arXiv subjects

Shyam Kumar Shrestha

Publications and source records attributed to Shyam Kumar Shrestha.

3 recordsLinked to original sources

arXiv2026

TCP BBR Performance over Wi-Fi~6: AQM Impacts and Cross-Layer Insights

We evaluate TCP BBRv3 on Wi-Fi 6 home networks under modern AQM schemes using a fully wireless testbed and a simple cross-layer model linking Wi-Fi scheduling, router queueing, and BBRv3's pacing dynamics. Comparing BBR Internet traffic with CUBIC across different AQMs (FIFO, FQ-CoDel, and CAKE) for uplink, downlink, and bidirectional traffic, we find that FIFO destabilizes pacing and raises delay, often letting CUBIC dominate; FQ-CoDel restores fairness and controls latency; and CAKE delivers the best overall performance by keeping delay low and aligning BBRv3's sending and delivered rates. We also identify a Wi-Fi-specific effect where CAKE's rapid queue draining, while improving pacing alignment, can trigger brief retransmission bursts during BBRv3's bandwidth probes. These results follow from the interaction of variable Wi-Fi service rates, AQM delay control, and BBRv3's inflight limits, leading to practical guidance to use FQ-CoDel or CAKE and avoid unmanaged FIFO in home Wi-Fi, with potential for Wi-Fi-aware tuning of BBRv3's probing.

cs.NI↗

arXiv2025

Understanding BBRv3 Performance in AQM-Enabled WiFi Networks

We present a modular experimental testbed and lightweight visualization tool for evaluating TCP congestion control performance in wireless networks. We compare Google's latest Bottleneck Bandwidth and Round-trip time version 3 (BBRv3) algorithm with loss-based CUBIC under varying Active Queue Management (AQM) schemes, namely PFIFO, FQ-CoDel, and CAKE, on a Wi-Fi link using a commercial MikroTik router. Our real-time dashboard visualizes metrics such as throughput, latency, and fairness across competing flows. Results show that BBRv3 significantly improves fairness and convergence under AQM, especially with FQ-CoDel. Our visualization tool and modular testbed provide a practical foundation for evaluating next-generation TCP variants in real-world AQM-enabled home wireless networks.

cs.NI↗

arXiv2024

Adapting Large Language Models for Improving TCP Fairness over WiFi

The new transmission control protocol (TCP) relies on Deep Learning (DL) for prediction and optimization, but requires significant manual effort to design deep neural networks (DNNs) and struggles with generalization in dynamic environments. Inspired by the success of large language models (LLMs), this study proposes TCP-LLM, a novel framework leveraging LLMs for TCP applications. TCP-LLM utilizes pre-trained knowledge to reduce engineering effort, enhance generalization, and deliver superior performance across diverse TCP tasks. Applied to reducing flow unfairness, adapting congestion control, and preventing starvation, TCP-LLM demonstrates significant improvements over TCP with minimal fine-tuning.

cs.NI↗