Order Statistic Filters are a class of non-linear filters whose output is a linear combination of the order statistics of the input. Analysis of these filters shows that these filters have good edge preservation properties and are very suitable for the removal of impulsive noise. In this thesis, algorithms for the computation of the Running Order Statistics (ROS) are implemented in real-time using the TMS320C30 digital signal processor (DSP). A practical application of the realized OSF for the demodulation of a double side band AM signal is also presented. |