Calibration of the characteristic frequency of an optical tweezer using a recursive least-squares approach

We describe the use of a recursive least-squares approach for calibration of the characteristic frequency of an optical tweezer. Unlike commonly used off-line calibration methods, the RLS algorithm does not require data averaging to reduce the effects of Brownian motion. We use computer simulations to demonstrate that the RLS algorithm can be used to efficiently calibrate the characteristic frequency of a trapped, 10-micron diameter polystyrene bead. However, experimental results suggest that applying the RLS method to an actual optical tweezer system is more difficult because of the measurement errors and laboratory noise.