基于奇异谱分析的γ能谱降噪算法  

Denoising algorithm of gamma energy spectra based on singular spectrum analysis

作　　者：赵思文 吴怡 王崇杰[1] ZHAO Si-wen;WU Yi;WANG Chong-jie(School of Physics and Electronic Technology,Liaoning Normal University,Dalian 116029,China)

机构地区：[1]辽宁师范大学物理与电子技术学院,辽宁大连116029

出　　处：《物理实验》2021年第9期11-17,22,共8页Physics Experimentation

摘　　要：针对传统频域滤波法无法有效解决γ能谱中噪声频谱与谱成分频谱的重叠问题,本文提出了基于奇异谱分析的γ能谱降噪方法,阐述了奇异谱分析降噪方法的基本原理,给出了降噪算法.通过分析γ能谱的奇异谱特征,给出了最优嵌入维数和γ能谱重构阶数的选取方法.实测60 CoHPGeγ能谱的降噪结果表明:奇异谱分析法可有效分离和消除γ能谱的加性噪声和乘性噪声,从而大幅提高了能谱信噪比.与传统方法相比,该方法算法简单,具有较强的降噪能力,且待定参量少,是有效的γ能谱降噪方法.In order to effectively solve the overlapping problem of noise spectrum and spectral component spectrum inγspectrum,aγenergy spectrum denoising method based on singular spectrum analysis was proposed in this paper.The basic principle of noise reduction method based on singular spectrum analysis was described and the algorithm was given.By analyzing the singular spectrum characteristics of theγenergy spectrum,a method to determine the optimal embedding dimension and the reconstruction order of theγenergy spectrum was given.The noise reduction results of the measured 60 CoHPGeγenergy spectrum showed that the singular spectrum analysis method could effectively separate and eliminate the additive noise and multiplicative noise of theγenergy spectrum,and greatly improve the signal-to-noise ratio of energy spectrum.Compared with the traditional method,this method was simple and had strong reduction ability,less undetermined parameters,and could be an effective method forγspectrum noise reduction.

关 键 词：Γ能谱 噪声去除 奇异值分解 线性滤波 同态滤波 

分 类 号：TL817.2[核科学技术—核技术及应用] TP274[自动化与计算机技术—检测技术与自动化装置]