Exact Minimax Estimation for Phase Synchronization

Abstract
We study the phase synchronization problem with measurements , where is an -dimensional complex unit-modulus vector and is a complex-valued Gaussian random matrix. It is assumed that each entry is observed with probability . We prove that the minimax lower bound of estimating under the squared loss is . We also show that both generalized power method and maximum likelihood estimator achieve the error bound . Thus, is the exact asymptotic minimax error of the problem. Our upper bound analysis involves a precise characterization of the statistical property of the power iteration. The lower bound is derived through an application of van Trees' inequality.
View on arXivComments on this paper