The purpose of the laboratory practical was to gain knowledge on the basics of MATLAB and its application to Statistical Signal Processing. The experiment was divided into two parts; the first part was on basics of MATLAB and its application to signal processing problems, the second part was on Speech processing using MATLAB. In the first chapter, after getting acquainted with basic matlab programming concepts, we proceeded to generating noise using the rand() and randn() functions. The concept of linear estimation was modelled to estimate the message in a signal corrupted by noise and the designed estimators were compared and justified to satisfy the Cramer-Rao Lower Bound theorem. In addition, a non linear estimation techniques; Least-Square criterion was developed to estimate the variables in a sum of sinusoidal signal corrupted by noise. The second chapter covered parameter estimation of an AR process. A sound signal was analyzed and a sample of the sound signal was synthesized and compared with the recorded sound signal.
To get a good understanding of the codes, you need to check out the practical report added as a pdf file