Performance of Sequential Phase Diversity with Dynamical Solar Scenes

Phase diversity techniques are usually based on the comparison of synchronously acquired pairs of focused-defocused images. This way, differences between both images are avoided except from random pixel variations due to noise on the detector and the phase diversity itself. In some astronomical instruments, though, the two images are not taken simultaneously. This work studies the impact of carrying out phase diversity with pairs of asynchronously acquired images while observing an evolving solar scene. We evaluate the performance of this technique as a function of the time gap between the images through the use of a magnetohydrodynamical simulation of the solar scene as observed by an instrument. We describe the incident wave front with two numbers of Zernike polynomials (20 or 32) to explore their effect on the wave front sensing accuracy and we employ two levels of noise to study their impact in the object restoration. We find that a time gap among our simulation images smaller than ~10 s has a negligible impact on the performance of the method. The rms error of the Zernike coefficients fitting worsens exponentially from there on, but the evolution is similar no matter the number of polynomials used in the fitting. Meanwhile, the quality of the object restoration benefits from lower noise levels, but it decreases linearly with the time gap independently of the amount of noise.