A TCP Throughput Model Based on Gilbert Packet Loss Pattern

ZENG Bin; ZHANG Da-fang; LI Wen-wei; XIE Gao-gang

您当前的位置：

首页 >

文章列表页 >

A TCP Throughput Model Based on Gilbert Packet Loss Pattern

更新时间：2025-07-16

- A TCP Throughput Model Based on Gilbert Packet Loss Pattern
- Acta Electronica Sinica Vol. 37, Issue 8, Pages: 1728-1732(2009)
- 作者机构：
  
  1. 湖南大学计算机与通信学院,湖南,长沙,410082
  2. 湖南大学软件学院,湖南,长沙,410082
  3. 中国科学院计算技术研究所,北京,100190
  4. 湖南大学计算机与通信学院湖南长沙,410082
  5. 湖南大学软件学院湖南长沙,410082
  6. 中国科学院计算技术研究所北京,100190
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP393
- Published：2009
- 稿件说明：
移动端阅览
ZENG Bin, ZHANG Da-fang, LI Wen-wei, et al. A TCP Throughput Model Based on Gilbert Packet Loss Pattern[J]. Acta Electronica Sinica, 2009, 37(8): 1728-1732.
DOI：

ZENG Bin, ZHANG Da-fang, LI Wen-wei, et al. A TCP Throughput Model Based on Gilbert Packet Loss Pattern[J]. Acta Electronica Sinica, 2009, 37(8): 1728-1732. DOI：

摘要

丢包机制是推导TCP吞吐量模型的关键

直接影响模型的准确性.本文利用四状态Gilbert丢包机制来描述端到端路径上的丢包行为

对TCP的拥塞控制过程进行建模

在此基础上提出了一种更精确的TCP吞吐量模型.实验表明

改进的模型能较好的与实际值相拟合

可以更精确地预测实际TCP数据流的吞吐量性能.

Abstract

TCP throughput is one of the key concepts of computer network

and its estimation is very important for many research and application domains.Packet loss pattern is crucial to deduce a TCP throughput model

and can affect model's precision directly.Therefore

a new TCP throughput model GT(Gilbert Throughput)is presented by modeling a TCP congestion control mechanism based on Gilbert four-state model that is used to represent the packet lost behavior of end-to-end internet path.Experiment results show that GT matches the results better than Goyal model

and can predict the throughput of TCP flows more precisely in actual network.

关键词

Keywords

references

Views

1509

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

TCP-Shape:Study on a Renovation Method of Network Congestion Control Algorithms

Empowering In-Network Gray Failure Detection with Programmable Switches

Multipath Transmission Control Mechanism Based on Deep Reinforcement Learning and Sub-flow Coupling Perception

Cross-Layer Congestion Control for Wireless Sensor Networks

Related Author

CHENG Jing

SHEN Yong-jian

ZHANG Da-fang

LI Wen-wei

LIU Hong-yan

ZHANG Dong

WU Chun-ming

QIN Jiu-ren

Related Institution

College of Computer and Communication

College of Computer and CommunicationChangshaHunan 410082China

College of Computer Science and Technology, Zhejiang University

College of Computer and Data Science, Fuzhou University

State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunication

⁰