Self-Calibration and Bilinear Inverse Problems via Linear Least Squares

TL;DR

This paper introduces a simple, efficient linear least squares method for solving self-calibration and bilinear inverse problems, including blind deconvolution, with theoretical guarantees and practical applications.

Contribution

It proposes a novel linear least squares approach for self-calibration in bilinear inverse problems, providing explicit guarantees and spectral methods with near-optimal sampling complexity.

Findings

Algorithms are numerically efficient and suitable for real-time deployment.

Theoretical guarantees and stability are established for the proposed methods.

Numerical simulations demonstrate the effectiveness of the approach.

Abstract

Whenever we use devices to take measurements, calibration is indispensable. While the purpose of calibration is to reduce bias and uncertainty in the measurements, it can be quite difficult, expensive, and sometimes even impossible to implement. We study a challenging problem called \emph{self-calibration}, i.e., the task of designing an algorithm for devices so that the algorithm is able to perform calibration automatically. More precisely, we consider the setup $\boldsymbol{y} = \mathcal{A}(\boldsymbol{d}) \boldsymbol{x} + \boldsymbol{\epsilon}$ where only partial information about the sensing matrix $\mathcal{A}(\boldsymbol{d})$ is known and where $\mathcal{A}(\boldsymbol{d})$ linearly depends on $\boldsymbol{d}$. The goal is to estimate the calibration parameter $\boldsymbol{d}$ (resolve the uncertainty in the sensing process) and the signal/object of interests $\boldsymbol{x}$ simultaneously. For three different models of practical relevance, we show how such a \emph{bilinear} inverse problem, including blind deconvolution as an important example, can be solved via a simple \emph{linear least squares} approach. As a consequence, the proposed algorithms are numerically extremely efficient, thus potentially allowing for real-time deployment. We also present a variation of the least squares approach, which leads to a~\emph{spectral method}, where the solution to the bilinear inverse problem can be found by computing the singular vector associated with the smallest singular value of a certain matrix derived from the bilinear system. Explicit theoretical guarantees and stability theory are derived for both techniques; and the number of sampling complexity is nearly optimal (up to a poly-log factor). Applications in imaging sciences and signal processing are discussed and numerical simulations are presented to demonstrate the effectiveness and efficiency of our approach.