analyze quic protocols

This paper addresses the insufficient co-optimization of congestion control and queue delay in QUIC. We propose a novel joint congestion and delay control mechanism. Its core innovation is the first integration of precise one-way queue delay estimation—derived from round-trip time (RTT) differentials—into QUIC, combined with the bandwidth-awareness principle of TCP Westwood+, enabling early congestion response. The mechanism operates entirely within the QUIC endpoint stack, requiring no modifications to underlying network infrastructure. Experimental evaluation across diverse real-world and emulated network scenarios demonstrates that our scheme significantly reduces end-to-end latency (by up to 35% on average) and packet loss rate (by up to 80%) compared to Cubic, BBRv2, and NewReno, while sustaining over 90% link utilization. These improvements substantially enhance performance for real-time communication applications.

This paper addresses the challenge that conventional TCP applications cannot leverage QUIC’s advantages—such as built-in encryption, stream multiplexing, and rapid loss recovery. To bridge this gap, we propose and implement a lightweight TCP-over-QUIC streaming tunnel, built in userspace using Rust and the Quinn QUIC library. It transparently proxies TCP connections and maps them onto QUIC streams without requiring modifications to legacy TCP applications. To our knowledge, this is the first systematic performance evaluation of such a tunnel under realistic network impairments. Experiments show that under 20% packet loss, the tunnel achieves significantly higher throughput than native TCP and exhibits superior stability under high latency and packet reordering. In ideal networks, it incurs only acceptable overhead from encryption and stream scheduling. Our core contribution is the empirical validation of QUIC-based TCP tunneling as both feasible and practically effective, delivering a low-overhead, robust engineering implementation.

This study systematically investigates the Quality-of-Experience (QoE) impact mechanisms of the QUIC protocol in multi-client video streaming scenarios. Focusing on two representative use cases—video-on-demand (VoD) and low-latency live (LLL) streaming—we employ a trace-driven simulation framework to analyze cross-layer interactions between mainstream QUIC implementations (featuring congestion control algorithms including Cubic and BBR) and adaptive bitrate (ABR) strategies (BOLA, Pensieve). We empirically reveal, for the first time, that identical congestion control algorithms exhibit substantial performance variation across different QUIC implementations, leading to measurable discrepancies in key QoE metrics—namely, stall ratio, startup latency, and average bitrate. Building upon this insight, we propose a QUIC–ABR cross-layer co-optimization framework that jointly adapts congestion window feedback and bitrate selection decisions. Our approach achieves quantifiable QoE improvements: a 32% reduction in stall ratio and an 18% increase in average video quality.

Ambiguities in the IETF QUIC specification (draft-29) hinder precise implementation and complicate compliance verification. Method: This work presents the first comprehensive formal model of draft-29, built within the Ivy framework and integrating state-machine modeling, SMT-based constraint solving, and differential testing to automate compliance validation across seven mainstream QUIC client/server implementations. Contribution/Results: Leveraging formal reverse analysis, we systematically uncover specification ambiguities and propose actionable remediation paths. Our approach identifies multiple critical compliance violations across implementations and pinpoints several interoperability-affecting specification ambiguities—directly informing ongoing IETF standard revisions. The methodology establishes a scalable, end-to-end framework for protocol formal verification, bridging high-level specifications with executable conformance checks while supporting both automated bug detection and specification refinement.

This work investigates the real-world deployment evolution and configuration heterogeneity of the QUIC protocol among hyperscale providers (e.g., Meta, Google, Cloudflare) since its 2021 standardization. Method: Leveraging one month of passive backscatter traffic from the CAIDA Internet telescope, we propose a fully passive identification framework. It exploits protocol-agnostic features—including SCID first-appearance patterns, packet aggregation behavior, and length distributions—to construct cross-domain QUIC deployment fingerprints. We further integrate QUIC packet reverse parsing, SCID semantic analysis, per-packet statistical modeling, and active measurement validation. Contribution/Results: For the first time without active probing, we infer vendor-specific RTO policies, retransmission mechanism differences, and load-balancer topology; we also quantitatively estimate server scale and hierarchical deployment. This work establishes a novel paradigm for passive monitoring and infrastructure inference of large-scale encrypted protocols.

This work addresses the challenge posed by HTTP/3’s reliance on the QUIC protocol, whose encryption and connection migration capabilities hinder stateful middleboxes—such as NATs, rate limiters, and load balancers—from accurately identifying and tracking network flows, often causing functional failures. To overcome this limitation, the paper presents the first general-purpose framework enabling stateful middleboxes to support QUIC connection migration. By leveraging QUIC protocol semantics, synchronizing connection state across instances, and co-designing with middlebox logic, the framework achieves reliable tracking of encrypted flows. A prototype implementation demonstrates that the approach preserves full middlebox functionality while incurring less than 5% overhead in both throughput and latency, and remains stable even under aggressive connection migration rates up to 100 Hz.

This study addresses the uncertain applicability of the QUIC protocol in resource-constrained IoT environments by proposing, for the first time, a systematic framework for customized QUIC configurations tailored to IoT constraints. Through protocol pruning based on optional QUIC features and comprehensive parameter optimization, this work validates both feasibility and performance advantages via extensive experimental evaluation. The results demonstrate that such customized configurations effectively overcome resource bottlenecks while maintaining protocol efficiency. Furthermore, key findings have been incorporated into IETF standardization drafts, providing both theoretical foundations and technical specifications to support large-scale QUIC deployment across IoT ecosystems. This research bridges the gap between QUIC’s design assumptions and real-world IoT limitations, establishing a viable pathway for adopting modern transport protocols in constrained network scenarios.

This study addresses the gap in existing QUIC protocol security analyses, which predominantly focus on network traffic and neglect empirical validation of defense mechanisms within binary implementations. To bridge this gap, we propose BSISA—a novel methodology that uniquely integrates binary reverse engineering with system-level network traffic analysis—to evaluate the real-world defensive efficacy of four major QUIC server implementations against six distinct attack types. By employing multidimensional classifiers and attack scenario simulation, BSISA precisely identifies “declared but silent” defense functions and pinpoints critical code paths responsible for attack absorption. Experimental results demonstrate that BSISA achieves an overall accuracy of 45.8%, substantially outperforming single-modality approaches, and uncover severe availability risks in certain implementations—most notably, picoquic exhibits a failure rate exceeding 72% under specific attacks.

QUIC, as the transport layer of the next-generation Web stack (HTTP/3), natively provides security and performance improvements over TCP-based stacks. However, since QUIC provides end-to-end encryption for both data and packet headers, in-network assistance like Performance-Enhancing Proxy (PEP) is unavailable for QUIC. To achieve the similar optimization as TCP, some works seek to collaborate endpoints and middleboxes to provide in-network assistance for QUIC. But involving both host and in-network devices increases the difficulty of deployment in the Internet. In this paper, by analyzing the QUIC standard, implementations, and the locality of application traffic, we identify opportunities for transparent middleboxes to measure RTT and infer packet loss for QUIC connections, despite the absence of plaintext ACK information. We then propose PEMI as a concrete system that continuously measures RTT and infers lost packets, enabling fast retransmissions for QUIC. PEMI enables performance enhancement for QUIC in a completely transparent manner, without requiring any explicit cooperation from the endpoints. To keep fairness, PEMI employs a delay-based congestion control and utilizes feedback-based methods to enforce CWND. Extensive evaluation results, including Mininet and trace-driven dynamic experiments, show that PEMI can significantly improve the performance of QUIC. For example, in the Mininet experiments, PEMI increases the goodput of file transfers by up to 2.5$\times$, and reduces the 90th percentile jitter of RTC frames by 20-75%.

Existing speed measurement tools focus on peak throughput and poorly reflect users’ perceived responsiveness; emerging metrics such as “latency under load” show promise but their sensitivity to Active Queue Management (AQM) configurations remains unclear. Method: We empirically evaluate three mainstream AQM schemes—CoDel, FQ-CoDel, and SFQ—in a controlled network environment, systematically analyzing their impact on throughput and latency distributions, particularly latency under load. Results: AQM significantly alters speed test outcomes, with distinct latency-throughput trade-offs observed across algorithms under high load. Current measurement platforms, if uncalibrated for AQM, yield misleading latency estimates, undermining the reliability of policy and regulatory decisions. This study is the first to quantitatively characterize the structural impact of AQM on emerging speed metrics, providing critical empirical evidence to inform standardization of measurement tools and evidence-based network governance.