Neural-RLWE Double Ratchet for Real-Time Media: Fast Per-Frame Keys with Provable Base Security

Park et al. (2023) analyze E2EE in video conferencing and propose PQ KEM integration; Yadav (2023) surveys RLWE-based key exchange and Ding reconciliation techniques; Kim et al. (2023) improve the efficiency of neural cryptography synchronization (1-h random walk and batch schemes). This idea composes them: establish a PQ base key via RLWE KEM, then run a neural synchronization process seeded by the base key to generate extremely frequent subkeys (e.g., per-frame). The base key gives the security backbone; the neural layer acts like a high-throughput ratchet that is cheap to compute and inherently stream-friendly. Deviations in synchronization (unexpected convergence/failure rates) become detection signals for active interference, feeding into adaptive re-keying (as in Idea 1). The novelty is a clean separation of provable initial security (RLWE KEM) and a practical, measurement-driven subkey schedule inspired by neural cryptography’s strengths, with clear guardrails (e.g., if neural sync diverges, fall back to KDF-based rekey). Impact: PQ-ready, latency-friendly E2EE for large calls and streaming scenarios, with strong throughput and rapid forward secrecy without frequent heavyweight KEM invocations.

References:

1. End-to-End Post-Quantum Cryptography Encryption Protocol for Video Conferencing System Based on Government Public Key Infrastructure. Yeongjae Park, H. Yoo, Jieun Ryu, Young-Rak Choi, Ju-Sung Kang, Yongjin Yeom (2023). Applied System Innovation.
2. Improvement of the Efficiency of Neural Cryptography-Based Secret Key Exchange Algorithm. Juyoung Kim, Sooyong Jeong, Dowon Hong, Namsu Jho (2023). IEEE Access.
3. Work in Lattice-Based Cryptography: Key Exchange Protocols under RLWE-Based Problems and Ding Reconciliation Technique. Sonam Yadav (2023). International Journal for Research Publication and Seminars.