English

Almost Everywhere Generalized Phase Retrieval

Functional Analysis 2019-09-20 v1 Information Theory Algebraic Geometry math.IT

Abstract

The aim of generalized phase retrieval is to recover xFd\mathbf{x}\in \mathbb{F}^d from the quadratic measurements xA1x,,xANx\mathbf{x}^*A_1\mathbf{x},\ldots,\mathbf{x}^*A_N\mathbf{x}, where AjHd(F)A_j\in \mathbf{H}_d(\mathbb{F}) and F=R\mathbb{F}=\mathbb{R} or C\mathbb{C}. In this paper, we study the matrix set A=(Aj)j=1N\mathcal{A}=(A_j)_{j=1}^N which has the almost everywhere phase retrieval property. For the case F=R\mathbb{F}=\mathbb{R}, we show that Nd+1N\geq d+1 generic matrices with prescribed ranks have almost everywhere phase retrieval property. We also extend this result to the case where A1,,ANA_1,\ldots,A_N are orthogonal matrices and hence establish the almost everywhere phase retrieval property for the fusion frame phase retrieval. For the case where F=C\mathbb{F}=\mathbb{C}, we obtain similar results under the assumption of N2dN\geq 2d. We lower the measurement number d+1d+1 (resp. 2d2d) with showing that there exist N=dN=d (resp. 2d12d-1) matrices A1,,ANHd(R)A_1,\ldots, A_N\in \mathbf{H}_d(\mathbb{R}) (resp. Hd(C)\mathbf{H}_d(\mathbb{C})) which have the almost everywhere phase retrieval property. Our results are an extension of almost everywhere phase retrieval from the standard phase retrieval to the general setting and the proofs are often based on some new ideas about determinant variety.

Cite

@article{arxiv.1909.08874,
  title  = {Almost Everywhere Generalized Phase Retrieval},
  author = {Meng Huang and Yi Rong and Yang Wang and Zhiqiang Xu},
  journal= {arXiv preprint arXiv:1909.08874},
  year   = {2019}
}

Comments

27 pages

R2 v1 2026-06-23T11:20:01.689Z