SPARQ: Efficient Entanglement Distribution and Routing in Space-Air-Ground Quantum Networks

TL;DR

SPARQ tackles entanglement distribution in a heterogeneous space–air–ground quantum network by introducing a DQN-based routing policy trained on dynamic network graphs and a third-party entanglement distribution policy to minimize swap operations. The architecture spans satellites, high-altitude platforms, and ground nodes, with realistic FSO and fiber channels and mobility modeled via STK, and an upgraded SPARQ simulator integrated with QuNetSim and STK for evaluation. Empirical results show that TPED reduces memory usage by half and improves end-to-end fidelity by a few percent on average, while the DQN routing increases resolved teleportation requests by ~$39\%$ and achieves fidelity gains of up to ~${9}\%$ relative to baselines; training on dynamic topologies yields an additional ~${15}\%$ fidelity improvement over static snapshots, and the air layer contributes ~${23.5}\%$ fidelity gain over space–ground-only networks. Collectively, SPARQ demonstrates a viable path toward on-demand, high-fidelity entanglement distribution across a global quantum Internet prototype, with scalable, decentralized routing and efficient entanglement distribution.

Abstract

In this paper, a space-air-ground quantum (SPARQ) network is developed as a means for providing a seamless on-demand entanglement distribution. The node mobility in SPARQ poses significant challenges to entanglement routing. Existing quantum routing algorithms focus on stationary ground nodes and utilize link distance as an optimality metric, which is unrealistic for dynamic systems like SPARQ. Moreover, in contrast to the prior art that assumes homogeneous nodes, SPARQ encompasses heterogeneous nodes with different functionalities further complicates the entanglement distribution. To solve the entanglement routing problem, a deep reinforcement learning (RL) framework is proposed and trained using deep Q-network (DQN) on multiple graphs of SPARQ to account for the network dynamics. Subsequently, an entanglement distribution policy, third-party entanglement distribution (TPED), is proposed to establish entanglement between communication parties. A realistic quantum network simulator is designed for performance evaluation. Simulation results show that the TPED policy improves entanglement fidelity by 3% and reduces memory consumption by 50% compared with benchmark. The results also show that the proposed DQN algorithm improves the number of resolved teleportation requests by 39% compared with shortest path baseline and the entanglement fidelity by 2% compared with an RL algorithm that is based on long short-term memory (LSTM). It also improved entanglement fidelity by 6% and 9% compared with two state-of-the-art benchmarks. Moreover, the entanglement fidelity is improved by 15% compared with DQN trained on a snapshot of SPARQ. Additionally, SPARQ enhances the average entanglement fidelity by 23.5% compared with existing networks spanning only space and ground layers.

Figures (16)

• Figure 1: Proposed architecture of SPARQ. FSO channels are shown with green dashed lines and optical fiber channels are shown with red solid lines. S, R, and U indicate a source of entanglement, repeater, and end-user, respectively.
• Figure 2: Illustration of different topologies of the same SPARQ network while satellites are moving. Ground nodes are presented in green, satellite nodes are presented in black, and HAP nodes are presented in red. FSO channels are presented in green dashed lines and optical fiber channels are presented in red solid lines. The shown node positions do not reflect the actual position, instead, this is reflected by the edges and whether they are connected or disconnected.
• Figure 3: Illustration of the proposed routing model and entanglement distribution strategy. The TPED policy is employed here for entanglement distribution as it results in improved end-to-end entanglement quality, as will be detailed in Section \ref{['sec:simulation_results']}.
• Figure 4: Two different policies to establish entanglement between the source node $S$ and the destination node $D$, the circles inside nodes represent an entanglement state. The color of the entanglement state represents the node with which an entanglement state is shared. In this figure: (a) represents the network nodes and communication channels before establishing entanglement, (b) represents the intuitive approach, and (c) represents the proposed TPED policy, where the role of $N_1$ is to generate an entangled pair and send one part to $S$ and the other part to $N_2$.
• Figure 5: Relationship between transmissivity and entanglement fidelity.