基于感知域测量值的自适应压缩采样方法  

An Adaptive Compressed Sampling Method Based on Sensing Domain Measurement

作　　者：芮国胜 崔田田 田文飚 RUI Guosheng;CUI Tiantian;TIAN Wenbiao(Key Laboratory of Signal and Information Processing in Shandong Province,Naval Aeronautical University,Yantai 264001,China)

机构地区：[1]海军航空大学信号与信息处理山东省重点实验室,山东烟台264001

出　　处：《电讯技术》2023年第10期1553-1559,共7页Telecommunication Engineering

基　　金：国家自然科学基金资助项目(41606117,41476089,61671016)。

摘　　要：现有的自适应采样方法需要在采集端获得原始信号,这在实际的蒸发波导相关气象要素压缩感知过程中不易实现。针对这一问题,提出了一种以测量值数据特征为依据的采样率自适应设定方法。该方法以能量表征各信号块的重要性,以观测值能量近似估计原信号块能量,根据感知域中测量值的能量变化自适应地为各块设定采样数目。理论分析和实验表明,该方法的重构性能优于传统自适应方法,相同采样率下可获得更高的重构信噪比。The existing adaptive sampling methods require original signal at the collector,which is difficult to realize in actual compressed sensing process of evaporation duct related meteorological elements.To solve the problem,an adaptive sampling rate setting method is proposed which is based on the data characteristics of measured value.The new method uses energy to represent the importance of each signal block,and uses the energy of the measured value to approximately estimate the energy of the original signal block.It can adaptively set the sampling number for each block according to the energy change of the measured value in the sensing domain.Theoretical analysis and experiments show that the new method has better reconstruction performance than the traditional adaptive methods,and can obtain higher reconstruction signal-to-noise ratio at the same sampling rate.

关 键 词：蒸发波导 自适应压缩采样 压缩感知 感知域测量值 

分 类 号：TN911.72[电子电信—通信与信息系统]