Enhancing Quantum Network Establishment Through Multi-Objective Genetic Algorithm

Abstract

A quantum network connecting quantum devices is expected to enhance quantum computing and information processing by increasing the number of qubits, and distributing computation and processing. However, noise-induced decoherence remains a big challenge as a quantum network can experience multiple sources of noises. Optimizing a quantum network for a given noise regime thus can improve its performance and reliability. This research uses multi-objective optimization with a genetic algorithm to find optimal conditions for constructing a quantum network in specific configurations. We employed the qwanta quantum network simulator to investigate two scenarios of a three-node quantum network with a quantum repeater in the middle. Pareto frontier analysis enables us to visualize the trade-offs between a quantum state fidelity and throughput in a quantum network. The results provide parameters that achieve the fidelity higher than the targeted value of 0.828 for quantum key distribution in a linear network with equal distances between nodes. However, for a three-nodes network with unequal distances based on Thailand's geographical locations, the highest fidelity achieved is 0.706. Expanding a genetic algorithm's search space to include a broader parameter range can potentially improve the results. This research demonstrates potential deployment of a genetic algorithm with multiple-objective optimization in a quantum network.