Lingfen Sun

Voice quality prediction models and their application in VoIP networks

The primary aim of this paper is to present new models for objective, nonintrusive, prediction of voice quality for IP networks and to illustrate their application to voice quality monitoring and playout buffer control in VoIP networks. The contributions of the paper are threefold. First, we present a new methodology for developing perceptually accurate models for nonintrusive prediction of voice quality which avoids time-consuming subjective tests. The methodology is generic and as such it has wide applicability in multimedia applications. Second, based on the new methodology, we present efficient regression models for predicting conversational voice quality nonintrusively for four modern codecs (G.729, G.723.1, AMR and iLBC). Third, we illustrate the usefulness of the models in two main applications - voice quality prediction for real Internet VoIP traces and perceived quality-driven playout buffer optimization. For voice quality prediction, the results show that the models have accuracy close to the combined ITU PESQ/E-model method using real Internet traces (correlation coefficient over 0.98). For playout buffer optimization, the proposed buffer algorithm provides an optimum voice quality when compared to five other buffer algorithms for all the traces considered

QoE Prediction Model and its Application in Video Quality Adaptation Over UMTS Networks

The primary aim of this paper is to present a new content-based, non-intrusive quality of experience (QoE) prediction model for low bitrate and resolution (QCIF) H.264 encoded videos and to illustrate its application in video quality adaptation over Universal Mobile Telecommunication Systems (UMTS) networks. The success of video applications over UMTS networks very much depends on meeting the QoE requirements of users. Thus, it is highly desirable to be able to predict and, if appropriate, to control video quality to meet such QoE requirements. Video quality is affected by distortions caused both by the encoder and the UMTS access network. The impact of these distortions is content dependent, but this feature is not widely used in non-intrusive video quality prediction models. In the new model, we chose four key parameters that can impact video quality and hence the QoE-content type, sender bitrate, block error rate and mean burst length. The video quality was predicted in terms of the mean opinion score (MOS). Subjective quality tests were carried out to develop and evaluate the model. The performance of the model was evaluated with unseen dataset with good prediction accuracy ( ~ 93%). The model also performed well with the LIVE database which was recently made available to the research community. We illustrate the application of the new model in a novel QoE-driven adaptation scheme at the pre-encoding stage in a UMTS network. Simulation results in NS2 demonstrate the effectiveness of the proposed adaptation scheme, especially at the UMTS access network which is a bottleneck. An advantage of the model is that it is light weight (and so it can be implemented for real-time monitoring), and it provides a measure of user-perceived quality, but without requiring time-consuming subjective tests. The model has potential applications in several other areas, including QoE control and optimization in network planning and content provisioning for network/service providers.

New models for perceived voice quality prediction and their applications in playout buffer optimization for VoIP networks

Perceived voice quality is an important metric in VoIP applications. The quality is mainly affected by network impairments such as delay, jitter and packet loss. Playout buffer at the receiving side can be used to compensate for the effects of jitter based on a tradeoff between delay and loss. The main aim in this paper is to find an efficient perceived quality prediction method for perceptual optimization of playout buffer. The contributions of the paper are three-fold. First, we propose an efficient new method for predicting voice quality for buffer design/optimization. The method can also be used for voice quality monitoring and for QoS control. In the method, nonlinear regression models are derived for a variety of codecs (e.g. G.723.1/G.729/AMR/iLBC) with the aid of ITU PESQ and the E-model. Second, we propose the use of minimum overall impairment as a criterion for buffer optimization. This criterion is more efficient than using traditional maximum mean opinion score (MOS). Third, we show that the delay characteristics of voice over IP traffic is better characterized by a Weibull distribution than a Pareto or an exponential distribution. Based on the new voice quality prediction model, the Weibull delay distribution model and the minimum impairment criterion, we propose a perceptual optimization buffer algorithm. Preliminary results show that the proposed algorithm can achieve the optimum perceived voice quality compared with other algorithms under all network conditions considered.

Content Clustering Based Video Quality Prediction Model for MPEG4 Video Streaming over Wireless Networks

The aim of this paper is quality prediction for streaming MPEG4 video sequences over wireless networks for all video content types. Video content has an impact on video quality under same network conditions. This feature has not been widely explored when developing reference-free video quality prediction model for streaming video over wireless or mobile communications. In this paper, we present a two step approach to video quality prediction. First, video sequences are classified into groups representing different content types using cluster analysis. The classification of contents is based on the temporal (movement) and spatial (edges, brightness) feature extraction. Second, based on the content type, video quality (in terms of Mean Opinion Score) is predicted from network level parameter (packet error rate) and application level (i.e. send bitrate, frame rate) parameters using Principal Component Analysis (PCA). The performance of the developed model is evaluated with unseen datasets and good prediction accuracy is obtained for all content types. The work can help in the development of reference-free video prediction model and priority control for content delivery networks.

Perceived speech quality prediction for voice over IP-based networks

Perceived speech quality is the key metric for QoS in VoIP applications. Our primary aims are to carry out a fundamental investigation of the impact of packet loss and talkers on perceived speech quality using an objective method and, thus, to provide the basis for developing an artificial neural network (ANN) model to predict speech quality for VoIP. The impact on perceived speech quality of packet loss and of different talkers was investigated for three modern codecs (G.729, G.723.1 and AMR) using the new ITU PESQ algorithm. Results show that packet loss burstiness, loss locations/patterns and the gender of talkers have an impact. Packet size has, in general, no obvious influence on perceived speech quality for the same network conditions, but the deviation in speech quality depends on packet size and codec. Based on the investigation, we used talkspurt-based conditional and unconditional packet loss rates (which are perceptually more relevant than network packet loss rates), codec type and the gender of the talker (extracted from decoder) as inputs to an ANN model to predict speech quality directly from network parameters. Results show that high prediction accuracy was obtained from the ANN model (correlation coefficients for the test and validation datasets were 0.952 and 0.946 respectively). This work should help to develop efficient, nonintrusive QoS monitoring and control strategies for VoIP applications.

Content-Based Video Quality Prediction for MPEG4 Video Streaming over Wireless Networks

There are many parameters that affect video quality but their combined effect is not well identified and understood when video is transmitted over mobile/ wireless networks. In addition, video content has an impact on video quality under same network conditions. The main aim of this paper is the prediction of video quality combining the application and network level parameters for all content types. Firstly, video sequences are classified into groups representing different content types using cluster analysis. The classification of contents is based on the temporal (movement) and spatial (edges, brightness) feature extraction. Second, to study and analyze the behaviour of video quality for wide range variations of a set of selected parameters. Finally, to develop two learning models based on – (1) ANFIS to estimate the visual perceptual quality in terms of the Mean Opinion Score (MOS) and decodable frame rate (Q value) and (2) regression modeling to estimate the visual perceptual quality in terms of the MOS. We trained three ANFIS-based ANNs and regression based- models for the three distinct content types using a combination of network and application level parameters and tested the two models using unseen dataset. We confirmed that the video quality is more sensitive to network level compared to application level parameters. Preliminary results show that a good prediction accuracy was obtained from both models. However, the regression based model performed better in terms of the correlation coefficient and the root mean squared error. The work should help in the development of a reference-free video prediction model and Quality of Service (QoS) control methods for video over wireless/mobile networks.

A new method for VoIP quality of service control use combined adaptive sender rate and priority marking

Quality of service (QoS) control is an important issue in voice over IP (VoIP) applications because of the need to meet technical and commercial requirements. The main objective of this paper is to propose a new QoS control scheme that combines the strengths of adaptive rate and speech priority marking QoS control techniques to provide a superior QoS control performance, in terms of perceived speech quality. A second objective is to propose the use of an objective measure of perceived speech quality (i.e. objective MOS score) for adaptive control of sender behaviour as this provides a direct link to user-perceived speech quality, unlike individual network impairment parameters (e.g. packet loss and/or delay). Our results show that the new combined QoS control method achieved the best performance under different network congestion conditions compared to separate adaptive sender rate or packet priority marking method. Our results also show that the use of an objective MOS as the control parameter for the sender rate adaptation improves the overall perceived speech quality. The results reported here are based on a simulation platform that integrates DiffServ enabled NS-2 network simulator, a real speech codec (AMR codec) and the ITU-T standard speech quality evaluation tool (PESQ).

Video Quality Prediction Models Based on Video Content Dynamics for H.264 Video over UMTS Networks

The aim of this paper is to present video quality prediction models for objective non-intrusive, prediction of H.264 encoded video for all content types combining parameters both in the physical and application layer over Universal Mobile Telecommunication Systems (UMTS) networks. In order to characterize the Quality of Service (QoS) level, a learning model based on Adaptive Neural Fuzzy Inference System (ANFIS) and a second model based on non-linear regression analysis is proposed to predict the video quality in terms of the Mean Opinion Score (MOS). The objective of the paper is two-fold. First, to find the impact of QoS parameters on end-to-end video quality for H.264 encoded video. Second, to develop learning models based on ANFIS and non-linear regression analysis to predict video quality over UMTS networks by considering the impact of radio link loss models. The loss models considered are 2-state Markov models. Both the models are trained with a combination of physical and application layer parameters and validated with unseen dataset. Preliminary results show that good prediction accuracy was obtained from both the models. The work should help in the development of a reference-free video prediction model and QoS control methods for video over UMTS networks.

Video Quality Prediction Model for H.264 Video over UMTS Networks and Their Application in Mobile Video Streaming

Universal Mobile Telecommunication System (UMTS) is a third generation mobile communication systems that supports wireless wideband multimedia applications. The objective of this paper is to present a new model for non-intrusive prediction of H.264 encoded video quality over UMTS networks and to illustrate their application to video quality monitoring and adaptation in mobile wireless streaming services. First, we present an efficient regression model for predicting video quality non-intrusively for all content types. The model is predicted from a combination of a set of objective parameters in the application and physical layer in terms of the Mean opinion Score (MOS). The application layer parameters considered are the content type, sender bitrate and frame rate and the physical layer parameters are the block error rate modeled with 2-state Markov model for a mean burst length of 1.75. The performance of the proposed metric is evaluated with unseen dataset with good prediction accuracy. Second, we illustrate the application of the model in mobile streaming services by presenting a new Sender Bitrate (SBR) adaptation scheme at pre-encoding stage that is Quality of Experience (QoE) driven. The scheme was tested and evaluated in the NS2 based UMTS simulation network. Extensive simulation results demonstrate the effectiveness of the proposed adaptation scheme in terms of the MOS and especially at the UMTS network bottleneck access where perceived video quality is most affected. The proposed scheme was responsive to available network bandwidth and congestion and adapted the SBR accordingly maintaining acceptable quality in terms of the MOS. The proposed scheme has applications in network planning and content provisioning for network/service providers.

Learning models for video quality prediction over wireless local area network and universal mobile telecommunication system networks

Universal mobile telecommunication system (UMTS) is a third-generation mobile communications system that supports wireless wideband multimedia applications. The primary aim of this study is to present learning models based on neural networks for objective, non-intrusive prediction of video quality over wireless local area network (WLAN) and UMTS networks for video applications. The contributions of this study are two-fold: first, an investigation of the impact of parameters both in the application and physical layer on end-to-end video quality is presented. The parameters considered in the application layer are content type (CT), sender bitrate (SBR) and frame rate (FR), whereas in the physical layer block error rate (BLER) and link bandwidth (LBW) are considered. Secondly, learning models based on adaptive neural fuzzy inference system (ANFIS) are developed to predict the visual quality in terms of the mean opinion score for all contents over access networks of UMTS and WLAN. ANFIS is well suited for video quality prediction over error-prone and bandwidth restricted UMTS as it combines the advantages of neural networks and fuzzy systems. The ANFIS-based artificial neural network is trained using a combination of physical layer parameters such as BLER and LBW and application layer parameters of CT, SBR and FR. The proposed models are validated using unseen data set. The preliminary results show that good prediction accuracy was obtained from the models. This study should help in the development of a reference-free video prediction model and quality of service control methods for video over UMTS/WLAN networks.