Quantum nonlocality offers a secure way to produce random numbers: Their unpredictability is intrinsic and can be certified just by observing the statistic of the measurement outcomes, without assumptions on how they are produced. To do this, entangled pairs are generated and measured to violate a Bell inequality with the outcome statistics. However, after a projective quantum measurement, entanglement is entirely destroyed and cannot be used again. This fact poses an upper bound to the amount of randomness that can be produced from each quantum state when projective measurements are employed. Instead, by using weak measurements, some entanglement can be maintained and reutilized, and a sequence of weak measurements can extract an unbounded amount of randomness from a single state as predicted in [Phys. Rev. A 95, 020102(R) (2017)]. We study the feasibility of these weak measurements, analyze the robustness to imperfections in the quantum state they are applied to, and then test them using an optical setup based on polarization-entangled photon pairs. We show that the weak measurements are realizable, but can improve the performance of randomness generation only in close-to-ideal conditions.

• Received 29 January 2021
• Revised 9 April 2021
• Accepted 10 May 2021
• Corrected 26 October 2021

DOI:https://doi.org/10.1103/PhysRevA.103.062206