100 papers
This study demonstrates that the choice of graph construction technique significantly impacts IoT botnet detection performance, revealing that Gabriel graphs combined with Variational Autoencoders and Graph Attention Networks achieve the highest accuracy (97.56%) on the N-BaIoT dataset compared to other methods like k-NN and Shared Nearest Neighbor.
This paper introduces MCQUIC, a multicast extension for the QUIC transport protocol that enables scalable, secure, and private mass live content delivery over the internet by providing packet encryption, integrity verification, and automatic unicast fallback while remaining transparent to applications.
This paper presents a measurement-driven cross-layer diagnostic methodology for O-RAN that utilizes tail latency and radio-layer indicators to identify performance degradation and device-specific behaviors under varying distances and obstruction scenarios, offering practical "degradation flags" for network troubleshooting.
This paper formalizes the semantics of authoritative DNS response selection by proving that its expressive power is inherently bounded by protocol constraints, establishing a finite normal form and an intrinsic algebraic structure that enables principled reasoning about equivalence and composition across heterogeneous systems.
This paper introduces a topological model using Hodge decomposition to analyze complex serverless information flows, distinguishing between locally correctable components and persistent harmonic modes to derive practical remediation strategies like "dumping effects" that address structural inefficiencies without requiring complete architectural restructuring.
Q-StaR is a quasi-static routing scheme for Networks-on-Chip that leverages the N-Rank algorithm to extract long-term load trends from topology and traffic distribution, thereby guiding runtime route selection to significantly improve load balancing while preserving the simplicity and predictability of static routing.
This paper proposes the Dual-Graph Embedding with Transformer (DGET) framework, a supervised multi-task learning architecture combining Graph Neural Networks and Transformers, to efficiently solve the NP-hard resource allocation problem in hybrid RF-OWC IoT networks, achieving near-optimal scheduling with reduced Age of Information and lower computational complexity compared to traditional optimization methods.
This paper proposes PEPspace, a secure transport acceleration strategy for interplanetary networks that utilizes a Non-Transparent Secure Proxy (NTSP) architecture to split encrypted connections and combines rate-based congestion control, adaptive forward error correction, and optimized flow control to achieve stable, high-throughput, and low-latency data delivery across extreme-delay links.
This paper evaluates the performance of Delay Tolerant Network (DTN) routing protocols in a simulated Nepal earthquake rescue scenario, demonstrating their effectiveness in distress message transmission while highlighting critical trade-offs between reliability and resource utilization to guide future emergency communication designs.
This paper introduces GenCC, a framework that leverages large language models to automatically design optimized congestion control utility functions, achieving performance improvements of 37% to 142% over state-of-the-art protocols by utilizing generative code evolution or mathematical chain-of-thought strategies.
This paper introduces OAuthHub, a development framework that mitigates OAuth data overaccess by utilizing users' personal devices as intermediaries to enforce a centralized, just-in-time permission model, thereby significantly reducing both the code complexity and development time required for secure data sharing.
This paper addresses the failure of byte-sequence-based masked modeling in encrypted traffic classification by identifying a mismatch in inductive bias and proposing FlowSem-MAE, a protocol-native tabular masked autoencoder that leverages field-specific semantics and temporal patterns to significantly outperform existing methods with substantially less labeled data.
Tureis is a self-supervised, Transformer-based framework designed for edge deployment in smart homes that achieves fine-grained, multi-failure sensor localization by encoding heterogeneous sensor streams into compact features and using a masked reconstruction objective to identify faulty devices without requiring labeled data or human intervention.
This paper proposes a hierarchical Graph Attention Network (HR-GAT) model that leverages public deployment records to accurately estimate fine-grained spectrum demand across multiple cities, significantly outperforming existing baselines and providing regulators with actionable insights for efficient spectrum sharing and allocation.
Fly-PRAC is a novel packet recovery scheme that leverages algebraic relations among coded packets to recover corrupted data at intermediate nodes without full decoding, significantly outperforming existing methods like S-PRAC in terms of transmission efficiency and decoding delay, particularly in noisy environments.
This paper introduces Spyglass, a cost-effective spectrum sensor that utilizes a switched array and the Searchlite protocol to achieve single-shot, multi-channel Angle of Arrival (AoA) estimation with high accuracy in dense wireless environments.
This paper proposes a novel self-finetuning framework that enables Generative AI agents to autonomously learn continuous control for dynamic Radio Access Network slicing by distilling long-horizon experiences into model parameters via a bi-perspective reflection mechanism, thereby outperforming traditional Reinforcement Learning and standard LLM-based agents in sample efficiency and multi-objective optimization without relying on handcrafted reward signals.
This paper argues that achieving high-performance end-to-end data movement requires shifting focus from raw network bandwidth to a holistic hardware-software co-design approach, introducing the "Drainage Basin Pattern" to identify and resolve bottlenecks across six critical paradigms ranging from network latency to host-side factors.
FAST is an efficient scheduler designed to overcome the scalability and performance limitations of existing solutions for All-to-All(v) communication in dynamic Mixture-of-Experts workloads by addressing traffic skew and incast congestion while drastically reducing synthesis time on modern GPU clusters.
This paper proposes a cost-effective, open-source platform utilizing 1:10 scaled mini-cars equipped with ROS2 and an ETSI-compliant OScar stack to bridge the gap between simulation and field testing for cooperative, connected, and automated vehicle research, demonstrated through a validated intersection collision warning application.