This paper studies AQM deployment in a specific class of networks where routers/switches have a topological hierarchy, form acyclic paths, and adopt multipath routing and defines a Katz centrality-based metric to choose the most appropriate router for AQM deployment, and argues that doing so ensures the greatest stabilising effect.
Abstract
Excessive queueing delays constitute a significant impediment to latency-sensitive network applications. Although effective deployment of Active Queue Management (AQM) strategies has been proposed as a necessary solution, deployment remains sparse. This paper studies AQM deployment in a specific class of networks where routers/switches have a topological hierarchy, form acyclic paths, and adopt multipath routing. Our approach rests on the well-established premise that AQM deployment impacts packet-forwarding dynamics in networks carrying TCP flows, thus establishing a direct link between stability and network performance. We use fluid models for TCP and queue dynamics in the network, along with a simple threshold-based queue policy to outline a closed-loop model for the network. Stability analyses reveal that while the network is vulnerable to instability as the average round-trip time (RTT) of the TCP flows increases, it tolerates a much larger RTT without losing stability when the threshold-based AQM is deployed in an appropriate router. We then define a Katz centrality-based metric to choose the most appropriate router for AQM deployment, and argue that doing so ensures the greatest stabilising effect. Finally, packet-level simulations corroborate that the proposed deployment strategy ensures low-latency operation of the network.
This research bridges the gap between theoretical queuing models and practical routing strategies, contributing to the development of more efficient routing protocols for MANETs and demonstrates significant improvements in delay, throughput, routing overhead and node lifetime under realistic traffic conditions.
Due to cross-domain heterogeneity, unpredictable traffic patterns, and limited real-time visibility into network conditions, mobile backbone networks are increasingly experiencing performance degradation. This study presents a novel cross-domain AI-driven telemetry pipeline that facilitates intelligent, high-performance routing across the core network, transport, and radio access domains in order to address these issues. The suggested approach captures fine-grained network information, such as latency, link utilization, queue depth, and packet loss, in real time by combining streaming telemetry with a uniform cross-layer data aggregation paradigm. Using this telemetry, a lightweight AI-powered predictive routing engine forecasts congestion and uses adaptive path selection to dynamically improve routing choices. The suggested method greatly increases routing efficiency and network resilience by employing the Ant colony optimization (ACO) algorithm to provide improved proactive and context-aware traffic steering, in contrast to conventional reactive routing protocols. Within the mobile backbone, the innovation is found in the smooth integration of AI-driven decision intelligence and cross-domain telemetry. Experiments show significant gains in throughput stability, end-to-end latency, and resource usage, confirming the usefulness of the suggested framework for next-generation mobile networks.
A new Enhanced Intelligent-based Energy and Mobility, and Obstacle-aware Clustering (EIEMOC) protocol to control the network congestion while meeting End-to-End Delay (E2D) constraints in delay-constrained FANET applications.
J. Rajeswari, R. Kousalya· International Journal of Ele...· 0 citations
Starlink has become the largest commercial LEO satellite network, yet little is known about its internal queue management and bandwidth allocation mechanisms. Prior measurement studies have documented performance variations but lack the granularity to explain the underlying causes. We present the first microscopic characterization of Starlink's transmission behavior, using controlled measurements from multiple terminals to capture per-packet dynamics at microsecond precision. Our analysis uncovers several previously undocumented mechanisms. Starlink employs head-drop queuing rather than tail-drop, with capacities of approximately 1500 and 4000 packets on downlink and uplink, respectively. Bandwidth allocation is demand-driven, starting from a baseline of 100/30 Mbps on the downlink and uplink that ramps up by 3.4×/2× over 400 ms when flows sustain queue pressure. Active queue management aggressively induces packet loss to control queue occupancy, especially on the uplink. These mechanisms reset every 15 seconds during Starlink's reconfiguration cycle. We also find flow-level queuing that isolates latency between concurrent flows while coupling their loss on the downlink. These findings reveal that Starlink's queue management creates fundamentally different operating conditions than terrestrial networks.
Hendrik Cech, Nitinder Mohan, Jorg Ott· 1 citation
(English) The optimization of Transmission Control Protocol (TCP) congestion control algorithms in geosynchronous satellite networks remains a significant research challenge due to high propagation delays, limited bandwidth and dynamic link conditions in such environments. This thesis presents a comprehensive analysis of various TCP congestion control mechanisms and Active Queue Management (AQM) policies in Performance Enhancing Proxy (PEP) geosynchronous satellite environments.
The study evaluates three primary TCP variants CUBIC, YeAH and BBRv1 under different AQM configurations, including TBF, CoDel, FQ-CoDel and FQ-PIE. Using a TCP-Splitting Satellite Emulation Framework (TSEF), we analyze the interaction between these TCP congestion control algorithms and AQM techniques to identify optimal strategies for balancing throughput, latency and fairness. Results indicate that while TCP CUBIC remains a robust option in geosynchronous satellite networks, its effectiveness is enhanced with appropriate AQM policies such as FQ-PIE and FQ-CoDel, specially with ECN enabled. These AQMs demonstrate superior latency management and fair bandwidth allocation among different TCP flows.
This research contributes to the ongoing development of satellite communication protocols by offering empirical insights into TCP behavior in satellite environments and providing recommendations for optimizing congestion control and queue management strategies. The findings form the basis for improving real-world satellite internet performance, particularly in applications demanding low-latency and high-throughput connectivity.
(Català) L’optimització dels algorismes de control de congestió del Transmission Control Protocol (TCP) en xarxes satel·litàries geoestacionàries continua sent un repte de recerca significatiu a causa dels elevats retards de propagació, l’amplada de banda limitada i les condicions dinàmiques dels enllaços. Aquesta tesi presenta una anàlisi exhaustiva de diversos mecanismes de control de congestió TCP i polítiques de Active Queue Management (AQM) en entorns satel·litaris geoestacionaris basats en Performance Enhancing Proxy (PEP).
L’estudi avalua tres variants principals de TCP, CUBIC, YeAH i BBRv1 sota diferents configuracions d’AQM, incloent-hi CoDel, FQ-CoDel i FQ-PIE. Mitjançant la nostra plataforma d’emulació, el TCP-Splitting Satellite Emulation Framework (TSEF), s’analitza la interacció entre aquests algorismes de control de congestió i tècniques d’AQM per identificar estratègies òptimes per equilibrar l’utilització, la latència i l’equitat en l’ús de l’amplada de banda. Els resultats indiquen que, tot i que TCP CUBIC continua tenint un rendiment sòlid en xarxes satel·litàries geoestacionàries, la seva efectivitat es veu millorada amb l’aplicació de polítiques AQM adequades, com FQ-PIE i FQ-CoDel, especialment amb ECN activat. Aquests mecanismes d’AQM demostren una millor gestió de la latència i una distribució equitativa de l’amplada de banda entre els diferents fluxos TCP.
Aquesta recerca contribueix al desenvolupament continu dels protocols de comunicació per satèl·lit, oferint informació empírica sobre el comportament del TCP en entorns satel·litaris i proporcionant recomanacions per a l’optimització del control de congestió i la gestió de cues. Els resultats serveixen de base per millorar el rendiment real d’Internet per satèl·lit, especialment en aplicacions que requereixen una baixa latència i una alta velocitat de transmissió de dades.
(Español) La optimización de los algoritmos de control de congestión del Transmission Control Protocol (TCP) en redes satelitales geoestacionarias continúa representando un desafío relevante en el ámbito de la investigación, debido a los elevados retardos de propagación, el ancho de banda limitado y las condiciones dinámicas del enlace.
Esta tesis ofrece un análisis exhaustivo de diversos mecanismos de control de congestión de Transmission Control Protocol (TCP), así como de políticas de Active Queue Management (AQM), en entornos satelitales geoestacionarias que emplean Performance Enhancing Proxy (PEP). El estudio compara tres variantes principales del protocolo TCP: CUBIC, YeAH y BBRv1, bajo distintas configuraciones de AQM, incluyendo CoDel, FQ-CoDel y FQ-PIE. Mediante nuestra plataforma de emulación, el TCP-Splitting Satellite Emulation Framework (TSEF), se analiza la interacción entre estos algoritmos de control de congestión y las técnicas de AQM, con el objetivo de identificar estrategias óptimas que logren un equilibrio entre rendimiento, latencia y equidad. Los resultados evidencian que, aunque TCP CUBIC continúa siendo una opción robusta para redes satelitales geoestacionarias, su eficacia se ve significativamente mejorada cuando se emplean políticas de AQM adecuadas, como FQ-PIE y FQ-CoDel, especialmente con ECN activado. Estas últimas destacan por su capacidad para gestionar eficientemente la latencia y distribuir de manera justa el ancho de banda.
Esta investigación contribuye al avance de los protocolos de comunicación satelital, proporcionando evidencia empírica sobre el comportamiento del TCP en dichos entornos y formulando recomendaciones para optimizar tanto las estrategias de control de congestión como las de gestión de colas. Los hallazgos obtenidos constituyen una base sólida para la mejora del rendimiento en redes satelitales reales, en particular en aquellas aplicaciones que requieren baja latencia y alta capacidad de transmisión.