DSP GUI Demonstrations

DSP GUI demos made during course work to illustrate various signal processing algorithms:

Each of the downloads contains a readme file explaining the purpose behind the demo, the list of files with the demo & how to run the demo. Some of the downloads also contain the presentations made before the demo to illustrate the concepts.

    Calculates the coefficients of an AR (autoregressive) process given the autocorrelation coefficients and the number of poles using Levinson Durbin recursion. The demo also plots the reflection coefficient and the error sequence. 

2. Image Halftoning by Vishal Monga

    This demo lets you design digital halftones by Edge Enhancement error Diffusion. The quality of the halftone is evaluated on the scale of various Image Quality Metrics. The demo also allows one to save the image produced by using the save option.

 3. Modelling Methods by Shailesh Patil

     This demo illustrates three different methods namely Shank, Prony & Pade's method to model a given signal. It also plots the error sequence and the minimum mean square error. The given signal can be easily modified to a different one by changing the ModellinGUI.m.

4. FIR Wiener Filters by Arvind Rao

    Various applications of FIR Wiener filters are demonstrated by the above demo. These applications include filtering, prediction & noise cancellation. The demo uses examples from the course text book (Monson H. Hayes, Statistical Digital Signal Processing and Modeling, John Wiley & Sons Inc. 1996) .

5. Periodogram Estimation by Zukang Shen

    Power spectrum estimation of sinusoids (single & double) and white noise using periodogram method is demonstated. The demo allows user to define the number of white noise sample, the frequency of sinusoids and other parameters.

6. Lattice Filters by Raghu Raj

    In this demo we provide a convenient tool to design lattice filters from rational transfer function coefficients or from pole-zero model, and vice-versa.  The GUI allows the user to visualize the effect of incrementally changing the lattice filter coefficients.

 7. Modified Periodogram by Changyong Shin

     Demonstrates various modified periodogram methods to estimate power spectrum of sinusoids & an AR process. The modification demonstrated are Bartlett, Welch, and Blackman-Tukey. The demo allows user to enter the frequency & amplitude of sinusoid plus the variance of white noise in the case of sinusoids, while for an AR process one can set the AR coefficients.

8. Frequency Estimation by Alex  Tsankov

   The demo estimates the frequency information of an arbitrary sinusoidal signal using the Pisarenko, MUSIC, Eigenvalue, and Minimum Norm methods.  The signal is assumed to consist of a sum of p complex exponentials distorted by additive white noise.

9. Kalman Filters by Farooq Sabir

    This demo illustrates discrete Kalman filtering. The AR system parameters are input to get kalman filtered output.

10. LMS by Huihui Wang

      Various flavors of the LMS algorithm are demonstrated by the demo. These include steepest descent linear prediction, LMS linear prediction and NLMS noise cancellation.

11. Adaptive Identification by Sayantan Choudhury

      In this demo we are trying to identify an unknown system (plant) of the form B(z)/A(z) using the LMS algorithm. The demo uses both Matlab(R) & LabView(R). It outputs the estimated parameters of the system and the error plot.

12. Parametric Spectrum Estimation by Joel Tropp 

      This demo generates 4th order autoregressive processes with arbitrary coefficients. It demonstrates how different parametric spectrum estimation techniques work for the AR process.

Last updated 1/8/2003 

Mail comments to : Robert W. Heath