Recursive Least Square Lattice Implementation

I implemented the "Recursive LSL Algorithm Using A Posteriori Estimation Errors" described in Haykin's "Adaptive Filter Theory" book. However, I'm not getting the performance. I have seen published papers using the same algorithm and implemented them, but I still didn't get it right. Can someone point me to a verified implementation of the Recursive LSL or give me some pointers on where the most common mistakes are? Start a New Thread