Parameters Optimization of Decoy-State Phase-Matching Quantum Key Distribution Based on the Nature-Inspired Algorithms

Phase-matching quantum-key distribution (PM-QKD) has achieved significant results in various practical applications. However, real-time communication requires dynamic adjustment and optimization of key parameters during communication. In this letter, we predict the PM-QKD parameters using nature-inspired algorithms (NIAs). The results are obtained from an exhaustive traversal algorithm (ETA), which serves as a benchmark. We mainly study the parameter optimization effects of the two NIAs: ant colony optimization (ACO) and the genetic algorithm (GA). The configuration of the inherent parameters of these algorithms in the decoy-state PM-QKD is also discussed. The simulation results indicate that the parameters obtained by the ACO exhibit superior convergence and stability, whereas the GA results are relatively scattered. Nevertheless, more than 97% of the key rates predicted by both algorithms are highly consistent with the optimal key rate. Moreover, the relative error of the key rates remained below 10%. Furthermore, NIAs maintain power consumption below 8 W and require three orders of magnitude less computing time than ETA.