摘要
距离特征量反映了目标距离变化规律,该观测量可由基于LOFAR谱图的距离特征量提取方法得到。为解决单一频段提取的距离特征量精度不高的问题,本文基于最优加权平均法,提出了多频段距离特征量值提取技术。针对该方法在实际应用中无法准确得到距离特征量解算值误差的标准差,提出了一种对方差进行实时估计的动态加权融合方法。试验数据处理结果表明,融合后精度明显提高。
The distance of an object can be described by the distance characteristic feature. A new method for extracting distance characteristic feature based on Low-Frequency Array （LOFAR） spectrum has been presented. In view of low precision of distance characteristic feature extracted from single frequency band, an optimal weighted average method is applied to the extraction process in this paper. Since the variance of distance characteristic feature can not be accurately obtained in the practical application, a weighted fusion algorithm is given which estimates the variance in the real time and allocating weights dynamically based on optimal weight allocation principle. The experimental results show that the precision of distance characteristic feature is one time higher by applying fusion.
出处
《应用声学》
CSCD
北大核心
2012年第5期372-378,共7页
Journal of Applied Acoustics
基金
"十二五"预研基金项目(No.51303070407)
水下测控技术重点实验室资助项目(No.613122)
关键词
距离特征量
最优加权平均
动态加权
融合
Distance characteristic feature, Optimal weighted average, Dynamically weighting, Fusion
