↖ HELIOLINK TECHNOLOGIES
LABORATORIES SECURITIES →
HelioLink logo HelioLink Laboratories
// Research program

Every federal system must switch to quantum-resistant encryption. The new math is heavy. We work on the hardware that carries it.

NIST finalized the first post-quantum cryptography standards in August 2024 (FIPS 203, 204, 205; HQC added March 2025), and NSA's CNSA 2.0 timeline requires national-security systems to adopt them by 2033. The algorithms exist; running them fast on small, power-limited devices is the unsolved part. Our research program addresses that deployment gap: better entropy sources and photonic computational throughput.

TRACK 1

Quantum random number generation for cryptographic entropy

Cryptographic security reduces, at its base, to the unpredictability of key material. Software pseudo-random generators are deterministic; their output is only as unpredictable as their seed and state. Quantum random number generators derive randomness from quantum-mechanical processes (photon arrival statistics, vacuum fluctuations) whose outcomes no amount of computation can predict.

QRNG hardware is commercially established; validated entropy sources under NIST SP 800-90B exist. Our interest is narrower: integrating photonic entropy generation into the same optical systems that carry the quantum key channel, so that key generation and key distribution share hardware rather than accumulating it.

TRACK 2

Photonic acceleration of post-quantum algorithms

NIST's lattice-based schemes impose substantially higher computational cost than the classical cryptography they replace, and the burden falls hardest on constrained devices such as avionics and battery-powered sensors. The Office of the National Cyber Director estimates federal migration costs alone at roughly $7.1 billion between 2025 and 2035; throughput on constrained hardware is a large part of why.

Photonic integrated circuits perform certain operation classes, notably large matrix multiplication and parallel arithmetic via wavelength division multiplexing, at high speed and low power. Those are the operation classes that dominate lattice-based PQC: number-theoretic-transform butterflies for ML-KEM and ML-DSA, parallel hash trees for SLH-DSA. We investigate mapping these kernels onto photonic hardware, analogous to how existing cryptographic accelerators offload specific functions to dedicated silicon.

Problems still open, for us and for the field: electro-optical interface bandwidth, deterministic timing for side-channel resistance, and a FIPS 140-3 certification pathway for modules containing photonic components. This track is the research behind H.U.G.0, the photonic security module we are designing; the full analysis is published in our white paper (PDF).

// Scope of claims

What we claim

  • The combination of quantum entropy sources and photonic computation is a viable research direction for PQC's performance problem, grounded in demonstrated physics.
  • The licensed NASA transceiver architecture materially lowers the integration barrier for QKD on single-aperture platforms.
  • Organizations holding data with 10+ year confidentiality requirements should treat "harvest now, decrypt later" as a present risk, consistent with CISA/NSA/NIST joint guidance.

What we do not claim

  • We have not built a production photonic PQC accelerator. No one has. There is no product announcement hiding on this page.
  • Photonic acceleration is not required for PQC migration; electronic accelerators and capable general-purpose hardware work today.
  • Quantum computers are not breaking RSA tomorrow. Annual expert surveys such as the Global Risk Institute's quantum threat timeline generally place cryptographically relevant machines a decade or more away. The urgency comes from data longevity.

Working on an adjacent problem?

We collaborate with university groups, consortium members, and program teams on photonics, entropy, and PQC deployment questions.

contact@heliolink.net