Modeling and Mitigation of Noise and Interference Toolbox in MATLAB

This free toolbox provides MATLAB functions and demonstrations for statistical modeling and mitigation of certain kinds of noise and interference in acoustic systems, power lines, wireless communications and wireless sensor networks. 

The noise and interference can come from other sources in the same frequency band of operation or in adjacent frequency bands.  The toolbox enables a user to

1. generate impulsive noise/interference
2. fit measured data to impulsive noise models
3. apply nonlinear filters to mitigate impulsive noise
4. improve detection performance of a signal in impulsive noise

In communication systems, additive noise has traditionally been assumed to be Gaussian. Redesigning the communication receiver with an impulsive noise model in mind can achieve a 10x-100x reduction in bit error rate or an increase of up to 2x in bit rate in interference-limited channels. The toolbox supports single- and multiple-antenna systems.

For wireless networks, we have shown analytically that

• A symmetric alpha stable distribution models RF interference in wireless sensor networks and femtocell networks, and
• A Gaussian mixture models RF interference in cellular networks, hotspots, and dense Wi-Fi networks.

The toolbox supports the symmetric alpha stable distribution, Gaussian mixture model, and Middleton Class A distribution.  The Middleton Class A distribution is a special case of a Gaussian mixture model.

The toolbox is available free of charge at

http://users.ece.utexas.edu/~bevans/projects/rfi/software/index.html