期刊名称：International Journal of Advanced Research In Computer Science and Software Engineering
印刷版ISSN：2277-6451
电子版ISSN：2277-128X
出版年度：2012
卷号：2
期号：4
出版社：S.S. Mishra
摘要：Nois es presen t in communication channels are distu rbin g and the recovery of th e original signals from the path with out any n oise is very difficul t task . Th is is achie ved by de noisi ng te chni ques that remove n oises from a digital sign al. Many denoisi ng techni ques h ave been proposed for the re moval of noise s from the digital audio sign als . But th e effective ness of th ose techn ique s is le ss. He re MMS E and Harr Wavelet Transforms are used for denois ing. Output S ignal to Noise Ratio (S NR) and Mean S quare Error is calculate d & compared using differen t tech niqu es. Harr wave let provides be tter resul t as compared to M MS E Trans form. An alysi s is done on noisy s peech sign al corrupted by wh ite n oise at 0 ¨C 18 dB sign al to noise rati o levels. Analysis is don e on noisy spee ch si gnal corrupted by white guassi an n oise S ignal to Noise Rati o levels at 0-18 db. . O utput signal to noi se ratio and mean s quare error is calculate d & compared usin g MMS Eand, HAAR wave let. S imul ation & re sults are performe d in MATLAB 7.10.0 (R2010a).
