![]() jAudio 1.0 Aggregator Appendix |
||
|
2D Polynomial ApproximationThis specific aggregator was first released in an August 2010 working paper by Daniel McEnnis. It transforms a 2D matrix of signal feature vectors into a set of coeffecients of the polynomial function f(x,y) that bests matches the given signal. It is calucalated by constructing a matrix of coeffecients by substituting concrete data points into each term that coeffecient is attached to (such as x^2*y^3) and a answer matrix is created from the signal at that matrix index. The coeffecients are then calculated by a least squares minimization matrix solver (colt matrix function). |