The Problems with Distortion 1.2 2003/12/16 14:41:23 US/Central 2003/12/17 16:00:57.038 US/Central Elizabeth Gregory lizzardg@rice.edu Jason A Buck jbuck@rice.edu Elizabeth Gregory lizzardg@rice.edu Mark S. Husband mhusband@rice.edu Richard G. Baraniuk richb@rice.edu Jason A Buck jbuck@rice.edu distortion nonlinear harmonics Taylor Series In trying to model distortion, a few noise problems and seemingly failed method arise. Let's say you go out to Radio Shack and buy a few cheap cables that let you plug in your slammin' new guitar into your soundcard, just so you can play around with different distortions. You want try playing with MATLAB, using the Taylor Series method's familiar discrete power series: y n 0 N a n x n where x is your original signal, y is your distorted signal, and a n is the constant for each item of the sum. The first problem you notice in playing back your wav file is the noise; after all, those were some really cheap cables! However, from an introductory electrical engineering class and another signals class, you've learned how to filter out noise, so that should be easy to deal with later. You notice the next problem when trying to execute the Taylor Series approximation of the signals. You may set some harmonics' constants to zero, but as you increase different coefficients of higher and higher powers, the only difference is that the music has a lower volume, and the volume of the noise is higher! Why isn't this working?