A novel method is presented for estimating a phase sequence from an ensemble of measured phase sequences that have noise and ambiguities of integral multiples of 2π rad. The maximum likelihood equation is approximately solved at a small measurement-error limit to obtain the new estimator. Statistical performance of the new estimator and the result of a numerical simulation of high-resolution imaging through the turbulent atmosphere are compared with those of the conventional phase unwrapping method which eliminates stepwise the 27π ambiguities by assuming some continuity constraints.
© 1983 Optical Society of America
Kazuyoshi Itoh and Yoshihiro Ohtsuka, "Phase estimation based on the maximum likelihood criterion," Appl. Opt. 22, 3054-3057 (1983)