Comparison of behavioral performance using on-line, closed-loop BMI driven by a 10th order UKF, a Kalman filter, and a 10 tap Wiener filter fit using ridge regression.

Performance was measured as signal-to-noise ratio and correlation coefficient of the BMI-controlled cursor position to the target position. The bottom row shows mean difference of each algorithm compared against the Kalman filter, where larger numbers are better. Bold numbers indicate the best value in each row.