改进型分批估计与自适应加权融合方法研究

针对智能家居火灾监测中数据准确性低冗余大等问题,提出了一种量测数据预处理与改进型分批估计自适应加权数据融合相结合的算法。首先,该算法根据格罗贝斯准则对单个传感器测量数据序列进行一致性检验,从而剔除疏失误差数据;其次,考虑传感器受恶劣因素影响导致量测波动较大,引入环境因子并改进分批估计算法计算单个传感器最优监测值;最后,针对不同方位多传感器误差分布不均匀的特点,提出了根据权值最优分配原则实现自适应加权数据融合。实验结果表明,该算法得到的融合结果误差小,能够有效提高数据准确性,降低冗余量,具有较好的稳定性能。

Research on Improved Batch Estimation and Adaptive Weighted Fusion Method

An algorithm combining measured data preprocessing with a modified batch estimation adaptive weighted fusion is proposed to solve the problems of low data accuracy and high redundancy in smart home fire monitoring.Firstly,it removed the missing error data according to the consistency test of the Grubbs criterion for measured data from current sensor,and then the optimal estimation was calculated based on improved batch estimation by introducing environmental factors.Finally,with the principle of optimal weights,adaptive weighted fusion was implemented to solve the inhomogeneous distribution of errors for sensors employed in different areas.Experiments show that this algorithm can make less error of the fusion result and can effectively improve the data accuracy and reduce the redundancy.The proposed algorithm has better stability.