Introduction - If you have any usage issues, please Google them yourself
One-dimensional speech signal reconstruction, orthogonal matching pursuit, the first of its sub-frame, and then again a reconstruction of a final calculation of the average frame reconstruction error