BISCAY: Practical Radio KPI Driven Congestion Control for Mobile Networks

2025-09-02

AI Summary

To address degraded mobile application performance caused by severe bandwidth fluctuations in cellular networks, this paper proposes an on-device, wireless KPI-driven congestion control scheme. We first leverage OpenDiag—a kernel-level diagnostic tool—to achieve millisecond-granularity, real-time wireless KPI collection on unrooted Android smartphones, doubling measurement resolution. We then design a KPI-aware bandwidth estimation layer that dynamically adjusts the TCP congestion window while maintaining full compatibility with standard protocol stacks. Deployed on commercial 5G devices, our scheme demonstrates, across diverse 4G/5G scenarios, over 90% reduction in both average and tail latency compared to BBR and CUBIC, with throughput maintained or improved. The core innovation lies in a lightweight, high-precision, on-device real-time congestion control framework that tightly integrates wireless KPIs into transport-layer decision-making.

Technology Category

• Machine Learning: Online Learning & Bandits
• Application Domains: Mobility, Driving & Flight
• Search and Optimization: Applications

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Security and Privacy: Large-scale security measurements
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Mobile application performance relies heavily on the congestion control design of the underlying transport, which is typically bottlenecked by cellular link and has to cope with rapid cellular link bandwidth fluctuations. We observe that radio KPI measurements from the mobile device chipset can be exploited for precise and timely measurement of available bandwidth on the cellular link. Building on this insight, we propose Biscay, a practical and radio KPI-driven congestion control system design for mobile networks. Biscay leverages OpenDiag, the in-kernel real-time radio KPI extraction tool we introduce in this paper, along with our KPI-based accurate bandwidth determination layer towards dynamically adjusting the congestion window to optimally use the available bandwidth while keeping delay to the minimum. Our solution is practical and deployable, as shown through our implementation of Biscay and OpenDiag on unrooted Android 5G phones. We extensively evaluate Biscay against different state-of-the-art congestion control designs including BBR and CUBIC with emulations driven by real measurement traces as well as real-world experiments spanning diverse 4G and 5G scenarios, and show that it provides significant average and tail delay improvements (typically over 90% reduction) while yielding better or similar throughput. These gains are enabled by 100% improvement in the granularity of on-device radio KPI measurements with OpenDiag compared to existing alternatives like MobileInsight.

Problem

• Optimizing congestion control for mobile networks
• Leveraging radio KPIs for precise bandwidth measurement
• Minimizing delay while maximizing throughput utilization

Innovation

• Leverages real-time radio KPI extraction tool
• Uses KPI-based bandwidth determination layer
• Dynamically adjusts congestion window for optimization