Modular Aggregation as a Debiasing Method for Non-Stationary Discrete Sources: Convergence and Numerical Validatio
The paper proposes a modular aggregation method to extract high-quality randomness from imperfect physical sources, even when those sources change over time (non-stationary). By summing outcomes modulo m, the resulting distribution rapidly converges to a uniform distribution. This technique is robust and useful for practical applications like Quantum Random Number Generators.