多传感器数据融合技术应用研究

当前多传感器数据融合技术中，多是基于支持度矩阵的最优加权多传感器融合方法，但在支持度矩阵的计算中，相对距离的选取多是基于时域峰值，时域峰值在测量中存在较大干扰和不确定性，因此通过时域峰值不能很好地确定两个传感器之间的相互支持程度．本文提出了一种基于互相关函数的支持度矩阵计算方法，然后对支持度高的传感器进行最优加权融合．通过仿真表明：此种方法无需传感器的任何先验知识，能够客观显示各传感器的可靠程度，比其他方法具有更高的融合精度．

Application of Multi-sensor Fusion in Data Processing

In present, most of multi - sensor data fusion technologies are based on supporting matrix and optiinal weight. But in the calculation of supporting matrix, the relative distance is mainly based on the peak value in time domain. There is interference and uncertainty in the measurement of the peak value, so the support degree of two sensors cannot be well determined through the peak value. This paper puts forward a method for calculating the supporting matrix based on cross-correlation function. Then we calculate the fusion of sensors with good support degree in SNR. Experimental analysis shows that this method does not need any prior knowledge of sensors, can show the reliability degree of each sensor objectively, and has high fusion accuracy.