Title: Dynamic bandwidth estimation and congestion avoidance based on network traffic in mobile cloud

Authors: S.P. Tamizhselvi; M. Vijayalakshmi

Addresses: Department of Computer Science and Engineering, KCG College of Technology, Anna University, Chennai, 600097, Tamil Nadu, India ' Department of Information Science and Technology, College of Engineering, Anna University, Chennai, 600025, Tamil Nadu, India

Abstract: Smartphone faces many QoS challenges due to network traffic, bandwidth, congestion, delay and packet loss. We propose a novel framework to solve the issues, namely, network traffic-aware dynamic bandwidth estimation and congestion avoidance in the mobile cloud. Three algorithms in the cloud, namely, network bandwidth cloud estimation (NBCE) algorithm, cloud estimated queueing delay (CEQD) algorithm, and cloud bandwidth congestion avoidance (CBCA) algorithm. NBCE utilises the actual bandwidth based on different mobile network traffic. CEQD determines the delay to minimise the packet loss. Finally, CBCA reduces the congestion using estimated bandwidth and queue length. The algorithms were implemented in the public cloud Amazon Web Service (AWS) and evaluated. NBCE utilises the actual bandwidth of 88%, 80%, and 67% for different network traffic compared with existing TCP variants to provide improved network performance. CEQD minimises the average delay to 20 ms, and CBCA improves 5%, 4%, 4% goodput in other network traffic.

Keywords: network traffic; bandwidth estimation; queueing delay; congestion avoidance; mobile cloud; quality of service; packet loss rate; throughput; goodput; congestion window.

DOI: 10.1504/IJSNET.2022.122978

International Journal of Sensor Networks, 2022 Vol.39 No.1, pp.41 - 55

Received: 03 Oct 2020
Accepted: 12 May 2021

Published online: 19 May 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article