小波包分解
小波
第二代小波变换
哈尔小波转换
平稳小波变换
离散小波变换
小波变换
吊装方案
计算机科学
多贝西小波
数学
级联算法
语音识别
算法
模式识别(心理学)
人工智能
作者
M. Y. Gokhale,Daljeet Kaur Khanduja
出处
期刊:Int'l J. of Communications, Network and System Sciences
[Scientific Research Publishing, Inc.]
日期:2010-01-01
卷期号:03 (03): 321-329
被引量:98
标识
DOI:10.4236/ijcns.2010.33041
摘要
This paper explains a study conducted based on wavelet packet transform techniques. In this paper the key idea underlying the construction of wavelet packet analysis (WPA) with various wavelet basis sets is elaborated. Since wavelet packet decomposition can provide more precise frequency resolution than wavelet decomposition the implementation of one dimensional wavelet packet transform and their usefulness in time signal analysis and synthesis is illustrated. A mother or basis wavelet is first chosen for five wavelet filter families such as Haar, Daubechies (Db4), Coiflet, Symlet and dmey. The signal is then decomposed to a set of scaled and translated versions of the mother wavelet also known as time and frequency parameters. Analysis and synthesis of the time signal is performed around 8 seconds to 25 seconds. This was conducted to determine the effect of the choice of mother wavelet on the time signals. Results are also prepared for the comparison of the signal at each decomposition level. The physical changes that are occurred during each decomposition level can be observed from the results. The results show that wavelet filter with WPA are useful for analysis and synthesis purpose. In terms of signal quality and the time required for the analysis and synthesis, the Haar wavelet has been seen to be the best mother wavelet. This is taken from the analysis of the signal to noise ratio (SNR) value which is around 300 dB to 315 dB for the four decomposition levels.
科研通智能强力驱动
Strongly Powered by AbleSci AI