Pith. sign in

cs.NI

Networking and Internet Architecture

Covers all aspects of computer communication networks, including network architecture and design, network protocols, and internetwork standards (like TCP/IP). Also includes topics, such as web caching, that are directly relevant to Internet architecture and performance. Roughly includes all of ACM Subject Class C.2 except C.2.4, which is more likely to have Distributed, Parallel, and Cluster Computing as the primary subject area.

Papers reviewed in the last 7 days lead, then the papers readers actually read. Ranking is not a quality score.

sort pith recommended most recent

2-bit rectenna backscatters encrypted data on two polarizations

An 868 MHz energy harvesting antenna switches between co‑ and cross‑polarized OOK backscatter to secure BLE sensors.

· “A New Backscattering Dual-Polarized Rectenna for Wireless Power Transfer and IoT Applications”

open re-runnable review →
Figure from the paper

Single firmware handles bins of any depth with sub‑cm accuracy

CleanCity‑BinSense parameterizes fill‑level sensing to slash waste collection costs in developing‑city deployment.

· “CleanCity-BinSense: An IoT-Enabled Smart Waste Management System with Configurable Real-Time Fill Monitoring and Nearest-Neighbor Route Optimization”

open re-runnable review →
Figure from the paper

Autonomous agents negotiate shared resources across 6G operator domains

An AI-native extension of the GSMA Operator Platform balances latency and energy via decentralised multi-agent negotiation.

· “AI-Native Orchestration in the 6G Continuum: Evolving Operator Platforms with Agentic AI”

open re-runnable review →
Figure from the paper

AI agents reconcile NaaS and RAN policies for autonomous 6G

Proposed architecture interprets natural language intents to automate coordination of external service bookings with internal network…

· “Toward Fully Autonomous 6G Networks: AI-driven Operational Efficiency and Optimization”

open re-runnable review →
Figure from the paper

Federated fuzzy causal model boosts marine fault diagnosis to 87% F1

A framework that reasons about operating mechanisms and supports counterfactual 'what-if' queries achieves 98.98% AUROC and outperforms…

· “SeaCausal-FL: Federated Fuzzy Causal Learning for Maritime IoT Fault Diagnosis and Counterfactual Reasoning”

open re-runnable review →
Figure from the paper

Task-aware V2X broadcasts keep platoon collision rate at 0.99 percent

Latent scene compression beats JPEG and JPEG 2000 on hazard-relevant quality at 0.06 bits per pixel.

· “Task-Oriented Semantic Communication for Hazard Warning and Remote Operation in Connected Vehicle Platoons”

open re-runnable review →
Figure from the paper

Collision-time packet counts cut missed heavy hitters by 12%

A per-flow packet count, consulted only when a slot is about to be overwritten, lifts F1 from 0.84 to 0.91.

· “Programmable Data Plane Switch based Heavy Hitter Flow Detection using Packet-Count and Packet-Size features”

open re-runnable review →
Figure from the paper

Two-stage backpressure routing unifies interface choice and packet forwarding

By modeling each interface as a virtual subnode, the method jointly assigns packets to modalities and routes them across the network.

· “Biased Backpressure Routing for Multihop Wireless Networks with Heterogeneous Interfaces”

open re-runnable review →
Figure from the paper

Entropy + random forest catches SYN floods at 99.9987% accuracy

Per-flow SHAP explanations arrive at 1,953 flows per second, though false positives rise to ~5.8% under random splits.

· “XAI-SDN: An Explainable Entropy-Guided Machine Learning Framework for Real-Time DDoS Detection in Software Defined Networks”

open re-runnable review →
Figure from the paper

The paper presents an all-optical data center network architecture where underloaded…

A wavelength borrowing architecture for optical spine-leaf data center networks with a tunable borrowing degree B that balances…

· “A Wavelength Borrowing Architecture for Optical Data Center Networks - Extended Version”

open re-runnable review →
Figure from the paper

This paper tests whether a language model's own uncertainty can flag risky network…

Sampling-based predictive uncertainty usefully ranks translation risk for LLM-generated network configurations, and parameter-token versus…

· “Uncertainty Signals for Network Intent Translation: Risk Ranking and Ambiguity Localization”

open re-runnable review →
Figure from the paper

Urgency-aware routing lifts on-time delivery by up to 40 percent

A heuristic-seeded reinforcement-learning protocol reaches full-online reliability with a seventh of the interaction budget.

· “From Prior-Guided Heuristics to Deployable Agents: Accelerating Demonstration-Driven Reinforcement Learning for Deadline-Constrained Network Control”

open re-runnable review →
Figure from the paper

This paper proposes PRISM, a distributed multiple access protocol that lets user devices…

A distributed medium access protocol (PRISM) uses multi-agent deep reinforcement learning to exploit spatial semantic redundancies among…

· “A Semantic-Aware Multiple Access Scheme Leveraging Spatial Redundancy for Uplink-Dominant Network Services”

open re-runnable review →
Figure from the paper

Joint optimization cuts pose prediction error by 35% in ISCC

By jointly optimizing beamforming, model depth, and compute frequency, the proposed framework maintains accurate human pose tracking under…

· “CRB-Guided Sensing and Resource Allocation for Human Pose Prediction in Integrated Sensing, Communication, and Computation Systems”

open re-runnable review →
Figure from the paper

Preventive architecture hits 100% SLA for 5G slices

TriSLA blocks violations before deployment using predictive ML and explainable AI, outperforming reactive methods by 48.8 percentage points.

· “TriSLA: A Preventive and Closed-Loop SLA-Aware Architecture for Multidomain Decision-Making with Explainable Artificial Intelligence in 5G Networks”

open re-runnable review →
Figure from the paper

This paper shows that using flattened 2D maps instead of full 3D geometry for simulating…

Flattening 3D road geometry into 2D for V2X propagation simulation can eliminate occlusion intervals, shifting average signal loss by 10 dB…

· “Highly Detailed Simulation for Connected Automated Vehicle Cooperative Driving”

open re-runnable review →
Figure from the paper

Decision-driven status updates outperform age-based by 18–21%

Two-agent RL framework learns when to send compressed state and how to accept, reject, or offload tasks, achieving the highest on-time…

· “Update for Decisions, Not Freshness: Goal-Oriented Status Updating and Selective Offloading at the Network Edge”

open re-runnable review →
Figure from the paper

Contextual bandit cuts wearable energy by 26% without dropping latency

A cross-layer controller allocates coding redundancy and compute placement by physiological information value, achieving the lowest energy…

· “Physiological Information Reliability: Cross-Layer Adaptive Resource Allocation for Cardiovascular Sensing”

open re-runnable review →
Figure from the paper

browse all of cs.NI → full archive · search · sub-categories