Novel approaches for online playout delay prediction in VoIP applications using time series models

Voice over IP (VoIP) applications requires a buffer at the receiver to minimize the packet loss due to late arrival. Several algorithms are available in the literature to estimate the playout buffer delay. Classic estimation algorithms are non-adaptive, i.e. they differ from more recent approaches basically due to the absence of learning mechanisms. This paper introduces two new formulations of adaptive algorithms for online learning and prediction of the playout buffer delay, the first one being based on the standard Box-Jenkins autoregressive model, while the second one being based on the feedforward and recurrent neural networks. The obtained results indicate that the proposed algorithms present better overall performance than the classic ones.     

Playout buffering of speech packets based on a quality maximization approach

To combat jitter problems in voice streaming over packet networks, playout buffering algorithms are used at the receiver side. Most of the proposed solutions rely on two main operations: prediction of delay statistics for future packets; setting of the end-to-end delay so as to limit or avoid packet losses. In recent years, a new approach has been presented, which is based on using a quality model to evaluate the impact of both packet loss and delay on the voice quality. Such a model is used to find the buffer setting that maximizes the expected quality. In this paper, we present a playout buffering algorithm whose main contribution is the extension of the new quality-based approach to the case of voice communications affected by bursty packet losses. This work is motivated by two main considerations: most of IP telephony applications are characterized by bursty losses instead of random ones; the human perception of the speech quality is significantly affected by the temporal correlation of losses. To this purpose, we make use of the extensions proposed in the ETSI Tiphon for the ITU-T E-Model so as to incorporate the effects of loss burstiness on the perceived quality. The resulting playout algorithm estimates the characteristics of the loss process varying the end-to-end delay, weights the loss and the delay effects on the perceived quality, and maximizes the overall quality to find the optimal setting for the playout buffer. The experimental results prove the effectiveness of the proposed technique.     

Adaptive playout scheduling algorithm tailored for real-time packet-based voice conversations over wireless ad-hoc networks

The effective provision of real-time, packet-based voice conversations over multi-hop wireless ad-hoc networks faces several stringent constraints not found in conventional packet-based networks. Indeed, MANETs (mobile ad-hoc networks) are characterized by mobility of all nodes, bandwidth-limited channel, unreliable wireless transmission medium, etc. This environment will surely induce a high delay variation and packet loss rate impairing dramatically the user experienced quality of conversational services such as VoIP. Indeed, such services require the reception of each media unit before its deadline to guarantee a synchronous playback process. This requirement is typically achieved by artificially delaying received packets inside a de-jitter buffer. To enhance the perceptual quality the buffering delay should be adjusted dynamically throughout the vocal conversation.In this work, we describe the design of a playout algorithm tailored for real-time, packet-based voice conversations delivered over multi-hop wireless ad-hoc networks. The designed playout algorithm, which is denoted MAPA (mobility aware playout algorithm), adjusts the playout delay according to node mobility, which characterizes mobile ad-hoc networks, and talk-spurt, which is an intrinsic feature of voice signals. The detection of mobility is done in service passively at the receiver using several metrics gathered at the application layer. The perceptual quality is estimated using an augmented assessment approach relying on the ITU-T E-Model paradigm while including the time varying impairments observed by users throughout a packet-based voice conversation. Simulation results show that the tailored playout algorithm significantly outperforms conventional playout algorithms, specifically over a MANET with a high degree of mobility.     

A Garch-based adaptive playout delay algorithm for VoIP

Network delay, packet loss and network delay variability (jitter) are important factors that impact on perceived voice quality in VoIP networks. An adaptive playout buffer is used in a VoIP terminal to overcome jitter. Such a buffer-control must operate a trade-off between the buffer-induced delay and any additional packet loss rate. In this paper, a Garch-based adaptive playout algorithm is proposed which is capable of operating in both inter-talkspurt and intra-talkspurt modes. The proposed new model is based on a Garch model approach; an ARMA model is used to model changes in the mean and the variance. In addition, a parameter estimation procedure is proposed, termed Direct Garch whose cost function is designed to implement a desired packet loss rate whilst minimising the probability of consecutive packet losses occurring. Simulations were carried out to evaluate the performance of the proposed algorithm using recorded VoIP traces. The main result is as follows; given a target Packet Loss Rate (PLR) the Direct Garch algorithm produces parameter estimates which result in a PLR closer than other algorithms. In addition, the proposed Direct Garch algorithm offers the best trade-off between additional buffering delay and Packet Loss Rate (PLR) compared with other traditional algorithms.     

HMM delay prediction technique for VoIP

This paper proposes a new algorithm for predicting audio packet playout delay for voice conferencing applications that use silence suppression. The proposed algorithm uses a hidden Markov model (HMM) to predict the playout delay. Several existing algorithms are reviewed to show that the HMM technique is based on a combination of various desirable features of other algorithms. Voice over Internet protocol (VoIP) applications produce packets at a deterministic rate but various queuing delays are added to the packets by the network causing packet interarrival jitter. Playout delay prediction techniques schedule audio packets for playout and attempt to make a reasonable compromise between the number of lost packets, the one-way delay and the delay variation since these criteria cannot be optimized simultaneously. In particular, this paper will show that the proposed HMM technique makes a good compromise between the mean end-to-end delay, end-to-end delay standard deviation and average packet loss rate.     

Adaptive playout scheduling and loss concealment for voice communication over IP networks

The quality of service limitation of today's Internet is a major challenge for real-time voice communications. Excessive delay, packet loss, and high delay jitter all impair the communication quality. A new receiver-based playout scheduling scheme is proposed to improve the tradeoff between buffering delay and late loss for real-time voice communication over IP networks. In this scheme the network delay is estimated from past statistics and the playout time of the voice packets is adaptively adjusted. In contrast to previous work, the adjustment is not only performed between talkspurts, but also within talkspurts in a highly dynamic way. Proper reconstruction of continuous playout speech is achieved by scaling individual voice packets using a time-scale modification technique based on the Waveform Similarity Overlap-Add (WSOLA) algorithm. Results of subjective listening tests show that this operation does not impair audio quality, since the adaptation process requires infrequent scaling of the voice packets and low playout jitter is perceptually tolerable. The same time-scale modification technique is also used to conceal packet loss at very low delay, i.e., one packet time. Simulation results based on Internet measurements show that the tradeoff between buffering delay and late loss can be improved significantly. The overall audio quality is investigated based on subjective listening tests, showing typical gains of 1 on a 5-point scale of the Mean Opinion Score.     

Delay reduction techniques for playout buffering

Receiver synchronization of continuous media streams is required to deal with delay differences and variations resulting from delivery over packet networks such as the Internet. This function is commonly provided using per-stream playout buffers which introduce additional delay in order to produce a playout schedule which meets the synchronization requirements. Packets which arrive after their scheduled playout time are considered late and are discarded. In this paper, we present the Concord algorithm, which provides a delay-sensitive solution for playout buffering. It records historical information and uses it to make short-term predictions about network delay with the aim of not reacting too quickly to short-lived delay variations. This allows an application-controlled tradeoff of packet lateness against buffering delay, suitable for applications which demand low delay but can tolerate or conceal a small amount of late packets. We present a selection of results from an extensive evaluation of Concord using Internet traffic traces. We explore the use of aging techniques to improve the effectiveness of the historical information and hence, the delay predictions. The results show that Concord can produce significant reductions in buffering delay and delay variations at the expense of packet lateness values of less than 1%     

Adaptive VoIP playout scheduling: assessing user satisfaction

Delay and packet loss dramatically affect the quality of voice-over-IP (VoIP) calls and depend on the playout buffer scheme implemented at the receiver. The choice of playout algorithm can't be based on statistical metrics without considering the perceived end-to-end conversational speech quality. The authors present a method for evaluating various playout algorithms that extends the E-model concept by estimating user satisfaction from time-varying transmission impairments. This article evaluates several playout algorithms and shows a correspondence between the authors' results and those obtained via statistical loss and delay metrics.     

Softwarization and virtualization of VoIP networks

Nowadays, voice over IP (VoIP) is a cost-effective and efficient technology in the communications industry. Free applications for transferring multimedia on the Internet are becoming more attractive and pervasive day by day. Nevertheless, the traditional, close, and hardware-defined nature of the VoIP networks’ structure makes the management of these networks more complicated and costly. Besides, its elementary and straightforward mechanisms for routing call requests have lost their efficiency, causing some problems, such as SIP servers’ overload. In order to tackle these problems, we introduce VoIP network softwarization and virtualization and propose two novel frameworks in this article. In this regard, we take advantage of the SDN and NFV concepts such that we separate data and control planes and provide the possibility for centralized and softwarized control of this network. This matter leads to effective routing. The NFV also makes this network’s dynamic resource management possible by functions virtualization of the VoIP network. The proposed frameworks are implemented in a real testbed, including Open vSwitch and Floodlight, examined by various scenarios. The results demonstrate an improvement in signaling and media quality in the VoIP network. As an example, the average throughput and resource efficiency increased by at least 28% and the average response time decreased by 34%. The overall latency has also been reduced by almost 39%.

     

Dynamic Video Playout Smoothing Method for Multimedia Applications

Multimedia applications including video data require the smoothing of video playout to prevent potential discontinuity. In this paper, we propose a dynamic video playout smoothing method, called the Video Smoother, which dynamically adopts various playout rates in an attempt to compensate for high delay variance of networks. Specifically, if the number of frames in the buffer exceeds a given threshold (TH), the Smoother employs a maximum playout rate. Otherwise, the Smoother uses proportionally reduced rates in an effort to eliminate playout pauses resulting from the emptiness of the playout buffer. To determine THs under various loads, we present an analytic model assuming the Interrupted Poisson Process (IPP) arrival. Based on the analytic results, we establish a paradigm of determining THs and playout rates for achieving different playout qualities under various loads of networks. Finally, to demonstrate the viability of the Video Smoother, we have implemented a prototyping system including a multimedia teleconferencing application and the Video Smoother performing as part of the transport layer. The prototyping results show that the Video Smoother achieves smooth playout incurring only unnoticeable delays.     

Evaluating perceived voice quality on packet networks using different random neural network architectures

Voice over Internet Protocol (VoIP) is one of the fastest growing technologies in the world. In VoIP speech signals are transmitted over the same network used for data communications. The internet is not a robust network and is subjected to delay, jitter, and packet loss. It is very important to measure and monitor the quality of service (QoS) the users experience in VoIP networks; this is not an easy task and usually requires subjective tests. In this paper we have analyzed three non-intrusive models to measure and monitor voice quality using Random Neural Networks (RNN). A RNN is an open queuing network with positive and negative signals. We have assessed the voice quality based on various parameters i.e. delay, jitter, packet loss, and codec. In our approach we have used the Mean Opinion Score (MOS) calculated using a Perceptual Evaluation of Speech Quality (PESQ) algorithm to generate data for training the RNN model. We have studied two feed-forward models and a recurrent architecture. We have found that the simple feed-forward architecture has produced the most accurate results compared to the other two architectures.     

VoIP中为提高语音质量所采用的关键技术

VoIP(Voice over IP)即IP电话,是将话音编码、压缩转换成数据包,在IP网络中进行传输的技术。为应对传统电话公司的竞争,IP电话的语音质量成为决定其未来命运的关键因素。该文首先介绍了几个界定QoS的参数和目前评价IP电话业务语音质量的三种模型：MOS模型、PSQM模型、E模型,然后重点介绍了在终端和网络上提高VoIP语音质量所采用的一些关键技术,应用于终端的技术中比较重要的是语音的编码与压缩、差错控制等,而应用于网络的技术则是解决IP QoS的两种基本模型：综合业务模型和区分业务模型。     

Neural and fuzzy computation techniques for playout delay adaptation in VoIP networks

Playout delay adaptation algorithms are often used in real time voice communication over packet-switched networks to counteract the effects of network jitter at the receiver. Whilst the conventional algorithms developed for silence-suppressed speech transmission focused on preserving the relative temporal structure of speech frames/packets within a talkspurt (intertalkspurt adaptation), more recently developed algorithms strive to achieve better quality by allowing for playout delay adaptation within a talkspurt (intratalkspurt adaptation). The adaptation algorithms, both intertalkspurt and intratalkspurt based, rely on short term estimations of the characteristics of network delay that would be experienced by up-coming voice packets. The use of novel neural networks and fuzzy systems as estimators of network delay characteristics are presented in this paper. Their performance is analyzed in comparison with a number of traditional techniques for both inter and intratalkspurt adaptation paradigms. The design of a novel fuzzy trend analyzer system (FTAS) for network delay trend analysis and its usage in intratalkspurt playout delay adaptation are presented in greater detail. The performance of the proposed mechanism is analyzed based on measured Internet delays.     

基于E-model的VoIP语音质量评估的研究

为准确评估VoIP）语音质量,对E—model算法进行了深入研究,剖析了E—model算法的组成部分Id,Is,Ie,A,探讨了丢包、延迟和抖动对VoIP质量的影响,并应用该算法对多种语音编码进行了评估。实验证明该客观评估算法主观与客观相关度高,有较强的适应性,可靠性,实用性,完全可用于VoIP语音质量评估。     

Coolstreaming: Design, Theory, and Practice

Peer-to-peer (P2P) technology has found much success in applications like file distributions and VoIP yet, its adoption in live video streaming remains as an elusive goal. Our recent success in Coolstreaming system brings promises in this direction; however, it also reveals that there exist many practical engineering problems in real live streaming systems over the Internet. Our focus in this paper is on a nonoptimal real working system, in which we illustrate a set of existing practical problems and how they could be handled. We believe this is essential in providing the basic understanding of P2P streaming systems. This paper uses a set of real traces and attempts to develop some theoretical basis to demonstrate that a random peer partnership selection with a hybrid pull-push scheme has the potentially to scale. Specifically, first, we describe the fundamental system design tradeoffs and key changes in the design of a Coolstreaming system including substreaming, buffer management, scheduling and the adopt of a hybrid pull-push mechanism over the original pull-based content delivery approach; second, we examine the overlay topology and its convergence; third, using a combination of real traces and analysis, we quantitatively provide the insights on how the buffering technique resolves the problems associated with dynamics and heterogeneity; fourth, we show how substream and path diversity can help to alleviate the impact from congestion and churns; fifth, we discuss the system scalability and limitations.     

Analysis of Skype VoIP traffic in UMTS: End-to-end QoS and QoE measurements

In the future Internet, multi-network services will follow a new paradigm in which the intelligence of the network control is gradually moved to the edge of the network. This impacts both the objective Quality of Service (QoS) of the end-to-end connection as well as the subjective Quality of Experience (QoE) as perceived by the end user. Skype already offers such a multi-network Voice-over-IP (VoIP) telephony service today. Due to its ease of use and a high sound quality, it becomes increasingly popular in the wired Internet.UMTS operators promise to offer large data rates which should suffice to support VoIP calls in a mobile environment. However, the success of those applications strongly depends on the corresponding QoE. In this work, we analyze the theoretically achievable as well as the actually achieved quality of IP-based voice calls using Skype. This is done performing measurements in both a real UMTS network and a testbed environment. The latter is used to emulate rate control mechanisms and changing system conditions of UMTS networks. The results show in how far Skype over UMTS is able to keep pace with existing mobile telephony systems and how it reacts to different network characteristics. The investigated performance measures comprise the QoE in terms of the MOS value and the QoS in terms of network-based factors like throughput, packet interarrival times, or packet loss.     

QoS-aware path switching for VoIP traffic using SCTP

Voice over Internet protocol (VoIP) has been a prevalent multimedia service nowadays. It allows us to transmit voice data over IP networks. However, quality of service (QoS) is a major challenge to VoIP services. It must provide similar quality to traditional public switched telephone network or cellular phone services. Therefore, QoS related protocols have become important for real-time applications. Multi-protocol label switch (MPLS) is one of the important techniques to improve the network performance from QoS point of view. It employs label swapping to speed up packet forwarding. However, when a large number of users utilize VoIP services, the network congestion issue still exists. It causes delay, jitter and packet loss that affect VoIP QoS. In this paper, we propose a QoS-aware path switching strategy by using stream control transmission protocol (SCTP) in MPLS network to improve the VoIP traffic. This was done by employing SCTP selective acknowledgment mechanism to report the transmission parameters of primary path and to determine the criteria to switch to backup path. Simulation results show significant improvement in VoIP QoS.     

Generation of synthetic video traffic using time series

Generating traffic has always been an important part of network simulations but has turned to an even more challenging task with modern networks. The statistical properties of the input stochastic processes traced in the networks used all along Information Era turned out to be complicated and difficult to reproduce. Taking into account successful efforts in modeling Internet traffic with FARIMA time series models, this paper attempts to extend their applicability and employ them to generate synthetic video traffic. It is known that FARIMA can model both the Short Range (SRD) and Long Range Dependence (LRD) existing in video traffic; however the traces it produces fail to describe correctly the moments (mean, standard deviation, skewness, kurtosis) of the distribution behind the data. Since an efficient traffic generator should capture both the statistical properties and queuing behavior of video traffic we experiment with models such as FARIMA with Student's t errors and FARIMA-GARCH with Normal and Student's t errors, improving somewhat the accuracy of the generated traffic. Furthermore, the paper suggests the projection of the traces generated by a FARIMA model to values of a Lognormal distribution. It is shown that such a methodology produces synthetic traces that can emulate very closely the behavior of real traces. In order to quantify closeness the generated traces are fed into a simple FIFO queuing system with finite buffers, where loss probability is calculated and compared to that experienced by the corresponding real traces. Using five different real traces, MPEG-4 or H.263, it is shown that the proposed methodology produces traffic generators that can capture satisfactorily several statistical properties of the real traffic and also its queuing behavior for a wide range of buffer sizes and service rates.     

Accurate resource estimation for homogeneous VoIP aggregated traffic

Modern VoIP codecs like G.729, G.723.1 or AMR can generate traffic during voice inactivity periods for Comfort Noise Generation (CNG). This feature alters the classical on–off pattern typically used to model the traffic generated by codecs with a Silence Suppression scheme. Therefore, the traffic generated due to CNG leads to severe inaccuracies in the dimensioning analysis done through traditional models based on multiplexing on–off sources like MMPP or fluid model.This paper addresses the VoIP dimensioning issue. First, we extend the traditional MMPP and fluid analytical models to include those traffic sources which perform the CNG feature. Second, we propose a simple but efficient algorithm which can be applied in dimensioning or admission control to find out the bandwidth reservation required to guarantee delay and loss in a packet-switch multiplexer node for VoIP traffic. Results are validated by simulations and VoIP traces and demonstrate a significant improvement in accuracy with respect to current on–off-based approaches.