/line-spectra

MUSIC, Min Norm and ESPRIT for line spectra estimation. Based on the book "Spectral Analysis of Signals", P. Stoica & R. Moses

Primary LanguagePythonMIT LicenseMIT

Signal analysis for line spectra signals

Various algorithms for the line spectra estimation, such as

  • MUltiple SIgnal Classification (MUSIC) method
  • Min Norm method
  • Estimation of Signal Parameters via Rational Invariance Techniques (ESPRIT) method

Based on the book "Spectral Analysis of Signals", P. Stoica & R. Moses

Methods results

To illustrate the results, let's look at the real spectrum of a testing sinal, and its pseudo spectrum generated by MUSIC method.
Real spectrum Pseudo spectrum (MUSIC)

Below you can see the real and the estimated sinusoidal components of the testing signal. Note that the results of the Min Norm method is very similar.
MUSIC estimation ESPRIT estimation

Creating and running tests

To create your custom test, you can generate_signal and then apply_method from utils.py.

To run any test, simply go to the directory 'source' and type in the following command in your terminal.

  python test.py