| （英） |
In quantum networks, qubits are transmitted via quantum teleportation using quantum entanglement established between the transmitting and receiving nodes. In quantum teleportation, when the entanglement fidelity is low, purification operations are required to enhance fidelity. Additionally, the number of quantum entanglements available at each node is limited, and the frequency of these operations significantly impacts the qubit transmission time. Consequently, the transmission time and fidelity of qubits vary depending on the connectivity of the links; however, a detailed examination of quantum network topology has not yet been conducted. In this paper, we first establish a topology design method based on genetic algorithms. Subsequently, we explore a topology design method for quantum networks using deep reinforcement learning. Through numerical examples, we investigate the effectiveness of topology design using genetic algorithms and then discuss the potential impact of applying deep reinforcement learning. |