QoS mapping of VoIP communication using self-organizing neural network

Author

Masugi, Masao

Author_Institution

Network Service Syst. Labs., NTT Corp., Tokyo, Japan

fYear

2002

fDate

2002

Firstpage

13

Lastpage

17

Abstract

The paper proposes a QoS mapping method for VoIP communications in real network environments. To take total account of the effects of several QoS-related parameters, we use a self-organizing neural network, which can map high-dimensional data into simple geometric relationships on a low-dimensional display. For training the self-organizing neural network, we measure several QoS-related parameters, such as PSQM (perceptual speech quality measure) and PSQM+, end-to-end delay, and packet loss rate, for three VoIP systems. Evaluation results confirm that our method can effectively evaluate the total QoS level composed of several QoS-related factors.

Keywords

Internet telephony; delays; integrated voice/data communication; learning (artificial intelligence); quality of service; self-organising feature maps; telecommunication computing; PSQM; QoS mapping; VoIP communication; end-to-end delay; high-dimensional data; low-dimensional display; packet loss rate; perceptual speech quality measure; self-organizing neural network; voice data integration; Costs; Delay estimation; Displays; IP networks; Internet telephony; Loss measurement; Neural networks; Protocols; Quality of service; Speech analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

IP Operations and Management, 2002 IEEE Workshop on

Print_ISBN

0-7803-7658-7

Type

conf

DOI

10.1109/IPOM.2002.1045749

Filename

1045749