Case Study
Undergraduate Quantum Education: Quantum Key Distribution with Qubi
A one-hour session at the University of Maryland that builds a quantum key distribution protocol from scratch in front of the room, using a single entangled pair of Qubis. Nothing is derived on a board. Every rule is shown physically first, a student volunteer takes one half of the pair to run the key exchange live, and the class reasons its way to why an eavesdropper cannot hide.
Taught by Sohum Thakkar, CEO · sohum@qolour.com
What students take away
By the end of the hour, students can explain each of the following.
A qubit is a physical thing
A Qubi’s orientation stands for the spin direction of a real electron, so “quantum state” becomes something students can point at and rotate rather than an abstraction.
The basis is a choice
You decide which axis to measure along. Measuring the same way twice gives the same answer; measuring a different way does not. That free choice is the whole hinge of the protocol.
Entanglement, built by hand
The controlled-NOT gate is shown physically as the operation that turns two independent qubits into a correlated pair, which is what the protocol distributes.
Why eavesdropping is detectable
Students reason through why an eavesdropper who has to guess the basis leaves a measurable error rate behind, and why announcing the basis publicly gives nothing away.
Before the protocol
What a qubit actually does
The protocol only makes sense once three facts are established by hand: a qubit has a direction, you choose the axis you measure it along, and measuring the same way twice gives you the same answer back. Each is demonstrated before it is used.
The protocol
Building quantum key distribution, one problem at a time
The workshop follows the shape of the protocol itself. Each step solves the hole opened by the step before it, so students arrive at the finished scheme by fixing their own broken version rather than being handed the answer. Clips carrying the instructor’s explanation wait for you to start them.
Share an entangled pair
One particle of an entangled pair goes to each party. In the room this is done by hand: two Qubis are knocked together to entangle them, and the CNOT is explained as the operation that makes correlation out of two independent qubits.
Measure, and record the result
Both parties measure their own qubit and write down what they got. Repeat, and a shared string of bits builds up on both sides. In real life you never see the entanglement itself, only the measurement outcomes.
Notice the key isn’t secure yet
The two strings match, so it looks like a shared key. But if Eve intercepts a qubit on its way across, she reads the same bit and neither party is any the wiser. A naive entangled key leaks silently.
Randomise the basis
Instead of always measuring up-and-down, each party independently and at random either turns their qubit 90 degrees first or doesn’t. Turn the same way as each other and the results still agree; turn differently and it is a coin flip.
Broadcast the bases, not the results
Afterwards each party announces publicly, to anyone listening, which trials they turned on — never what they measured. Trials where the two turned differently are thrown away. This is the move the workshop puts a student volunteer on stage to perform.
Catch Eve in the error rate
Because the bases are only announced after the fact, Eve has to guess. On the trials the two parties kept, she guesses wrong half the time, and half of those flip the result — a 25% error rate where there should be none. Eavesdropping stops being invisible.
Spend half the key to check the other half
Out of a thousand trials where the bases matched, the two parties publicly compare the first five hundred. If every one agrees, nobody was listening, and the remaining five hundred become a secret key nobody else in the universe holds.