Solution for the "Grand Challenge on Bandwidth Estimation for Real-Time Communications"

[Duncan Thomson]

Regarding the Grand Challenge on Bandwidth Estimation for Real-Time Communications, posted here: ACM MMSys'21, Istanbul, Turkey

 

I believe we developed and published a fairly complete solution to that challenge over 20 years ago. 

 

The paper is available here: Dynamic Quality of Service and Adaptive Applications for a Variable Bandwidth Environment | The MITRE Corporation

 

The abstract:

 

DYNAMIC QUALITY OF SERVICE AND ADAPTIVE APPLICATIONS FOR A VARIABLE BANDWIDTH ENVIRONMENT

April 2000

Duncan Thomson, The MITRE Corporation

Dr. Nancy L. Schult, The MITRE Corporation

Dr. Mohammad Mirhakkak, The MITRE Corporation

 

 DOWNLOAD PDF (215.62 KB) 

This paper describes an approach for providing dynamic Quality of Service (QoS) support in a variable bandwidth network, which may include wireless links and mobile nodes. The QoS approach centers on the notion of providing QoS support at some point within a range requested by applications. The applications must be capable of adapting to the level of QoS provided by the network, which may vary during the course of a connection. The paper describes a new protocol called dynamic RSVP (dRSVP) that we have implemented to demonstrate and evaluate the dynamic QoS concept. The protocol and a new application programming interface (API) for this dynamic QoS are described. Finally, we briefly discuss our experience with adaptive streaming video and audio applications that work with the new protocol in a testbed network.

 

Also see:

IEEE Journal on Selected Areas in Communications ( Volume: 19, Issue: 10, Oct 2001)

Page(s): 1984 - 1997

Date of Publication: Oct 2001 

ISSN Information:

INSPEC Accession Number: 7076768

DOI: 10.1109/49.957313

Publisher: IEEE

 

[Zhixiong Niu]

Thanks for your information. We really appreciate your contributions in this topic  20 years ago.

 

Like TCP, it has been evolving for more than 2 decades. Many algorithms are developed for new reequipments, hardware, networks by different methods. It is still a very hot topic in both production and academia.

 

If you still interested in this BWE, welcome to participate this challenge and submit your models/algorithms.   

 

-- 

Best,

Zhixiong

--
You received this message because you are subscribed to the Google Groups "OpenNetLab Challenge" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opennetlab-chall...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/opennetlab-challenge/CO6PR09MB7349BB4F6CF5B666CB1FCD29AE709%40CO6PR09MB7349.namprd09.prod.outlook.com.