Empirical Validation of Quantum Randomness via NIST SP 800-22: A Scalable Experimental Framework

We present a scalable experimental framework for validating quantum-generated bitstreams using the NIST SP 800-22 statistical test suite. Quantum randomness was simulated using Qiskit’s aer_simulator, generating 10 independent sequences of 10,000,000 bits each. These bitstreams were evaluated across 15 statistical categories including Monobit, Block Frequency, FFT, Linear Complexity, and Excursion-based tests. All but one test passed the NIST-recommended thresholds for both pass rates and p-value distributions. Specifically, the Random Excursion (Standard) test narrowly missed the threshold with 52 out of 56 sequences passing. Despite this, the overall results confirm the high entropy and statistical randomness of the quantum-generated sequences. Our framework also incorporates a batch-based architecture that enforces streak continuity across sequence boundaries and uses SHA-256 hashing for integrity verification. This infrastructure supports reproducibility and lays the foundation for future exploration of probabilistic thresholds in ultra-large-scale quantum bitstreams. In particular, we plan to generate and analyze up to one trillion quantum bits to investigate whether there exists a natural limit to the length of repeated stochastic patterns (streaks) in true randomness.