Sudokshina Chatterjee discusses the voice, looking at how vocalists are trained, the different ways ...
Gives an insight into a variety of techniques used to uncover the needs and wants of their consumers...
This is the live video of 120 young student leaders in conversation with Jet Li at the Voice of Futu...
Lecture Series on Broadband Networks by Prof.Karandikar , Department of Electrical Engineering, IIT ...
What should governments do to fight climate change? The world is emitting huge amounts of greenhouse...
citeseer |
(0) (0 Votes)
|
Views: (1018) Date: (13-05-09) Pages: () |
Abstract: We analyze the effects of adding voice traffic to a packet switched network. We develop and validate separate simple and general analytical models for data and voice traffic and then combine them. Our analytical models assume no particular MAC layer and no priority of service. We analyze the effects on the average voice packet delay of the number of voice and data sessions, packaging of multiple voice frames in a single packet, contention at the MAC layer, and LAN utilization. Our first goal was to develop a model which can predict the number of real time voice sessions which can coexist with (TCP) data sessions. By appropriately parameterizing the model for narrowband voice codecs like G.723.1, we find some very interesting results. First, we find that hundreds of such voice sessions can share a 10 Mbps Ethernet link while maintaining appropriately low delays (below 30 ms). (See Figure 5) Next we see the surprising result that with as few as 5-10 simultaneous data sessions, the numbe...