OS-CMA for blind channel equalization and signal extraction

The optimal step-size constant modulus algorithm (OS-CMA) is based on a computationally efficient iterative technique to minimize the CM cost function. This technique computes algebraically the step size (adaption coefficient) globally minimizing the CM contrast in the search direction at each iteration.

OS-CMA's optimization technique is similar to that used in RobustICA. As a result, the OS-CMA shares many advantages of RobustICA, including high convergence speed and the ability to avoid spurious local extrema.

More details about the OS-CMA and its comparative performance analysis can be found in the references below.


Download the OS-CMA package (Release 1, November 21, 2014; Matlab version; 11KB).
Back to software