| IP.com Number | IPCOM000028322D |
|
|
|---|---|---|---|
| Dated | Jun 25, 2004 UTC | ||
| Size | 2 page(s) (25.6 KB) | ||
| Disclosed by |
|
||
| Country | Portugal |
|---|---|
| Language | English (United States) |
| Related Person(s) |
(CONTACT) Juergen Carstens |
| Copyright | SIEMENS AG 2004 |
This document was submitted to IP.com's Prior Art Database and this preview is designed to provide you with information regarding the contents of this document by displaying up to the first four pages of the document as scaled page renderings and displaying a limited amount of text which was extracted from the document on the Text Preview Tab.
To find out more on how to obtain the entire document, click the Download tab. There is a charge for downloading some Prior Art Database documents; please examine carefully whether you believe this document fills your needs before purchasing.
For more information about the Prior Art Database, visit the Learn section of this website. Thank you for visiting IP.com's Prior Art Database! You may wish to check out our Intellectual Property Library website before you leave.
S
VoIP Traffic Simulator Architecture
Idea: Elvino Rodrigues, PT-Lisbon; Nuno Silva, PT-Lisbon
A special software architecture permits to simulate N independent Voice over Internet Protocol (VoIP). The software architecture is based on a multithread and multiprocess approach, where each pair of threads represents a peer of VoIP conversation. Each peer of the conversation is located in one independent machine. Figure 1 depicts the main architecture.
The NTP server and NTP client blocks represent the processes responsible for the synchronization of the machines. The synchronization uses the protocol NTP (Network Time Protocol) and a proprietary algorithm that leads to a reduction of network delay and network asymmetry effects on the synchronization.
Two threads model each peer of the conversation. The master thread is responsible for sending the RTP (Real Time Protocol) voice packets, while the slave thread receives the voice packets from the other peer. The master thread at machine1 is also responsible for controlling the voice model. The voice model is based on artificial conversational speech concerning the standard ITU-T P.59 and coded accordingly a pre-defined codec. The codec used could be one of the following: G.723.1, G.729, G.728 and G.711.
The server and client major blocks represent a control layer, which allows to create the correct number of threads and to accomplish with the initial synchronization between peers. The server block is responsible for gathering...
Copyright © 2004-2010 IP.com. All Rights Reserved.