论文部分内容阅读
本系列文章的工作是在舰船噪声谱图的基础上,利用模糊神经网络对舰船进行分类识别。本文是系列文章的第三篇,研究提取双重频率功率谱中调制信息的方法及建立双重谱和平均功率谱模板图。对双重谱,在每个频道中利用最小二乘法消除趋势项,对高频调制作适当补偿。包络线谱的强弱用谱线低于零频谱强度的调制深度和诺线跳出基线的相对高度来表示,并通过隶属函数转换为模糊度量。双重谱模板图记忆了稳定出现的调制线谱和相对应的调制强度。平均功率谱的模板图记忆了多个典型样本的谱均值和对应的标准离差。
The work of this series of articles is based on the ship noise spectrum, the use of fuzzy neural network classification of the ship classification. This article is the third in a series of articles to study methods for extracting modulation information from dual-frequency power spectra and to create dual-spectrum and average power spectrum templates. For the double spectrum, the least squares method is used in each channel to eliminate the trend term, and the high frequency modulation is appropriately compensated. The intensity of the envelope spectrum is expressed as the modulation depth below which the spectral line falls below zero spectral intensity and the relative height above which the null line jumps out of the baseline and is transformed into a fuzzy measure by a membership function. The double spectrum template memorizes the steady modulated line spectrum and the corresponding modulation intensity. The template of the average power spectrum memorizes the spectral mean of many typical samples and the corresponding standard deviation.