Hybrid Quantum-Classical Avalanche TRNG: Experimental Validation on a Rigetti Ankaa-3 Superconducting QPU
Luiz Nascimento (Kayos Intelligence LLC
PROPOSAL · v1.0 · 2025-12-08 · human
Abstract
This work reports an experimentally validated hybrid quantum-classical avalanche mechanism executed on real quantum hardware and designed for use in true random number generation (TRNG) and cryptographic kernels. Using the Rigetti Ankaa-3 superconducting quantum processor (84 qubits) accessed via AWS Braket (us-west-1), we implemented and measured three circuits that jointly characterize diffusion, entanglement and bit-flip sensitivity in a quantum avalanche regime. On 7 December 2025 we executed three independent jobs, all completed successfully, totaling 6,144 measurement shots over 16 physical qubits and producing 455 kB of raw result data. The first experiment was an 8-qubit “quantum avalanche cascade” circuit combining Hadamard, CNOT and phase rotations to emulate diffusion over an 8-bit space. From 4,096 shots we observed 256 distinct basis states, with 82.9% of all outcomes in high-entropy regions and a mean Hamming distance of 3.95/8 from |00000000⟩, corresponding to a quantum avalanche score of 49.4%, close to the 50% diffusion expected from classical avalanche cryptography. The second experiment used 5 qubits to probe entanglement-driven diffusion (A→B). The empirical distribution showed non-trivial GHZ-like structure: the combined frequency of |00000⟩ and |11111⟩ accounted for 29.1% of all shots, defining an effective entanglement fidelity of 29.1% under realistic noise. The third experiment selected a non-adjacent 3-qubit subset (B→A) from the same device topology and obtained GHZ-like outcomes with 20.7% frequency, indicating bidirectional correlation and diffusion even in a reduced subspace. Taken together, these results provide, to the best of our knowledge, the first reproducible dataset in which quantum diffusion on a public QPU numerically approaches classical avalanche behavior and can be integrated into a kernel-level hybrid TRNG design. All experiments were performed on commercially accessible hardware and can be independently reproduced, making this work a concrete reference point for future research on quantum-assisted entropy sources, post-quantum cryptography, and hybrid quantum-classical security architectures.