Skip to content

What happens to an entropy source when voltage or temperature drifts

Entropy sources drift with temperature, supply voltage and age, and they can be pushed deliberately. Drift rarely shows in the output statistics — it shows in the physical quantities the entropy model depends on, which is why a certificate issued at evaluation cannot speak for a device in service years later.

Temperature

Every physical process in a semiconductor device has a temperature coefficient. Laser threshold current, photodiode responsivity, amplifier gain and ADC reference all move with temperature, and they do not move together.

A source characterised at 25 °C and deployed in a rack at 55 °C is operating somewhere the model was not tested. It may still be sound. Whether it is, is a question of measurement rather than assumption.

Supply voltage

Rail variation shifts bias points. A laser driven marginally below threshold at nominal supply may not return fully below threshold at the low end of tolerance, and if it does not, consecutive pulses are no longer phase-independent — which is the entropy mechanism itself.

The output would look unchanged. The entropy per sample would not be.

Ageing

Semiconductor lasers degrade slowly and predictably: threshold current rises over operating hours. Over a deployment measured in years, a device that started comfortably inside its envelope can drift toward its edge.

This is the case that most clearly defeats one-time characterisation. The device passed evaluation, and it is now different from the device that passed.

Deliberate interference

A source in a physically accessible device can be pushed: heated, undervolted, or exposed to injected signals. The attacker's goal is not to stop it — a stopped source is detected — but to reduce its entropy while keeping the output plausible.

That attack is defeated by measuring the physical parameters rather than the output, because reducing entropy while holding every physical quantity at its modelled value is a far harder problem than reducing entropy alone.

The conclusion this forces

None of these appear reliably in statistical tests, and all of them invalidate an entropy estimate. A static certificate describes a device that no longer exists in quite the same form. Continuous measurement of the model's quantities is the only thing that keeps the claim current.