Introduction - If you have any usage issues, please Google them yourself
W(n) represents the weight vector at n, W(n+1) represent weight vector at n+1. U represents the step size which control the convergence speed. The value for the gradient vector is calculated through covariance matrix R and cross-correlation r.