This demo, developed using the Processing programming language (similar to Java), shows a system identification task in which 'n' least-mean-squares adaptive filters (LMS-AFs), with different step sizes, try to estimate the plant coefficients. The 2D coordinates (x,y) of the point named 'Target' are collected in the so-called optimum vector wo = [x, y]', which is the system to be estimated. The interface allows the user to change parameters like measurement noise and random-walk noise. It also provides two types of learning curves for each one of the adaptive filters: mean-square error (MSE) and mean-square deviation (MSD). Source code and cross-platform applications:

This is a simple Python implementation of the Lempel-Ziv algorithm (data compression algorithm). It was developed during an Information Theory course in the first year of my doctorate. Source code and windows executable: