Time-bin QKD

Team Members: Wang Ziyan, Wu Jinbin

Summary

The project is to implement time-bin QKD on the encoding and decoding ends.

List of Equipment

Equipment	Manufacturers	Quantities
Function Generator AFG1022	Tektronix, Inc	1
Oscilloscope HDO6104	Teledyne LeCroy	1
NIM FAN-OUT Module	S-Fifteen Instruments	1
NIM Anti-coincidence Module	S-Fifteen Instruments	2
NIM cables		multiples
BNC cables		multiples

Experiment Setup

1. Producing a short pulse

The shortest pulse that can be generated by the function generator is about 40 nanoseconds, which is too long for our purpose. To create a shorter pulse, we performed pulse shortening using delayed Anti-Coincidence Logic.

A short electrical pulse was first generated using function generator with high level as 0V and low level as -1.0V, following NIM (Nuclear Instrumentation Module) Standard.

The input pulse was first passed through a fanout module to produce two identical copies, B and B'. A relative time delay, $\Delta t$, was introduced between the two signals before they were sent to the signal and veto inputs of the anti-coincidence module. The anti-coincidence module produces an output only when B is present while the delayed signal B' is absent. This operation can be represented as

$V_{out}=B\wedge \overline{B^{\prime }}$.

When the leading edge of B arrives, the output becomes high. After a delay $\Delta t$, the leading edge of B' reaches the veto input and suppresses the output. Consequently, the original pulse is converted into a much shorter pulse whose width is approximately determined by the relative delay:

$T_{\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{s}\mathrm{e}}\approx \Delta t$.

Thus, by controlling the delay between the two copies of the input signal, the output pulse width can be adjusted independently of the width of the original pulse. The short pulse is generated at the rising edge of the input signal.

Experiment Record

2026-10-5

1. Prof. Kurtsiefer explained the principles of experiments and the experimental apparatuses.

2. We learned how to use Oscilloscope HDO6104.

2026-10-6

1. We used the Function Generator AFG1022 to generate a short pulse that followed NIM Standard and had a width of about 40ns. The rise time (10% to 90%) of the pulse is approximately 2ns.

2. A NIM FAN-OUT Module was used to split the short pulse into two identical NIM pulses, with the rise time (10% to 90%) decreased to about 780ps.

3. The two identical NIM pulse were sent to the signal and veto ports of an NIM Anti-coincidence Module respectively, producing an output pulse with a width of about 2ns. The rise time of the pulse is about 780ps.

2026-10-14

1. The same as before, we used the Function Generator AFG1022 to generate a short pulse and then inputted the pulse into the NIM FAN-OUT Module.

2. We measured the three output ports and obtained three rise times (10% to 90%), thich are respectively 800ps, 760ps and 720ps.

3. In order to obtain the best signal, we inputted the pulse with the shortest rise time to the "signal" port, and the pulse with the second shortest rise time to the "veto" port. Then we got an output pulse with a width of about 1ns.

4. We connected the laser diode to the function generator and output square waves. The laser diode started emitting light form a 0.5-volt high level.

5. We measured the battery voltage to be 9.47 volts. This is much higher than the operating voltage range of the laser diode, so it cannot be used as a DC power supply.

2026-10-19

1. We connected the battery (the 9.47-volts one), the photodiode and a resistor (9.959 kOhm) in series together. And we used a voltmeter to measure the voltage of this resistor.

2. We connected a laser diode to the direct current power supply, and gradually increased the output voltage. The laser diode begined to emit red light that can be observed with the naked eye at about 1.65V.

3. We aligned the illuminated area with the sensor region of the photodiode. When the output voltage of DC supply was 0V, we read the voltage display value 11.80mV and recorded it as the ambient voltage (dark current is included in it).

4. Starting from 1.60V and gradually increasing, we read the voltage values from the voltmeter every 0.5V and recorded them.

5. After subtracting the ambient voltage from the measured voltage and then dividing by the resistance, we obtained the photocurrent from the laser diode. Then we divede these photocurrent datas by response in 780nm (temporally consider the red light from laser diode is 780nm), and we obtained the power of the laser diode.

6. We draw a graph showing the output power of the laser diode in relation to the input voltage.