Extrapolating Zernike Moments to Predict Future Optical Wave-fronts in Adaptive Optics Using Real Time Data Mining

Abstract

We present the details of predicting atmospheric turbulence by mining Zernike moment data obtained from simulations as well as experiments. Temporally correlated optical wave-fronts were simulated such that they followed Kolmogorov phase statistics. The wave-fronts reconstructed either by modal or zonal methods can be represented in terms of Zernike moments. The servo lag error in adaptive optics is minimized by predicting Zernike moments in the near future by using the data from the immediate past. It is shown statistically that the prediction accuracy depends on the number of past phase screens used for prediction and servo lag time scales. The algorithm is optimized in terms of these parameters for real time and efficient operation of the adaptive optics system. On an average, we report more than 3% improvement in the wave-front compensation after prediction. This analysis helps in optimizing the design parameters for sensing and correction in closed loop adaptive optics systems.