Implementation and validation of an adaptive FEC machanism for video transmission

This research focuses on investigating the FEC mechanism as an error recovery over a wireless network.The existing adaptive FEC mechanism faces a major drawback, which is the reduction of recovery performance by injecting too many excessive FEC packets into the network. Thus, this paper proposes th...

Full description

Saved in:
Bibliographic Details
Main Authors: Ghazali, Osman, Harun, Nur Ziadah
Format: Article
Language:English
Published: Universiti Utara Malaysia 2012
Subjects:
Online Access:http://repo.uum.edu.my/7015/1/jict114.pdf
http://repo.uum.edu.my/7015/
http://jict.uum.edu.my/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This research focuses on investigating the FEC mechanism as an error recovery over a wireless network.The existing adaptive FEC mechanism faces a major drawback, which is the reduction of recovery performance by injecting too many excessive FEC packets into the network. Thus, this paper proposes the implementation of an enhanced adaptive FEC (EnAFEC) mechanism for video transmission together with its validation process.There are two propositions in the EnAFEC enhancement, which include block length adaptation and implementation, and suitable smoothing factor value determination.The EnAFEC adjusts the FEC packets based on the wireless network condition so that excessive FEC packets can be reduced.The proposed enhancement is implemented in a simulation environment using the NS-2 network simulation.The simulation results show that EnAFEC generates less FEC packets than the other types of adaptive FEC (EAFEC and Mend FEC).In addition, a validation phase is also conducted to verify that the proposed enhancement is functioning correctly, and represents a real network situation.In the validation phase, the results obtained from the simulation are compared to the outputs of the other adaptive FEC mechanisms.The validation results show that the mechanism is successfully implemented in NS2 since the number of packet loss falls under the overlapping confidence intervals.