Examples

The Information Reconciliation Library for CV-QKD Systems provides example use cases in C++ and Python, demonstrating the library’s usage. These examples are located in the tests folders.

C++

An example demonstrating the use of the compiled library in C++ is located in the IR_lib/cpp/tests directory. This example assumes that the transmitted and received data are saved in a file. The data is first read from the file, and then the reconciliation process is performed using the library. The statistics of the reconciliation process are printed to the console. The script that utilizes the library is provided in the src/use_case.cpp file.

To compile the example, the library must first be compiled as described in Installation Guide. Then, navigate to the IR_lib/cpp/tests directory and execute the following commands:

mkdir build
cd build
cmake ..
make

The compiled executable is located in the IR_lib/cpp/tests/bin directory.

Python

Examples demonstrating the use of the compiled library in Python are located in the IR_lib/tests directory. These examples first sample random data and transmit them to the receiver over an additive white Gaussian noise (AWGN) channel. The receiver then performs the reconciliation process using the library. The statistics of the reconciliation process are printed to the console.

To compile the library, follow the instructions described in Installation Guide. Then, you can use the Jupyter notebook provided in the IR_lib/tests directory to run the examples.

The first example, named QCrypt_script.ipynb, simulates the reconciliation process for a fixed reconciliation rate across different channel signal-to-noise ratio values and displays the statistics.

By modifying the slow and fast decoder options and the MDR dimension, users can compare the performance of the decoders. The results of the test cases are presented in Fig. 2 for the two extreme rate values supported by the library: rate 0.01 and rate 0.2.

Fig. 2 Results of test cases. FER, BER, and NoI represent frame error rate, bit error rate, and number of decoding iterations, respectively. The results are shown for MDR dimensions of 8, 4, 2, and 1, from left to right. The slow decoder is denoted by points, while the fast decoder is represented by lines. Decoding durations are normalized to the number of parallel decoding operations and were obtained using an AMD EPYC™ 7713P 64-Core Processor.

The second example, named QCrypt_script_fixed_beta.ipynb, demonstrates the performance of the reconciliation process for different channel signal-to-noise ratio values at a fixed reconciliation efficiency. Users can also change the modulation of the transmitted states between Gaussian and 4-state (BPSK) modulation and compare the efficiency of the multi-dimensional reconciliation.