English

Performance bound of the intensity-based model for noisy phase retrieval

Information Theory 2021-12-30 v2 math.IT

Abstract

The aim of noisy phase retrieval is to estimate a signal x0Cd\mathbf{x}_0\in \mathbb{C}^d from mm noisy intensity measurements bj=aj,x02+ηj,  j=1,,mb_j=\left\lvert \langle \mathbf{a}_j,\mathbf{x}_0 \rangle \right\rvert^2+\eta_j, \; j=1,\ldots,m, where ajCd\mathbf{a}_j \in \mathbb{C}^d are known measurement vectors and η=(η1,,ηm)Rm\eta=(\eta_1,\ldots,\eta_m)^\top \in \mathbb{R}^m is a noise vector. A commonly used model for estimating x0\mathbf{x}_0 is the intensity-based model x^:=\mboxargminxCdj=1m(aj,x2bj)2\widehat{\mathbf{x}}:=\mbox{argmin}_{\mathbf{x} \in \mathbb{C}^d} \sum_{j=1}^m \big(\left\lvert \langle \mathbf{a}_j,\mathbf{x} \rangle \right\rvert^2-b_j \big)^2. Although one has already developed many efficient algorithms to solve the intensity-based model, there are very few results about its estimation performance. In this paper, we focus on the estimation performance of the intensity-based model and prove that the error bound satisfies minθRx^eiθx02min{η2m1/4,η2x02m}\min_{\theta\in \mathbb{R}}\|\widehat{\mathbf{x}}-e^{i\theta}\mathbf{x}_0\|_2 \lesssim \min\Big\{\frac{\sqrt{\|\eta\|_2}}{{m}^{1/4}}, \frac{\|\eta\|_2}{\| \mathbf{x}_0\|_2 \cdot \sqrt{m}}\Big\} under the assumption of mdm \gtrsim d and aj,j=1,,m,\mathbf{a}_j, j=1,\ldots,m, being Gaussian random vectors. We also show that the error bound is sharp. For the case where x0\mathbf{x}_0 is a ss-sparse signal, we present a similar result under the assumption of mslog(ed/s)m \gtrsim s \log (ed/s). To the best of our knowledge, our results are the first theoretical guarantees for the intensity-based model and its sparse version. Our proofs employ Mendelson's small ball method which can deliver an effective lower bound on a nonnegative empirical process.

Keywords

Cite

@article{arxiv.2004.08764,
  title  = {Performance bound of the intensity-based model for noisy phase retrieval},
  author = {Meng Huang and Zhiqiang Xu},
  journal= {arXiv preprint arXiv:2004.08764},
  year   = {2021}
}

Comments

40 pages

R2 v1 2026-06-23T14:56:40.081Z