Stochastic Models and Techniques for Sparse Signals
Real-world data such as multimedia, biomedical, and telecommunication signals are formed of specific structures. However, these structures only determine some general properties of the data while the unknown or unpredictable parts are assumed to be random. This fact suggests that we can use stochastic models to explain real-world signals. Processes such as Gaussian white noise or Gaussian ARMA processes are well-known examples which are extensively used in modeling some components of the natural signals.