Efficient progressive readout of a register of (qu)bits

Abstract

Recently, a series of articles by Combes et al. has shown that it was possible to greatly improve the measurement rate of a register of qubits for given detector resources by means of a clever feedback control scheme. However, this speed-up came at an exponential cost in terms of complexity and memory use. In this article, I propose a simple efficient algorithm --exponentially more frugal in memory and less complex to implement-- which is asymptotically as fast. I use extensively the implicit classicality of the situation to provide a slightly more straightforward interpretation of the results. I compute the speed-up rates exactly in the case of the proposed model and in the case of the open-loop scheme of Combes et al. and prove that they indeed provide the same asymptotic speed-up.