Enhanced packet loss recovery for real time PC-based GigE vision AVI systems
by Khaled Taouil; Tarek Jellad; Zied Chtourou
International Journal of Communication Networks and Distributed Systems (IJCNDS), Vol. 14, No. 4, 2015

Abstract: GigE vision could be a viable solution for seamless and cost effective integration of machine vision in manufacturing processes. Transmission delays, inherent to streaming and lost packets protocol implemented within the GigE, often jeopardise the implementation of networked PC-based machine vision. This is related to GigE vision open standard handling data integrity and error control through control and stream protocols (GVSP and GVCP) that implement data resend functions. In this work, we study the effect of data loss on the real time performance of distributed GigE vision AVI systems. We also introduce an optimised error recovery function based on definition of packets of interest (POIs) that allows for selective lost packet recovery based on their pertinence to subsequent machine vision processing. This scheme alleviates the traffic overhead by avoiding non-useful retransmissions. This mechanism allows reducing the latency and improving the predictability of such systems.

Online publication date: Fri, 29-May-2015

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Communication Networks and Distributed Systems (IJCNDS):
Login with your Inderscience username and password:

    Username:        Password:         

Forgotten your password?


Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com