Quantum Two-Mode Squeezing Radar and Noise Radar: Correlation Coefficients for Target Detection

TL;DR

This paper develops a statistical framework for quantum two-mode squeezing radars, using correlation coefficients derived from covariance matrices to improve target detection, and provides explicit performance predictions based on simulated data.

Contribution

It introduces a method to estimate correlation coefficients for QTMS radars and derives explicit ROC curves for target detection performance.

Findings

Estimates follow a Rice distribution related to the true correlation coefficient.

Explicit ROC expressions enable performance prediction.

Simulation results validate the theoretical approach.

Abstract

Quantum two-mode squeezing (QTMS) radars and noise radars detect targets by correlating the received signal with an internally stored recording. A covariance matrix can be calculated between the two which, in theory, is a function of a single correlation coefficient. This coefficient can be used to decide whether a target is present or absent. We can estimate the correlation coefficient by minimizing the Frobenius norm between the sample covariance matrix and the theoretically expected form of the matrix. Using simulated data, we show that the estimates follow a Rice distribution whose parameters are simple functions of the underlying, "true" correlation coefficient as well as the number of integrated samples. We obtain an explicit expression for the receiver operating characteristic curve that results when the correlation coefficient is used for target detection. This is an important first step toward performance prediction for QTMS radars.