A fast Newton method is presented for solving the entropy maximization prob
lem in the Bayesian statistical approach to phase estimation. The method re
quires only O(n log n) instead of standard O(n(3)) floating point operation
s per iteration, while converging in the same rate as the standard Newton m
ethod. The method is described and related computational issues are discuss
ed. Numerical results on simple test cases are also presented to demonstrat
e the behavior of the method.