Presentation is loading. Please wait.

Presentation is loading. Please wait.

New Models for Perceived Voice Quality Prediction and their Applications in Playout Buffer Optimization for VoIP Networks University of Plymouth United.

Similar presentations


Presentation on theme: "New Models for Perceived Voice Quality Prediction and their Applications in Playout Buffer Optimization for VoIP Networks University of Plymouth United."— Presentation transcript:

1 New Models for Perceived Voice Quality Prediction and their Applications in Playout Buffer Optimization for VoIP Networks University of Plymouth United Kingdom {L.Sun; E.Ifeachor}@plymouth.ac.uk Dr. Lingfen Sun Prof Emmanuel Ifeachor

2 ICC 2004, Paris France, 20-24 June 2004 2 Outline  Background  Speech quality for VoIP networks  Current status  Aims of the project  Main Contributions  Novel non-intrusive voice quality prediction models  Novel perceptual-based speech quality optimization (e.g. jitter buffer optimization) mechanism  Conclusions and Future Work

3 ICC 2004, Paris France, 20-24 June 2004 3 Background – Speech Quality for VoIP Networks  VoIP speech quality: end-user perceived quality (MOS), an important metric.  Affected by IP network impairments and other impairments.  Voice quality measurement: subjective (MOS ) or objective (intrusive or non-intrusive) SCN IP Network Gateway SCN: Switched Comm. Networks (PSTN, ISDN, GSM …) End-to-end Perceived speech quality Intrusive measurement Non-intrusive measurement MOS Reference speech Degraded speech

4 ICC 2004, Paris France, 20-24 June 2004 4 Current Status and Problems  Lack of an efficient non-intrusive speech quality measurement method  E-model (a complicated computational model)  Based on subjective tests to derive models/parameters, time- consuming and expensive. Only limited models exist  Lack of perceptual optimization control methods  only based on individual network parameters for buffer optimization and QoS control purposes  not perceptual-based optimization control

5 ICC 2004, Paris France, 20-24 June 2004 5 Aims of the Project IP Network Receiver Voice source Voice receiver Encoder Sender Packetizer Jitter buffer Decoder De- packetizer Non-intrusive measurement MOS End-to-end perceived voice quality (MOS)  To develop novel and efficient method/models for non-intrusive quality prediction,  To apply the models for perceptual-based optimization control ( e.g. buffer optimization or adaptive sender-bit-rate QoS control).

6 ICC 2004, Paris France, 20-24 June 2004 6 Novel Non-intrusive Voice Quality Prediction  Based on intrusive quality measurement (e.g. PESQ) to predict voice quality non-intrusively which avoids subjective tests.  A generic method which can be applied to audio, image and video. VoIP Network New model (packet loss, delay, codec …) Predicted MOSc PESQ E-model Measured MOSc delay MOS(PESQ) Reference speech Degraded speech Intrusive method (regression or ANN models) Non-intrusive method

7 ICC 2004, Paris France, 20-24 June 2004 7 New Structure to Obtain MOS c  PESQ can only predict one-way listening speech quality (expressed as MOS).  By a new combined PESQ/E-model structure, a conversational speech quality (MOSc) can be obtained as Measured MOSc. PESQ Delay model MOS  R  I e IeIe End-to-end delay E-model MOSc IdId Reference speech Degraded speech MOS (PESQ)

8 ICC 2004, Paris France, 20-24 June 2004 8 Regression based Models (1)  Nonlinear regression models are derived for I e based on PESQ/PESQ-LQ  Further combine I e with I d to obtain MOS c. MOS (PESQ) I e model IeIe E-model MOSc I d model IdId Delay (d) Codec Packet loss Reference speech Degraded speech Speech database Encoder Loss model Decoder Nonlinear regression model (I e model) Predicted I e PESQ/ PESQ-LQ MOS  R  I e Measured I e (a) (b)

9 ICC 2004, Paris France, 20-24 June 2004 9 Regression based Models (2)  I e can be modelled by a logarithm fitting function with the form of  Parameters for different codecs (PESQ) ParametersAMR(H)AMR(L)G.729G.723.1iLBC a16.6830.8621.1420.0612.59 b*10030.114.2612.7310.249.45 c14.9631.6622.4525.6320.42

10 ICC 2004, Paris France, 20-24 June 2004 10 Regression Models for AMR (12.2Kb/s) e.g. for AMR (12.2Kb/s), The goodness of fit is: SSE = 2.83 and R 2 = 0.998 MOS vs. packet loss and delay

11 ICC 2004, Paris France, 20-24 June 2004 11 Perceptual-based Buffer Optimization  Motivation:  only based on individual network parameters (e.g. delay or loss)  targeting only minimum average delay or minimum late arrival loss, not maximum MOS.  There is a need to design buffer algorithm to achieve optimum perceived speech quality.  Contribution  A perceptual-based optimization jitter buffer algorithm o Use regression based models for buffer optimization o Use a minimum impairment criterion instead of traditional maximum MOS score o A Weibull delay distribution based on trace analysis o A perceptual-based optimization of playout buffer algorithm

12 ICC 2004, Paris France, 20-24 June 2004 12 Impairment Function I m  Define: impairment function I m Playout delay d Weilbull distribution buffer loss  b

13 ICC 2004, Paris France, 20-24 June 2004 13 Minimum Impairment Criterion  Define: minimum impairment criterion Given:network delay d n, network loss  n and codec type Estimate: an optimized playout delay d opt Such that: minimize I m can be reached. d 1 d 2 d 3 d 4 Minimum I m

14 ICC 2004, Paris France, 20-24 June 2004 14 Perceptual-based Optimization Buffer Algorithm For every packet i received, calculate network delay n i If mode == SPIKE then if n i  tail*old_d then mode = NORMAL elseif n i > head*d i then mode = SPIKE; old_d = d i else -update delay records for the past W packets endif At the beginning of a talkspurt If mode == SPIKE then d i = n i else -obtain ( , ,  ) for Weilbull distribution for the past W packets -search playout d which meets minimum I m criterion endif

15 ICC 2004, Paris France, 20-24 June 2004 15 Performance Analysis and Comparison (1)  Selected five traces from UoP to CU (USA), DUT (Germany), BUPT (China), and NC (China).  Traces 1 and 3 with high delay variation and traces 2, 4, 5 with low delay variation TraceDelay (ms) Jitter (ms) Loss (%) 115316.21.1 2460.80.3 318619.514.3 4160.74.4 51500.2

16 ICC 2004, Paris France, 20-24 June 2004 16 Performance Analysis and Comparison (2)  “p-optimum” algorithm achieves the optimum voice quality for all traces.  “adaptive” algorithm achieves sub-optimum quality with low complexity.

17 ICC 2004, Paris France, 20-24 June 2004 17 Conclusions and Future Work  Conclusions  The development of a new methodology and regression models to predict voice quality non-intrusively.  Demonstrated the application of new non-intrusive voice quality prediction models to perceptual-based optimization of playout buffer algorithms.  Future Work  To consider buffer adaptation during a talkspurt in order to achieve the best trade-off between delay, loss and end-to-end jitter.  To extend the work to improve the performance of multimedia services (e.g. audio/image/video) over IP networks

18 ICC 2004, Paris France, 20-24 June 2004 18 Contact Details  http://www.tech.plymouth.ac.uk/spmc  Dr. Lingfen Sun L.Sun@plymouth.ac.uk  Prof Emmanuel Ifeachor E.Ifeachor@plymouth.ac.uk  Any questions? Thank you!


Download ppt "New Models for Perceived Voice Quality Prediction and their Applications in Playout Buffer Optimization for VoIP Networks University of Plymouth United."

Similar presentations


Ads by Google