Buffer management for scalable video streaming

L Favalli,A Morales Figueroa

doi:10.4108/eai.13-9-2017.153337

Abstract

In a video streaming scenario, to cope with bandwidth variations, buffer management and intelligent drop packets policies play a critical role in the final quality of the video received at the user side. In this context, we present a buffer management strategy implemented at the source of a video co

Highlights

Delivering variable bit-rate video streaming over the Internet is a challenging task
In this paper we present a buffer management strategy with selective packet discard for Medium Grain Scalable (MGS) video encoded H.264 video sequences, based on two main blocks: 1)A Bandwidth Estimation (BE) approach based on Hidden Markov Model (HMM), where the states sequence simulates the bandwidth of the network and are matched with the rates of the MGS encoder
The available bandwidth can be modeled by N states, each one representing a certain level of availability

Summary

Introduction

Delivering variable bit-rate video streaming over the Internet is a challenging task. In this paper we present a buffer management strategy with selective packet discard for Medium Grain Scalable (MGS) video encoded H.264 video sequences, based on two main blocks: 1)A Bandwidth Estimation (BE) approach based on Hidden Markov Model (HMM), where the states sequence simulates the bandwidth of the network and are matched with the rates of the MGS encoder. When the source transmission buffer length exceeds a given threshold, the buffer management strategy takes place, dropping low priority packets and sending the most important packets fitting the available bandwidth.

Bandwidth Estimation Model

Hidden Markov Model

Buffer management for MGS video sequences

Simulations results

Findings

Conclusions