PYTHON · NUMPY

The denominator that closes it

Show that the fix is not a stricter certificate but the missing task denominator, and close the loop back to the course's first module.

01Challenge

You have a channel that a safety certificate permits and an energy monitor cannot see. Decide what would actually close it — and check your answer against what the standards already do and do not define.

Predict firstA covert channel runs under the energy alarm. What closes it?

02Model

The instinct is to tighten the certificate. That fails, and the reason is worth understanding. The adversary's best channel was slowing down, and slowing down is the safe direction. A safety envelope is structurally unable to object to excess caution, so making it stricter pushes an adversary further into the region the envelope already blesses. A certificate that bounds danger does not bound waste. So look at what each layer actually bounds. The certificate bounds the trajectory. The energy monitor bounds instantaneous power. Neither bounds joules per unit of work completed. And a policy exploiting this channel is, by construction, spending more energy than its task requires. A monitor normalised by completed work sees that immediately, because dawdling is precisely what it measures. And that quantity is the one you already met in the very first module of the prerequisite course, as a gap: no published standard defines robot energy per task. It was a comparability problem. It turns out to be a security problem too.

03Guided practice

Model a task-normalised monitor and check whether it sees what the power monitor missed. Then list which standard supplies which term.

04Feedback

PASS WHEN You found the monitor that sees what the other two cannot.

If your run fails, check:

The power monitor reports normal for both runs, because every individual reading it sees is legitimate.
The task monitor sees the covert run immediately: same work, more joules. Dawdling is exactly what it measures.
And this is the quantity PAI-280 identified as missing from every published standard. ISO 20140 defines efficiency as a ratio and leaves both terms to you; ISO/TS 25213 measures consumption and supplies no task denominator either.
The gap was named as a comparability problem. It is also the thing that closes a covert channel.

05Retrieve & space

Why does making the safety certificate stricter fail to close this channel?
What does each of the three layers bound, and which one is missing?
Where had you already met the missing quantity in this course sequence?

06Mastery & project

When a monitor misses something, ask what quantity it normalises by — and whether the right denominator exists in any standard at all.

Pick a monitored system. Identify what its alarms normalise by. Propose the denominator that would make its alarms meaningful, and check whether any standard defines it.

GO FURTHER · THE LIVE PLAYGROUND

Open the instrument ↗ · The channel you are closing.
Back to where the gap was named ↗ · PAI-280's reporting record: the task denominator no standard supplies.

← Under the alarm Back to course →

PYTHON · NUMPY

The denominator that closes it

Same covert policy, two monitors. One watches watts, the other watches joules per task.

You read

the arrays and values already in scope

You change

the code you write in each cell

Fixed

the dataset and the checks that grade you