Quantum entanglement is the cornerstone of quantum network communication, yet the qubit resources required to establish it are severely limited. Moreover, users within the network impose strict delay constraints on entanglement requests due to various factors, such as the real-time requirements of upper-layer applications or the finite coherence time of userside quantum memories. Consequently, effectively scheduling scarce qubit resources to serve these time-sensitive requests presents a significant challenge. This difficulty is exacerbated by the partial observability of quantum systems and the time-varying nature of qubit generation rates, rendering traditional scheduling algorithms ineffective. Furthermore, while Deep Reinforcement Learning (DRL) offers a promising avenue for quantum network optimization, existing research primarily focuses on network-layer routing, overlooking the finer-grained entanglement scheduling problem subject to strict delay and resource constraints. To bridge this gap, we propose the Deep Quantum Request Scheduler (DQRS), a novel DRL-based algorithm designed to maximize service quality under these dual constraints. Specifically, DQRS integrates a spatio-temporal branching architecture and virtual action slots within an ActorCritic framework. This design enables robust scheduling with continuous actions while adaptively managing fluctuating qubit budgets. Extensive simulation results demonstrate that DQRS achieves up to 44 % performance improvement over traditional methods and a 4 % gain compared to state-of-the-art DRL benchmarks across various resource constraints and observability settings, validating the robustness of the proposed approach.
Quantum entanglement is the cornerstone of quantum network communication, yet the qubit resources required to establish it are severely limited. Moreover, users within the network impose strict delay constraints on entanglement requests due to various factors, such as the real-time requirements of upper-layer applications or the finite coherence time of userside quantum memories. Consequently, effectively scheduling scarce qubit resources to serve these time-sensitive requests presents a significant challenge. This difficulty is exacerbated by the partial observability of quantum systems and the time-varying nature of qubit generation rates, rendering traditional scheduling algorithms ineffective. Furthermore, while Deep Reinforcement Learning (DRL) offers a promising avenue for quantum network optimization, existing research primarily focuses on network-layer routing, overlooking the finer-grained entanglement scheduling problem subject to strict delay and resource constraints. To bridge this gap, we propose the Deep Quantum Request Scheduler (DQRS), a novel DRL-based algorithm designed to maximize service quality under these dual constraints. Specifically, DQRS integrates a spatio-temporal branching architecture and virtual action slots within an ActorCritic framework. This design enables robust scheduling with continuous actions while adaptively managing fluctuating qubit budgets. Extensive simulation results demonstrate that DQRS achieves up to 44 % performance improvement over traditional methods and a 4 % gain compared to state-of-the-art DRL benchmarks across various resource constraints and observability settings, validating the robustness of the proposed approach.
Quantum networks represent promising foundations for secure communication, distributed quantum computing, and advanced quantum-enabled services. However, their deployment faces practical challenges including limited quantum resources, short coherence times, and environmental disturbances. Effective end-to-end entanglement request scheduling is critical to addressing these challenges, as it directly affects resource utilization and network reliability. Although end-to-end entanglement service rate is one of the representative performance measures in the quantum networks, the explicit optimization of the rate under the realistic constraints is relatively unexplored in the previous studies. This paper proposes an offline reinforcement learning (RL)-based scheduling framework, employing a decision transformer integrated with graph attention networks, to specifically optimize service rates within practical operational constraints, such as the single-time-slot usage limitation of quantum links. Our approach adaptively leverages network topology and operational dynamics to enhance scheduling decisions. Simulation studies conducted on the NetSquid platform across four quantum network topologies demonstrate that our model consistently outperforms both a conventional rule-based method and a baseline offline RL method in terms of service rate, while preserving fidelity and maintaining delays within acceptable levels. These results confirm the effectiveness of the proposed method for practical quantum network management.
Quantum networks, leveraging the principles of quantum mechanics, enable unconditionally secure communication and are emerging as a critical component of next-generation infrastructure. However, the scarcity of quantum network resources results in contention among requests for shared links and qubits. Such contention prolongs processing times and degrades resource utilization efficiency, which in turn leads to imbalanced completion times between high- and low-priority requests. This paper addresses the request scheduling problem by proposing DRL-QN, a scheduling framework based on Reinforcement Learning (RL) that optimizes scheduling sequences, reduces the completion time of network requests, and improves fairness. Specifically, to help the agent better perceive request contention and the network environment, request characteristics and network resources are described as the state, enabling accurate scheduling decisions. Furthermore, to ensure a balanced learning objective, the agent’s training is guided by a multi-objective reward function that simultaneously promotes low completion time and fairness. Extensive simulations demonstrate that the proposed DRL-QN method significantly reduces network completion time by 16.22%–21.25% compared with existing schemes, while achieving fairness comparable to the fairness-optimal method and improving by approximately 6.79%–13.61% over other approaches.
Abstract Entanglement-based quantum networks require intelligent routing mechanisms that not only maximize fidelity and connectivity but also ensure service differentiation under varying traffic conditions. In this work, we propose a novel QoS-aware traffic modeling and scheduling framework integrated into a reinforcement learning-based quantum routing agent. Unlike prior approaches that assume deterministic or uniform request arrival patterns, our model introduces stochastic and bursty traffic scenarios using Erlang and ON/OFF distributions to emulate realistic quantum workloads. Each entanglement request is assigned a set of QoS attributes, including fidelity requirements, time-to-live constraints, and priority levels. We then extend the reward function of the agent to consider deadline compliance and resource efficiency in addition to fidelity. Through simulation, we compare the proposed method against state-of-the-art baselines, including DQRA and Proactive RL, under both best-effort and QoS-prioritized scheduling modes. Results demonstrate that our approach maintains higher success rates and lower latency for high-priority requests, while preserving entanglement fidelity and efficient resource allocation across varying network loads. This study highlights the importance of integrating QoS mechanisms into quantum routing to support scalable and differentiated quantum services
Abstract Quantum communication plays a key role in enabling distributed Noisy Intermediate-Scale Quantum (NISQ) systems; however, practical quantum networks are often affected by link fluctuations, limited quantum resources, and entanglement degradation. These factors make reliable routing and scheduling difficult, particularly in large-scale and dynamically changing environments. Classical methods such as Historical Weighted Fidelity Routing (HWFR) and Dynamic Multi-Priority Quantum Scheduling (DMPQS) generally perform well only when network conditions remain stable, and their performance degrades when entanglement success rates vary or relay nodes become congested. A major challenge therefore lies in operating under uncertain and time-varying fidelity conditions. Most existing approaches focus on either routing or scheduling in isolation and do not explicitly consider future fidelity variations, which can lead to inefficient resource usage and increased end-to-end latency. In this work, we propose a Deep Reinforcement Learning-based Quantum Communication Optimization Framework (DRL-QCOF) that jointly addresses routing and scheduling. The framework employs a Bi-LSTM-based Quantum Fidelity Prediction Network (QFPN) to estimate future link fidelity F_"pred" (t)from time-series features including link length, noise coefficient, entanglement success probability, and relay utilization. These predictions are then used by a Deep Reinforcement Learning-based Adaptive Quantum Scheduling Agent (DRAQS-Agent), implemented using an Actor–Critic architecture with a continuous reward formulation, to guide routing and scheduling decisions across distributed nodes. Simulation results indicate that the proposed approach improves the average link fidelity to a maximum of 0.88, compared to 0.78–0.86 achieved by existing methods. The end-to-end latency is reduced from 104 ms to 83 ms, while baseline schemes remain above 110 ms under the same conditions. In addition, the average response time is reduced to approximately 147 ms, resource utilization increases to 71–83%, and the success ratio improves from 93% to 97%, resulting in improved end-to-end reliability. The cumulative reward stabilizes at around 16.0, with convergence observed within 300 training episodes, and the achieved throughput reaches 118–121 entangled pairs per second, confirming the effectiveness of the proposed framework in dynamic NISQ network scenarios.
Quantum communications are gaining momentum in finding applications in a wide range of domains, especially those require high-security data transmissions. On the other hand, machine learning has achieved numerous breakthrough successes in various application domains including networking. However, currently, machine learning is not as much utilized in quantum networking as in other areas. With such motivation, we propose a machine-learning-powered entanglement routing scheme for quantum networks that aims to accommodate maximum numbers of demands (source-destination pairs) within a time window. More specifically, we present a deep reinforcement routing scheme that is called Deep Quantum Routing Agent (DQRA). In short, DQRA utilizes an empirically designed deep neural network that observes the current network states to schedule the network’s demands which are then routed by a qubit-preserved shortest path algorithm. DQRA is trained towards the goal of maximizing the number of resolved requests in each routing window by using an explicitly designed reward function. Our experiment study shows that, on averse, DQRA is able to maintain a rate of successfully routed requests above 80% in a qubit-limited grid network, and about 60% in extreme conditions i.e. each node can act as a repeater exactly once within a window. Furthermore, we show that the complexity and the computational time of DQRA are polynomial in terms of the sizes of the quantum networks.
In this paper, a novel Deep Q-Network (DQN) based scheduling method to optimize delay time and fairness among entanglement requests in quantum repeater networks is proposed. The scheduling of requests determines which pairs of end nodes should be entangled during the current time slot, while other pairs are placed in a queue for future slots. However, existing research on quantum networking often relies on simple statistical models to capture the behavior of quantum hardware, such as the failure rate of establishing entanglement. Moreover, current quantum simulators do not support network behaviors, including handling, pending, and dropping requests. To bridge the gap between quantum deployments and network behaviors, in this paper a dynamic network model is presented, encompassing quantum simulations, random topologies, and user modeling. The DQN based scheduling scheme allows us to balance the conflicting objectives of minimizing delay time and maximizing fairness among these entanglement requests. The proposed technique was evaluated using simulations, with results showing that the proposed DQN achieves higher performance compared to Greedy, Proportional fair and FIFO scheduling schemes.