Properties of The Discrete Sinc Quantum State and Applications to Measurement Interpolation

TL;DR

This paper investigates properties of the discrete sinc quantum state used in phase estimation and introduces new estimators that improve measurement precision by analyzing measurement outcome frequencies.

Contribution

It proposes and analyzes novel estimators based on the two most frequent measurement outcomes, enhancing precision without extra qubits.

Findings

The Ratio-Based Estimator provides a closed-form expression for the encoded value.

The Coin Approximation Estimator effectively uses Bernoulli process parameters.

Additional properties of the discrete sinc state are identified for future estimator design.

Abstract

Extracting the outcome of a quantum computation is a difficult task. In many cases, the quantum phase estimation algorithm is used to digitally encode a value in a quantum register whose amplitudes' magnitudes reflect the discrete sinc function. In the standard implementation the value is approximated by the most frequent outcome, however, using the frequencies of other outcomes allows for increased precision without using additional qubits. One existing approach is to use Maximum Likelihood Estimation, which uses the frequencies of all measurement outcomes. We provide and analyze several alternative estimators, the best of which rely on only the two most frequent measurement outcomes. The Ratio-Based Estimator uses a closed form expression for the decimal part of the encoded value using the ratio of the two most frequent outcomes. The Coin Approximation Estimator relies on the fact that the decimal part of the encoded value is very well approximated by the parameter of the Bernoulli process represented by the magnitudes of the largest two amplitudes. We also provide additional properties of the discrete sinc state that could be used to design other estimators.