中多无线传感器监测数据融合精度低的问题，提出了一种改进型的分批估计自适应加权融合算法。算法首先对单个无线温度传感节点一段时间内所采集的数据根据容许函数阈值剔除误差较大的数据，然后对该温度传感器的数据进行分批估计得出该节点某一段时间内的最优估计值，以此得到该区域所有无线温度传感节点最优估计值后，依据权值最优分配原则对每组传感器数据进行组内自适应加权融合，从而计算得到该段时间内温室大棚的温度精确值。通过实验验证得出：相对于平均值法与传统的分批估计算法，本算法数据融合易于实现，融合值相对误差值更低，稳健性更好。

An Improved Multiple Wireless Sensor Data Batch Estimation Adaptive Weighted Fusion Algorithm

To solve multiple wireless sensor monitoring data fusion in greenhouses the problem of low precision ,this paper proposes a modified batch estimation adaptive weighted fusion algorithm. Firstly,the algorithm based on the allowable threshold function eliminate a large error data,which collected over by the wireless temperature sensor within a period of time,then,batch estimate these processed data to derive optimal estimation of the value of the node,after this optimal estimate of all wireless temperature sensor nodes in the region,according to the principle of optimal allocation of weights to adaptive weighted fusion Within the group,finally,the algorithm calculates over the period of time the greenhouse temperature exact values. Through the experiment shows:compared with batch estima?tion algorithm and the traditional method of average value,this algorithm is easy to realize data fusion,the fusion value relative error is lower,better robustness.