信号量化和离散傅里叶变换的信息熵分析  

Information entropy analysis for signal quantization and discrete Fourier transforms

作　　者：曾金芳 黄佳妹 李超 ZENG Jinfang;HUANG Jiamei;LI Chao(Xiangtan University,College of Physics and Optoelectronic Engineering,Xiangtan 411100,China)

机构地区：[1]湘潭大学物理与光电工程学院,湖南湘潭411100

出　　处：《长江信息通信》2024年第3期64-68,共5页Changjiang Information & Communications

摘　　要：常规的描述方法不满足人们获取信号所携带信息的工作提出的要求,针对信号量化和获取频谱过程存在信息损失的问题,提出从信息域的角度分析量化和离散傅里叶变换的信息量变化,并以信息量来衡量处理的效果。基于信息熵推导离散傅里叶变换的熵表征,通过应用实例结果可知,当采样点数达到一定数值时,信号的输出熵趋于一个恒定值,结合频谱图发现再增加采样点数获取的频谱图几乎毫无变化,信息损失可忽略。实验数据结果具有一般普遍性,为原本依靠经验的工作提供理论支撑,在处理前进行效果预测可以降低实验成本和节约时间,为信号处理提供一种客观的分析方法。The conventional description method does not meet the requirements of people's work to obtain the information carried by the signal,and in view of the problem of information loss in the process of signal quantification and acquisition spectrum,it is proposed to analyze the change of quantification and discrete Fourier transform information from the perspective of information domain,and measure the effect of processing by the amount of information.Based on the entropy representation of the discrete Fourier transform derived from the information cntropy,it can be scen from the application example results that when the number of sampling points reaches a certain valuc,the output cntropy of thc signal tends to a constant value,and combined with the spectrogram,it is found that the spectrogram obtained by increasing the number of sampling points has almost no change,and the information loss is negligible.The experimental data results are generally universal,providing thcoretical support for the original cmpirical work,and the effect prediction before processing can reduce the experimental cost and save time,and provide an objctivc analysis method for signal procssing.

关 键 词：信息熵 互信息 离散傅里叶变换 量化 信号处理 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]