Joint Data Scheduling and FEC Coding for Multihomed Wireless Video Delivery

TL;DR

This paper introduces JDAFC, a novel method combining path selection, dynamic data allocation, and fountain coding to enhance live video streaming over multihomed wireless networks, outperforming existing solutions.

Contribution

The paper presents a new joint data scheduling and fountain coding approach specifically designed for multihomed wireless video delivery, addressing bandwidth limitations of single-path streaming.

Findings

JDAFC improves peak signal-to-noise ratio in video streaming.

JDAFC reduces end-to-end delay in wireless video delivery.

Simulation results show JDAFC outperforms reference solutions.

Abstract

This paper studies the problem of mobile video delivery in heterogenous wireless networks from a server to multihomed device. Most existing works only consider delivering video streaming on single path which bandwidth is limited causing ultimate video transmission rate. To solve this live video streaming transmission bottleneck problem, we propose a novel solution named Joint Data Allocation and Fountain Coding (JDAFC) method that contain below characters: (1) path selection, (2) dynamic data allocation, and (3) fountain coding. We evaluate the performance of JDAFC by simulation experiments using Exata and JVSM and compare it with some reference solutions. Experimental results represent that JDAFC outperforms the competing solutions in improving the video peak signal-to-noise ratio as well as reducing the end-to-end delay.