traffic engineering

To address the path optimization challenge for dynamic traffic engineering in software-defined networking (SDN), this paper proposes a real-time closed-loop control framework integrating deep reinforcement learning (DRL) with source routing. Methodologically, we design PolKA—a lightweight, P4-programmable source routing mechanism—and Hecate—a DRL-based system for real-time traffic analytics and path decision-making—achieving, for the first time, their coordinated closed-loop scheduling on a physical P4 testbed. Our key contribution is a data-plane-aware source routing integration paradigm that tightly couples path intelligence with programmable forwarding. Experimental results demonstrate a 37% reduction in end-to-end scheduling latency and a 52% decrease in link utilization variance, significantly enhancing network adaptability and operational controllability.

This study addresses the limitations of traditional approaches in efficiency and applicability for performance prediction of data flows in wired networks. It systematically reviews the decades-long evolution of network performance modeling, encompassing discrete-event simulation, queueing theory, network calculus, machine learning, and hybrid methods. The work innovatively proposes a unified taxonomy of modeling paradigms, revealing a paradigm shift from analytical and simulation-based techniques toward data-driven deep learning. It further provides a detailed analysis of how these approaches differ in evaluation objectives, underlying assumptions, and comparability. By clarifying the strengths, limitations, and appropriate application scenarios of each methodology, this research establishes a comprehensive reference framework to guide future advances in network performance modeling.

This work addresses the limitations of manually crafted static assume-guarantee contracts in cross-domain deterministic networking, which struggle to adapt to dynamic traffic and lack automated synthesis mechanisms. The paper presents the first framework that integrates large language models with formal verification to automatically synthesize both static and dynamic cross-domain contracts. Leveraging a large language model as a reasoning agent and combining network calculus, packet-level simulation, or real-world testbeds as verification oracles, the approach enables typed network modeling and precise determination of safe reconfiguration points. Evaluated on a TDM-PON infrastructure, the framework successfully generates dynamic 5G fronthaul contracts meeting a 100-microsecond deadline, achieving 3.5× higher bandwidth efficiency than static over-provisioning schemes with only 1.2 microseconds of latency error, while seamlessly supporting static 5G-TSN bridging scenarios.

Manual network configuration is error-prone and often violates standards, leading to misconfigurations and inconsistency. This paper proposes a UML-based model-driven approach: first, an extensible, semantically precise network configuration metamodel is defined to formally capture device capabilities, topological constraints, and change requirements; second, declarative model-to-command mapping rules and a code generator are developed to automatically translate high-level configuration models into vendor-specific CLI commands. To our knowledge, this is the first academically rigorous end-to-end automation framework that bridges UML modeling and production-grade network configuration. Evaluated in a real-world OSPF migration project at Shinshu University’s campus network, the approach fully automated the generation of all device configurations; post-deployment validation confirmed that network behavior strictly conformed to specifications—demonstrating correctness, practicality, and engineering feasibility.

Existing network simulation models neglect application-layer behavior, leading to traffic distortion and hindering robustness evaluation of monitoring and anomaly detection systems. To address this, we propose the first framework that models application-layer behavior as learnable and composable probabilistic processes. Specifically, it estimates probability density functions from real-world traffic traces and employs behavioral pattern convolution to generate dynamic, scalable, and realistic traffic. We further design a lightweight simulation engine supporting coexistence of multiple applications on a single machine and real-time behavioral modulation. Experimental results demonstrate that our approach significantly improves test coverage while achieving traffic distributions closely aligned with real-world scenarios. The open-source implementation has been validated on large-scale production networks.

This work addresses the challenge of automatically translating high-level service intents into effective Linux traffic control configurations, a task traditionally reliant on manual, low-level operations. The paper presents the first end-to-end framework that converts natural language or declarative intent specifications into standards-compliant Quality of Service (QoS) rules. The approach integrates queueing-theoretic semantic modeling, the LLaMA3 large language model, Active Queue Management (AQM)-guided prompting, and a rule-based validation mechanism to ensure correctness and compliance of the generated configurations. Experimental evaluation on 100 test intents demonstrates that LLaMA3 achieves a semantic similarity of 0.88 and a coverage of 0.87, outperforming baseline models by over 30%. Furthermore, AQM-guided prompting reduces output variability by a factor of three, significantly enhancing consistency and reliability.

This work proposes an end-to-end, reproducible supervised traffic flow classification framework that addresses the limitations of traditional port- or payload-based methods in the face of encrypted and increasingly diverse network traffic. The framework integrates practical considerations from real-world measurements, incorporating flow-based feature extraction, time-aware data splitting, leakage-proof experimental design, and interpretability analysis to mitigate common methodological pitfalls. Accompanied by an open-source Jupyter Notebook implementation, it provides a complete pipeline—from traffic capture and dataset construction to model training, evaluation, and deployment. Empirical validation on real-world encrypted traffic demonstrates the approach’s effectiveness, robustness, and practical deployability.

This study addresses energy efficiency optimization in communication networks during low-traffic periods by jointly optimizing network topology design and shortest-path routing. The approach ensures that all traffic demands can be satisfied within the activated subnetwork through dynamically adapted shortest paths. The authors propose, for the first time, a capacitated integer linear programming model that precisely captures dynamic shortest-path routing, complemented by provably effective strengthening constraints to accelerate solution convergence. A tailored column generation algorithm is developed to efficiently handle large-scale instances. Experimental results demonstrate that a simplified strategy—fixing routes and deactivating redundant links—achieves near-optimal performance, while the traffic-oblivious method TOCA exhibits superior efficacy in multi-demand scenarios.

Traditional wide-area network traffic engineering struggles to adapt to dynamic changes due to its centralized, periodic optimization mechanism, often resulting in minute-scale delays and suboptimal solutions. This work proposes OnlineTE, a distributed solver grounded in optimization decomposition theory that integrates centralized coordination with edge-driven execution. By enabling switches to immediately trigger re-optimization upon detecting link or traffic changes, OnlineTE achieves near-optimal path scheduling with second-scale responsiveness. The approach embeds multi-commodity load balancing (MLU) and max-flow problem formulations into a distributed algorithm, demonstrating in simulations with 750 nodes a tenfold performance improvement over the state-of-the-art while maintaining computational overhead well below the capacity of modern programmable switches.

This study addresses the lack of theoretical understanding regarding the interaction between congestion control algorithms (CCAs) and traffic policers in modern networks, which hinders the rational configuration of policing parameters. It is the first to systematically demonstrate that the interaction dynamics between CCAs and policers—whether based on virtual queues or token buckets—fundamentally differ from those with traffic shapers. The authors develop a formal analytical framework that integrates congestion control theory, virtual queue mechanisms, and token bucket models to derive precise configuration guidelines for key policing parameters, such as virtual queue capacity and assured rate thresholds. This work provides network operators with verifiable and tunable policing strategies, substantially enhancing the efficiency of network resource management.