Article

Characterizing voice and video traffic behavior over the Internet

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

In this paper, we present our research on characterizing voice and video traffic behavior in large-scale Internet Videoconferencing sys-tems. We built a voice and video traffic quality measurement testbed to collect Videoconferencing traffic traces from several sites all over the world that were connected to our testbed via disparate network paths on the Internet. Our testbed also featured the H.323 Beacon, an H.323 ses-sion performance assessment tool we have developed, and various other open-source and commercial tools. Our findings obtained by analyzing the collected traffic traces demonstrate the impact of: 1) end-point tech-nologies that use popular audio and video codecs and 2) network health status that is characterized by the variations of delay, jitter, lost and re-ordered packets in the network, on the end-user perception of audiovisual quality. The perceptual data used in our analysis includes both objec-tive and subjective quality measures. These measures were collected from our testbed experiments for a few sample tasks involving various levels of human interaction in Internet Videoconferences.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... A major issue is that this solution handles the listen and sleep periods, represented as x and y, in quite a conservative way. In particular, they prioritize QoS constraints, e.g., packets' delay, independent of the traffic identity, e.g., delayinsensitive traffic [23]- [25]. However, the trade-off between QoS requirements and desirable energy conservation results into shrinking energy saving even when this is not required. ...
... On the contrary, delay increases with the increase of the sleep period duration (d listen is fixed at 8 ms), since "wait-to-wakeup" time dominates the overall delay. In Fig. 7, d listen and d sleep are changed to cause delay of 10 up to 100 ms, which characterize voice and video traffic classes, correspondingly [23]- [25]. ...
... In particular, the results of Fig. 7 can be exploited as follows. Considering the first class of traffic, i.e., voice, the configuration (d listen , d sleep ) = (8 ms, 20 ms) seems to provide delay equal to 10 ms [23]- [25]. To verify this setup, we run the query Q 4 : This is the author's version of an article that has been published in this journal. ...
Article
Full-text available
Energy-efficient Passive Optical Networks (PONs) have gained significant interest since they are estimated to be the largest energy consumers among the wired access networks for the next ten years. In Ethernet PONs (EPONs), the equipment placed at the customer premises, i.e., the Optical Network Units (ONUs), has been shown to be responsible for almost 65% of the total EPON power consumption. Sleep mechanisms, implemented at ONUs’ side, can contribute to the EPONs’ energy-efficiency. However, the trade-off between energy saving and Quality of Service (QoS) requirements should be carefully tuned, especially when the downstream transmission is considered, to achieve the desirable results. In this paper, we propose a general framework that exploits formal methods as an approach for EPONs’ energy-efficiency. The idea is to build a holistic model representing both the state machine of the ONU in its details as well as the ONU’s communication with the Optical Line Terminal (OLT) under EPONs’ specifications. Verification results reveal that, guided by the QoS constraints and considering the transitions’ cost, an “aggressive” policy defining long sleep periods of the 100 ms order can be up to 24,9% beneficial for energy-efficiency compared to a “conservative” policy with sleep periods of the 10 ms order.
... Quality of Experience (QoE) is basically a subjective measure of end to end performance at the services level, from the point of view of the users. It has been shown in [5] and [6] that QoE is affected by network parameters like delay, jitter and loss. ...
... The E-model fundamentally addresses objective quality assessment for voice traffic. However, [5] and [6] show that reasonable correlation exists between the subjective quality assessment scores for audiovisual quality provided by the users and the objective quality assessment scores provided by video quality monitoring softwares like Telchemy VQMon tool that uses the Emodel and traces obtained from various videoconferencing tasks as an input for its analysis. Therefore, it is reasonable to use E-model as a tool for estimating QoE for audiovisual signal as well. ...
... The effect of jitter on QoE is much more pronounced than delay as evident from the gradient of the convergence plot. This trend was reported by other researchers before [5,6]. However, the methodology used by them to establish this claim was mostly experimental in nature. ...
Article
Multi-Protocol Label Switching (MPLS) technology has proven its worth for delivering new services while at the same time allowing migration from old to new generation networks. Avoidance of congestion is one of the major performance objectives of traffic engineering in MPLS networks. Load balancing can prevent the congestion caused due to inefficient allocation of network resources. Another important aspect in network performance is the end user perception of the quality delivered by the network called the Quality of Experience (QoE). The final arbiter of service performance is the end user whose opinion about quality is based on his or her perception. This end user perception of audiovisual quality is quantified by Mean opinion score (MOS). The network parameters that affect the MOS are delay, Jitter and loss. Though a number of multipath load balancing algorithms have been proposed in [1] and [2], none have proportioned traffic keeping the QoE constraint in mind. Here, a multipath load balancing algorithm is used to optimally split the incoming traffic based on the effect of average delay and jitter offered by the network so that the QoE measure of MOS is maximized. These initial results indicate that desirable QoE can be achieved by finite and small number of executions of an appropriate iterative load balancing algorithm once the step-size and the weights of the composite cost function representing combined effect of average delay and jitter are judiciously chosen.
... Audiovisual quality evaluation methods can be categorized into subjective and objective methods [9]. They are referred to as QoE measurement methods. ...
... After evaluating with T1-T3, one can see in Table 8 that MOS Audiovisual-MOBA3 as in (14) provides the lowest MAPE values when compared with other MOS Audiovisual models, while MOS Audiovisual-BM2 as in (12) provides the higher MAPE values. One can obviously see from Fig. 4, which shows the average MAPE values from Table 8, (8) or the MOS Audiovisual-BM1 model and (9) or the MOS Audiovisual-BM2 model provide average MAPE of 5.24% and 7.47% respectively. Meanwhile (12)- (14) or MOS Audiovisual-MOBA1 -MOS Audiovisual-MOBA3 models provide average MAPE values of 4.95%, 5.92% and 4.79% respectively. ...
Article
Full-text available
With the advancement of multimedia communications and information technologies, online video games have become very popular recreational activities worldwide, while Multiplayer Online Battle Arena (MOBA) games have gained immense popularity from game players in recent times. Therefore, based on quality of experience, audiovisual quality evaluation models for MOBA games have been proposed in this study. The subjective data from 210 game players (both males and females) about audio quality during conversation, the video quality, the overall quality, and the weighted percentage of video and voice have been collected from small competition events of the Garena ROV (Arena of Valor) in Hua Hin, Thailand. The data from 160 game players have been used to create the models, while the data from 50 game players have been applied for model performance evaluation. After developing the models, Mean Absolute Percentage Error (MAPE) technique has been utilized for model evaluation. On completion of this study, it was found that two thirds of the proposed models provide better performance than the other previous works. All of them, called MOSAudiovisual-MOBA1 model to MOSAudiovisual-MOBA3 model shows excellent performances with MAPE values of 4.95%, 5.92% and 4.75% respectively. Therefore, excellent performance of these proposed models becomes one of the contributions of this study. Also, this evidence of excellent model performance makes the proposed models ideal for utilization for assessment of audiovisual quality of the MOBA games. All audiovisual quality evaluation models provide excellent efficiencies.
... Audiovisual quality assessment methods can be divided into two main methods, subjective and objective [8]. Both of these are referred to as Quality of Experience (QoE) methods. ...
... years). It was used for MAPE calculation using (8), then the MAPE results were obtained, as shown in Table III, 2) This study did not follow [13][ [16][17], because those studies were based on objective assessment referring to network parameter (e.g., delay, loss and jitter) but this study was conducted based on subjective assessment. ...
... The investigation of packet delay is important for problems involving the mathematical modeling of network traffic, real-time applications [5], and Internet telephony and video, as well as for the study of network performance characteristics. For all such applications, network admin staff need to know the type of packet delay distribution occurring. ...
... (3) The truncated normal distribution is described by the following expression: (5) and j is network jitter or ip packet delay variation as follows [11]: ...
... This mechanism will ensure that each flow is following a specified seed, when its packets are generated, to reflect the realworld traffic and leads for high accuracy. To do that, a Seed Initializer method is used to initialize seeds for the variety streams based on the defined mathematical techniques in Calyam and Lee [17]. ...
... QoS max and QoS min are the effective deadline ranges for each flow j. Calyam and Lee [17] built a voice and video traffic measurement testbed to determine their effective deadline ranges. The Good range corresponds to delay values of 0-150 ms, the Acceptable range corresponds to delay values of 150-300 ms, while the Poor range corresponds to delay values > 300 ms. ...
Article
Full-text available
Packet networks are currently enabling the integration of traffic with a wide range of characteristics that extend from video traffic with stringent quality of service (QoS) requirements to the best-effort traffic requiring no guarantees. QoS guarantees can be provided in conventional packet networks by the use of proper packet-scheduling algorithms. As a computer revolution, many scheduling algorithms have been proposed to provide different schemes of QoS guarantees, with Earliest Deadline First (EDF) as the most popular one. With EDF scheduling, all flows receive the same miss rate regardless of their traffic characteristics and deadlines. This makes the standard EDF algorithm unsuitable for situations in which the different flows have different miss rate requirements since in order to meet all miss rate requirements it is necessary to limit admissions so as to satisfy the flow with the most stringent miss rate requirements. In this paper, we propose a new priority assignment scheduling algorithm, Hierarchal Diff-EDF (Differentiate Earliest Deadline First), which can meet the real-time needs of these applications while continuing to provide best-effort service to non-real time traffic. The Hierarchal Diff-EDF features a feedback control mechanism that detects overload conditions and modifies packet priority assignments accordingly.
... In Ethernet, one video frame is packetized in one Ethernet frame with sizes ranging from 65-1518 bytes. According to real measurements performed by [9,12], the most common size is ranging from 1025-1518 bytes. To be conservative, we will assume a video frame size of 1344 with a rate of 30 fps for our analytic and simulation work. ...
Article
Full-text available
The deployment of desktop videoconferencing, also known as video and voice over IP (VVoIP), over existing IP networks is gaining popularity. Such a deployment has become a major and challenging task for data network researchers and designers. An analytic approach for deploying videoconferencing is presented. The approach utilises queueing network analysis and investigates two key performance bounds for videoconferencing: delay and bandwidth. The approach can be used to assess the support and readiness of an existing IP network. Prior to the purchase and deployment of desktop videoconferencing equipment, the approach predicts the number of videoconferencing sessions or calls that can be sustained by an existing network while satisfying QoS requirements of all network services and leaving adequate capacity for future growth. As a case study, we apply our approach to a typical network of a small enterprise. In addition, we use OPNET network simulator to verify and validate our analysis. Results obtained from analysis and simulation are in line and give a close match.
... This mechanism will ensure that each flow is following a specified seed, when its packets are generated, to reflect the real-world traffic and leads for high accuracy. To do that, a Seed Initializer method is used to initialize seeds for the variety streams based on the defined mathematical techniques in [14] . ...
Article
Full-text available
Packet networks are currently enabling the integration of traffic with a wide range of characteristics that extend from video traffic with stringent QoS requirements to the best-effort traffic requiring no guarantees. QoS guarantees can be provided in conventional packet networks by the use of proper packet scheduling algorithms. As a computer revolution, many scheduling algorithms have been proposed to provide different schemes of QoS guarantees with Earliest Deadline First (EDF) as the most popular one. With EDF scheduling, all flows receive the same miss rate regardless of their traffic characteristics and deadlines. This makes the standard EDF algorithm unsuitable for situations in which the different flows have different miss rate requirements since in order to meet all miss raterequirements it is necessary to limit admissions so as to satisfy the flow with the most stringent miss rate requirements. In this study, we propose a new priority assignment scheduling algorithm, Hierarchal Diff-EDF (Differentiate Earliest Deadline First), which can meet the real-time needs of these applications while continuing to provide best effort service to non-real time traffic. The Hierarchal Diff-EDF features a feedback control mechanism that detects overload conditions and modifies packet priority assignments accordingly.
... max QoS and min QoS are the Effective Deadlines range for each flow j. Calyam and Lee in [22] had built a voice and video traffic measurement testbed to determine their effective deadline ranges. The Good range corresponds to delay values of (0-150) ms, the Acceptable range corresponds to delay values of (150-300) ms, while the Poor range corresponds to delay values > 300ms. ...
Article
Full-text available
Packet networks are currently enabling the integration of traffic with a wide range of characteristics that extend from video traffic with stringent QoS requirements to the best-effort traffic requiring no guarantees. QoS guarantees can be provided in conventional packet networks by the use of proper packet scheduling algorithms. As a computer revolution, many scheduling algorithms have been proposed to provide different schemes of QoS guarantees with Earliest Deadline First (EDF) as the most popular one. With EDF scheduling, all flows receive the same miss rate regardless of their traffic characteristics and deadlines. This makes the standard EDF algorithm unsuitable for situations in which the different flows have different miss rate requirements since in order to meet all miss raterequirements it is necessary to limit admissions so as to satisfy the flow with the most stringent miss rate requirements. In this paper, we propose a new priority assignment scheduling algorithm,Hierarchal Diff-EDF (Differentiate Earliest Deadline First), which can meet the real-time needs of these applications while continuing to provide best effort service to non-real time traffic. TheHierarchal Diff-EDF features a feedback control mechanism that detects overload conditions and modifies packet priority assignments accordingly. To examine our proposed scheduler model, we introduced our attempt to provide an exact analytical solution. The attempt showed that the solution was apparently very complicated due to the high interdependence between the system queues' service. Hence, the use of simulation techniques seems inevitable. The simulation results showed that the Hierarchical Diff-EDF achieved the minimum packet average delay when compared with both EDF and Diff-EDF schedulers.
... This period is enough for the prototype to reach a stable state. The experimental results indicated a maximum inter-arrival jitter of 10,38ms, a minimum of 0,85ms and an average of 7,03ms, resulting in good QoS [7]. ...
Article
Full-text available
The paper discuses how the imminent transition to the digital terrestrial television in UHF (DigitalSwitchover) could be employed towards enabling always-on connectivity and triple-play services access even from rural and dispersed locations, i.e. in areas where no termination/connection exists between the local PSTN exchanger and the optical fibre core backbone. Exploiting the European digital video broadcasting standard in regenerative configurations, the paper introduces an architecture that utilises the television stream as a common broadband infrastructure, capable to deliver not only custom digital television bouquets, but also and most predominant, to provide access to triple-play services. Utilising this television beam in backhaul (middle-mile) configurations it extends the core backbone to reach rural and dispersed locations, enabling therefore the immediate and cost-effective deployment (in these areas) of technologies that provide for always-on connectivity, such as WLAN, xDSL, etc.
... However, these traces reflect the individual frame sizes rather than the actual frame data, thereby excluding metrics that rely on actual frame data and playback. As a result, since video playback quality is strongly related to packet loss and end-to-end packet delays [15], the following objective measures, widely used in video content performance analysis [16][17][18][19][20][21], were employed: packet loss ratio (PLR), packet delay (PD), packet jitter, and minimum throughput. PLR is the number of corrupted, lost, or excessively delayed packets divided by the total number of packets expected at the video client station. ...
Article
WiMAX (Worldwide Interoperability for Microwave Access) embodies the IEEE 802.16 family of standards that provision wireless broadband access. With the IEEE 802.16e−2005 mobility amendment, WiMAX promises to address the ever-increasing demand for mobile high speed wireless data in fourth-generation (4G) networks. WiMAX market studies continue to project increased subscriber growth rates and planned carrier trials worldwide. Coupled with these increasing growth rates and higher WiMAX throughput rates, bandwidth intensive video on demand (VoD), Internet Protocol TV (IPTV), and mobile TV services are emerging in the forefront of the mobile arena.In this paper, we explore and derive optimum system level WiMAX parameters by quantifying network performance using such video-rich services. The video traffic will sufficiently load and stress the network to exploit the potential bandwidth, delay, and mobility limitations. We use the OPNET Modeler to engineer simulation sequences and explore the impact of channel bandwidth, time division duplex (TDD) frame size, advanced antenna systems support, and retransmission schemes using four objective performance metrics while streaming a feature-length movie to a Mobile WiMAX subscriber. The objective of this paper is to provide greater insight into Mobile WiMAX system performance using emerging, load intensive and delay sensitive media streaming services.
... The variability in the packet sizes of video traffic is dependent on the actual temporal and spatial nature of the video content being encoded. In Ethernet, one video frame is packetized in one Ethernet frame with sizes ranging from 65-1518 bytes.Figure 3 shows a histogram and the corresponding CDF of video packet (or Ethernet frame) size characteristics collected from well-known Internet testbed and reported in [17]. The video packet sizes characteristics correspond to an aggregated representation of video traffic from H.261, H.262 and H.263 video codec streams arising from desktop videoconferencing end-points. ...
Article
Abstract OPNET is a powerful,network,design and simulation tool that has gained,popularity in industry and academia. However, there exists no known simulation approach on how to deploy a popular real-time network,service such as videoconferencing. This paper demonstrates,how,OPNET can be leveraged,to assess the readiness of existing IP networks,to support,desktop,videoconference. To date, OPNET does not have built-in features to support videoconferencing or its deployment. The paper,offers remarkable,details on how,to model,and,configure OPNET for such a purpose. The paper,considers two,types of video traffic (viz. fixed and,empirical video packet,sizes). Empirical video packet sizes are collected from well-known,Internet traffic traces. The paper presents in-depth analysis and interpretation of simulation,results and shows how to draw,proper,engineering conclusions. r,2007 Published by Elsevier Ltd. Keywords: Network,Design; Videoconferencing; Simulation; OPNET ARTICLE IN PRESS www.elsevier.com/locate/jnca 1084-8045/$ - see front matter r,2007 Published by Elsevier Ltd.
... 6 shows that one of the outcomes of the queue instability is the high packet arrival time variance (i.e., jitter). While the results for backpressure scheduling are unacceptable [13], by using β = 1 and γ = 0.2, it is possible to decrease the interarrival distribution range by a factor of 2 compared to pure backpressure scheduling. If we compare this result to UDP case without scheduling, we observe that the inter-arrival distribution is more regular due to less fluctuations in the queues. ...
Conference Paper
Full-text available
Transmission scheduling to improve the capacity in wireless mesh networks (WMNs) is challenging. One promising candidate solution is backpressure scheduling, which was shown to provide throughput optimality and queue stability. Several recent practical systems that implement approximations of backpressure scheduling show performance benefits in WMNs that are based on IEEE 802.11, which is the most widely adopted MAC protocol. However, a detailed analysis of the queuing behavior is still missing for practical systems. In this paper, we experimentally show that backpressure scheduling over IEEE 802.11 presents significant queue instability due to the delayed interaction between MAC and network layers. To understand how and when queue instability occurs, we present a weighted backpressure scheduling (WBS), which uses two additional weight factors β and γ for the queue lengths of a node and its next-hop node when computing the backpressure value. By using WBS, we experimentally investigate the interactions between the components of backpressure scheduling implementations, i.e packet scheduling and link scheduling, and show that, usually, β >; γ leads to queue stabilization.
... Specifically, the quality of the one-way broadcast/multicast IP services was evaluated in terms of packet inter-arrival jitter, utilising a users/citizen accessing one IP-TV programme (MPEG-4, 1Mbps provided by the IP multicaster) via his UHF antenna. The experimental results indicated a maximum inter-arrival jitter of 10,38ms, a minimum of 0,85ms and an average of 7,03ms , resulting in good QoS [9].Figure 5a gives the graphical representation of the inter-arrival jitter during a 60 seconds evaluation period. This period is enough for the prototype to reach a stable state. ...
Article
Full-text available
This article introduces and validates a concept for the realization of a fusion IP/DVB networking environment that supports the provision of urban and rural always-on connectivity and triple-play services. Specifically, the proposed fusion environment realizes a unified infrastructure that changes traditional, passive urban users into active, information society participants capable of creating, manipulating, and distributing their own content/services over a commonly exploited infrastructure. Further, the environment enables triple-play services and always-on connectivity in rural areas, that is, in regions where there is no termination of the core backbone to the local PSTN/ISDN exchanger. Finally, the article describes important directions for future research to follow to exploit the proposed IP/DVB unified platform for the synergy of emerging broadcasting, telecommunications, wireless, and wired technologies, to alleviate the digital divide that currently exists not only among countries but also within most countries
... The literature also includes exhaustive analyses of video traffic over Internet in streaming systems [11], conferencing systems [12], or IPTV systems [13] where packet loss rates and patterns are detailed. These patterns can be used to emulate packet loss in further experiments under realistic conditions with the support of a suitable packet loss model. ...
Article
Full-text available
Quality of experience is of critical importance in streaming video services, because the traditional quality of service cannot represent the quality perceived by viewers. This work evaluates several objective quality metrics under realistic bursty packet loss conditions in the network, with the support of a packet loss model. Alignment of reference and streamed video sequences (with different levels of spatial-temporal information) are also explored as a technique to prevent inaccurate computation of objective metrics due to frame loss. Finally, the correlation between subjective and objective metrics for each motion level and the computing time of metrics are analysed. The most suitable objective metrics to characterize the real degradation in the quality perceived by viewers, for both off-line and real-time assessment, are proposed. The integration of motion, busty packet loss, sequence alignment after frame loss and computing time of metrics are the main contributions of this research work.
Article
Low rate TCP DoS (Shrew) and reduction of quality (RoQ) attacks have been proven to be detrimental to TCP traffic in the Internet. In this paper, we investigate the effect of these attacks on QoS sensitive, real time UDP traffic, like VoIP. Both of these attacks are difficult to detect, as the average attack traffic is low. These stealthy RoQ attacks are especially hard to detect as their periodicity is not well-defined, and the attacks can contain spoofed source as well as destination IP addresses. Extensive ns2 simulations have been performed under realistic scenarios to validate the effects of these attacks on real time VoIP traffic using the G711 and G729a audio codecs. Forward Error Correction (FEC) is also considered in our analysis to recover packets from the packet loss caused by attacks. Without FEC, these attacks can drastically reduce the quality of VoIP traffic, leading to denial of service. To evaluate the perceptual quality of voice under attacks, the network losses are mapped to user's quality perception by using the ITU E-model. The MOS range of the E-model decreases from best to medium due to the low rate TCP DoS and RoQ attacks even when FEC is deployed. In some of the attack scenarios, the packet losses are bursty, which can lead to dropped calls. In addition, the stealthy RoQ attacks with higher burst lengths are as effective as the shrew attack in impairing VoIP traffic. We have also evaluated the effects of the attack on the VoIP traffic on the DETERLAB testbed, which has specifically been built for conducting network security research, and the results are consistent with the ns2 simulation results. We believe our investigation is important to remove the misconception about low rate TCP DoS and RoQ attacks that they affect only TCP traffic in the Internet. Copyright © 2008 John Wiley & Sons, Ltd.
Conference Paper
In this present work, we propose a scheduling approach for heterogeneous traffic over network. The proposed scheduling not only satisfies the QoS requirements of the real time flows (class-1) but also continues providing best effort service to non real time flows (class-2). Many scheduling techniques have been proposed to respond to the temporal requirements of real-time jobs (flows). The technique using the algorithm Earliest Deadline First (EDF) is most used because the optimality of this algorithm has been proved for a number of criteria. However, this algorithm knows a complexity that increases with the size of queue. In order to not penalize the flows of class-2, we propose an approach for scheduling non-preemptive basis in two queues. The first is a hybrid queue for class-1 that combines the two policies First in First out (FIFO) and EDF to reduce the complexity of EDF. The second queue for class-2 is scheduled by FIFO policy. The performances of the proposed approach are evaluated by the miss deadline of the packet in the hybrid queue the class-1 and the average time of residence of packet in the FIFO queue of class-2. The simulation results underline that the existing errors are relatively weak which confirm the effectiveness of this approach.
Conference Paper
Packet networks are currently enabling the integration of traffic with a wide range of characteristics that extend from video traffic with stringent QoS requirements to the best-effort traffic requiring no guarantees. QoS guarantees can be provided in packet networks by the use of proper packet scheduling algorithms. In this paper, we propose a new priority assignment scheduling algorithm, hierarchal diff-EDF (differentiate earliest deadline first), which can meet the real-time needs of such applications while continuing to provide best effort service to non-real time traffic. The hierarchal diff-EDF features a feedback control mechanism that detects overload conditions and modifies packet priority assignments accordingly. The simulation results show that the hierarchical diff-EDF achieved the minimum packet miss ratio when compared with both EDF and diff-EDF schedulers.
Article
Various techniques have been proposed to achieve good QoS for diverse application types at network nodes. In order to supply the various QoS requirements for different kinds of applications, new scheduling policies need to be developed and evaluated. Many scheduling techniques have been proposed to respond to the temporal requirements of real-time flows. We propose a new scheduling Hierarchical Hybrid EDF/FIFO for network service, which considers the real time flow have higher priority to service. The technique uses the algorithm Earliest Deadline First (EDF) because the optimality of this algorithm has been proved for a number of criteria. However, this algorithm knows a complexity that increases with the size of queue. Our approach for scheduling is non-preemptive and based in two queues. The first queue for real time flow (class-1) combines the two policies EDF and First in First out (FIFO) to reduce the complexity of EDF. The second queue for non real time flow (class-2) is scheduled by FIFO policy. The work complexity of our approach is less or equal than O(n log 2 n) with n is the length of the hierarchical queue. The simulation results underline the effectiveness of this approach.
Article
Providing Quality-of-Service (QoS) guarantee requires for each router on the path of a traffic flow not to violate the flow's delay budget allocated to itself. Since the amount of load being offered to the router is determined by the budget, some imbalance in load among routers on the path may be alleviated by means of adjusting the budget. The equal allocation applied to the resource reservation protocol (RSVP) is simple to implement, but it has the drawback of a poor resource utilization. A load balancing method in which the delay budget being allocated to a router depends on its load state was developed to improve the drawback, but it's too complex to apply to the RSVP. This paper proposes an intra-path load balancing method not only applicable to the RSVP but also more effective in improving the drawback. The proposed method first partitions the end-to-end delay bound of a flow to routers by the RSVP and then let them adjust their budgets according to both the bottleneck state of the path and their links' bandwidth availabilities. The results of the simulation applying the proposed method to an evaluation network showed that the proposed method may provide the gain of 4 17 % compared to that in the legacy one in terms of the number of maximally admittable flows.
Conference Paper
Full-text available
This paper presents an analysis of the effects of multiple Quality of Experience (QoE) parameters applicable to the Network domain. The aim was to investigate the best candidate parameters for model development. The Taguchi orthogonal array method was used to achieve this aim by enabling a reduction in the number of trials required in our experiments whilst maintaining accuracy of the results. This method provided us with an analysis technique to identify the most and least influential parameters for QoE. We were also able to find relationships between these different parameters. The Quality Function Deployment (QFD) method was used to verify the results obtained by the Taguchi method. We conclude that packet reorder, jitter and packet loss are the most important parameters. These parameters are almost equally important for perceptive video quality. Packet reordering emerges as the most influential parameter but statistically all three are equal. Delay was found to be the least influential parameter. Overall perceptual quality is generally adversely affected by network parameters than content domain parameters.
Article
A key element in any system based on several interconnected computing and/or storage nodes is the interconnection network. Currently, one of the main concerns of high-speed interconnection network designers is how to improve network performance while using the minimum number of network resources. In that sense, in this paper we describe an efficient switch architecture suitable for any interconnect technology implementing deterministic source-based routing. This switch architecture uses the same network resources to provide two issues that improve network performance: Congestion Management and QoS support. We also present results to compare the effectiveness of this architecture to those of other proposals typically used to provide these issues in this context. These results have been obtained for synthetic traffic and for traces from parallel benchmarks and video frames. From the results, we can conclude that in any traffic scenario, our proposal is as effective as the previous ones, while requiring fewer resources and thus being much more cost-effective.
Conference Paper
Full-text available
Test-bed based research is an important aspect in mobile ad hoc networks (MANETs) and especially in vehicular ad hoc networks (VANETs). However because of the high associated cost, we do not see as many high quality experiment studies as simulation or analysis ones. In this paper, we present extensive experimental measurements of vehicle-to-vehicle communication on several typical scenarios in order to study the critical factors that affect multimedia applications over an IEEE 802.11 VANET. Unlike [4], we proof that in addition to short message transfer, the deployment of multimedia applications is also feasible. We found that network performances are clearly different according to the environment (city, highway, etc.), the distance between vehicles and the vehicles speed.
Conference Paper
Full-text available
In this paper we look at the signaling for session mobility using the Session Initiation Protocol (SIP) and the Session Description Protocol (SDP). Bearing in mind that perceived voice quality is sensitive mostly to delay variation (also called jitter), we work on bounding that value for providing a better voice Quality of Service over Wireless LAN (WLAN). For mobility we use a hybrid SIP-MIP stack. The basic idea is to combine the strengths of MIP for doing fast handovers and the strengths of SIP for powerful session adaptation capabilities. Numerical calculation figures as well as real implementation results are provided. It can be observed that the session update cycle during mobility can be modified in such a way that session delay variation bounds achieved are quite low and there are almost no spikes. We use composite performance metrics that we defined in previous work to analyze the correlation between the data patterns for session signaling delay in the classical case and our proposed case to better highlight the achieved performance delta.
Conference Paper
Ever since the emergence of Intelligent Transportation System (ITS), there has been a rapid integration of communications and automotive electronics industry, promoting developments and trends for the new generation of telematics. Apart from the development and standardization of telematics, its field test is however attracting less attention while being a most important aspect of the issue. The purposes of this research are to set up software/field tests and propose possible improvements for IEEE 802.11p standard. Then, obtaining information from software simulation to establish a controllable field for ground tests, and analyzing the indicative parameters of packet loss and latency of Dedicated Short Range Communication (DSRC) telematics system under mobile conditions, to complete a systematic testing procedure, statistics analysis as well as a transceiver design that will improve the transmission performance.
Article
Next-generation telematics solutions are being driven by the maturation of recently deployed intelligent transportation systems, assisted by the integration of and rapid collaboration with information communication technology markets and the automotive industry. Examples of this trend can be found in DSRC contexts in which Wi-Fi and WiMAX have become very relevant in vehicular environments. Comprehensive field trials were conducted and relevant measurements were collected in physical environments to investigate the possible environmental factors that occur in roadside-unit-to-vehicle communication scenarios. In these carefully controlled automobile experiments, the most representative parameters are measured and analyzed for DSRC in dynamic environments (e.g., packet loss, latency, and delay spread). Using a systematic testing procedure and through measurement analysis, the communication performance and the operating reliability of wireless communications can be effectively verified; insightful improvements and helpful suggestions and contributions may thus be achieved for telematics system designs. Furthermore, a practical, reasonable, and standardized testing procedure is proposed, and it can be utilized in equipment verification and product certification.
Conference Paper
Full-text available
This paper introduces and validates an experimental infrastructure of a fusion IP/DVB networking environment for the provision of triple play services at remote areas exploiting the synergy of Broadcasting, Internet and Telecommunication sectors technologies. This synergy constitutes a challenge for creating a networking platform which exploits the particularities and complementarities of these sectors for the support of ubiquitous services and always-on connectivity enabling passive rural citizens to become active Information Society participants. This paper describes important directions for exploiting the proposed unified platform alleviating the digital divide that currently exists not only among countries but also within most regions of the same country.
Conference Paper
This paper proposes a novel model for instigating the ldquoperfectrdquo DDoS attack. Our model uses a combination of network feedback and coordination of attack rates and protocols to achieve a high level of service deniability and low detectability. We demonstrate via extensive simulations that our attack model achieves a higher impact as compared to the shrew and the RoQ (reduction of quality) attacks while the detectability is low. The model adopts botnets in executing/realizing the perfect storm/attack in a real network. We show that executing the attack is not difficult at all. We then briefly describe failures of recent and most relevant detection approaches in mitigating these attacks. The failures of these schemes demonstrate low detectability of our proposed ldquoperfect stormrdquo.
Conference Paper
Several assessment sessions involving human observers were conducted on the output of a laboratory broadband digital television system to examine how packet reordering could affect the audiovisual subjective quality. Besides packet reordering extent, the system was also varied in set-top box, accompanying operating system, and content genre. It was found that current receivers are similarly exposed to the impairment and that their output quality becomes intolerable for more than 0.12% of reordered packets. Also, high-motion content appears more vulnerable to packet reordering. These findings suggest that packet reordering is a valid threat to broadband digital television, so we briefly discuss a few protective measures
Article
Wireless mesh networks are being rapidly deployed in both urban and rural areas impact-ing a large fraction of the world's population. As the profile of the average user changes so does the observed traffic: voice and video communica-tions are grabbing a bigger share since the net-works are used as a key communication tool, es-pecially in the areas without any other informa-tional infrastructure. However, real-time applica-tions require strict performance guarantees that cannot be provided by the current off the shelf so-lutions. Understanding the behavior of voice and video traffic in the wireless environment sets good foundations for designing an efficient solutions on all networking layers. In this paper we focus on analysis of large scale network traces, experi-menting with voice and video flows over the wire-less medium and, based on the derived results, de-signing a packet scheduler for a TDMA-like MAC layer protocol.
Conference Paper
The low rate TCP DoS (Shrew) and the reduction of quality (RoQ) attacks have been proved to be detrimental to the TCP traffic in the Internet. In this paper, we investigate the effect of these attacks on the QoS sensitive real time UDP traffic like VoIP. The stealthy RoQ attacks in particular are hard to detect as its periodicity is not well defined as compared to the shrew attack, and they can be launched even with the destination IP address spoofed! The attack potency formula is tailored for our work to observe the detrimental effect of the attack on the VoIP traffic. We study the effect of the attack flow on three QoS parameters of a VoIP flow, namely, the delay, the jitter, and the packet loss. We have performed extensive ns2 simulations under realistic scenarios to validate the effect of these attacks on the real time UDP traffic. Our results indicate that both these attacks are capable of driving a VoIP call from good to acceptable quality, i.e., reduction of quality, and from acceptable quality to complete denial of service. We believe that it is critical to investigate the characteristics of any attack in order to develop sound mitigation mechanisms; this work highlights one such characteristic of the low rate DoS attack.
Article
Full-text available
This paper describes VQmon, a non -intrusive monitoring technique for Voice over IP networks that is computational ly efficient and suitable for integrating or embedding into VoIP gateways or IP Phones. This uses an extended version of the ITU G.107 E-Model incorporating the effects of time varying packet loss and "recency". A 4 state Markov model is used to represent the time distribution of packet loss during a VoIP call. QoS, Voice over Packet, E model, subjective quality A. INTRODUCTION Voice over IP networks differ from conventional telephone networks in that voice quality is affected by a wider variety of network impairments and can vary from call to call and even during a call. It is therefore desirable to monitor call quality in order that service providers can properly provision networks and that network resources are properly allocated. Passive monitoring systems examine operating characteristics of a system in order to assess or measure performance level. This may involve examining elements of the system, for example buffer levels, or examining the data stream being transmitted through the system. This contrasts with Active measurement systems in which test data is inserted into the system and used to obtain performance measurements. This paper describes a passive monitoring system (VQmon) for Voice over IP networks that is able to monitor per-call quality, providing feedback to a service management or CDR (Call Detail Record) system. The VQmon monitoring system also considers the effects of time varying impairments such as bursty packet loss and recency.
Article
Full-text available
H.323 protocol based Voice and Video conferencing solu-tions are established popular technologies both in industry and academia. However, these bandwidth intensive appli-cations are often plagued by various performance problems. In this paper we describe a few common end-to-end per-formance problems noticed over several years in our ITU-T H.323 protocol-based "Video and Voice over IP" (VVoIP) infrastructure and provide a tool for troubleshooting these performance issues. Based on our operations experiences, we are developing a tool called the "H.323 Beacon" to mea-sure, monitor and qualify the performance of H.323 sessions. We describe the development architecture and feature set of the H.323 Beacon, which can be used by a novice end-user, network engineer or conference-operator using VVoIP sys-tems. H.323 Beacon is an application-specific network mea-surement tool that provides H.323-protocol specific evidence and other information necessary to troubleshoot H.323 ap-plication performance in the network and at the host (end-to-end). We also present 2 use-cases for the H.323 Beacon that demonstrate the utility of the tool for VVoIP perfor-mance debugging. Finally, we outline essential best prac-tices for prevention and efficient resolution of intermittent and imminent failures of H.323 VVoIP applications.
Conference Paper
Full-text available
The popularity of H.323 applications has been demonstrated by the billions of minutes of audio and video trac seen on the Internet every month. Our objective in this paper is to obtain Good, Acceptable and Poor performance bounds for network metrics such as delay, jitter and loss for H.323 applications based on objective and subjective quality assessment of various audio and video streams. To obtain the necessary data for our analysis we utilize the H.323 Beacon tool we have developed and a set of Videoconferencing tasks performed in a LAN and also with end-points located across multiple continents, connected via disparate network paths on the Internet.
Article
A paradox in the videoconferencing and multimedia communications industry is that while there are defined international standards such as H.320 and H.323 which set foundations for network and vendor interoperability, there are no standards at all for performance evaluation. It should therefore come as no surprise that every vendor of videoconferencing claims "great quality," a wonderfully subjective measure, while also claiming 15 fps at 128 kbps, a wonderfully quantitative measure with no real substance behind it (Frame rates vary with motion content, among other things). Even worse perhaps is that there is no recognized industry consensus of what really determines conferencing quality, especially video quality. Caveat emptor. One of the issues at play is that perceived multimedia quality can be captured only partially by quantitative measurements. Other aspects are equally important, but largely subjective. Generally these are too complex to capture with a single quantitative figure. For example, some vendors have engineered their products to maintain constant frame rate by sacrificing clarity when there is a high motion component. This represents a problem since there are no established units of measurement for how clear an image appears, or whether the video is smooth or choppy. In mid-1997, Intel Corporation began a program to develop a suite of quantitative and qualitative measurements to help users evaluate audio and video quality in videoconferencing systems. We created a laboratory test stand and then evaluated commercially available, standards-compliant products from both Intel and other leading vendors. This article will discuss the test development program and the technical factors, both quantitative and qualitative, which affect perceived conferencing quality.
Conference Paper
Time, day, location and instantaneous network conditions largely dictate the quality of Voice over IP calls. In this paper we present the results of over 18000 VoIP measurements, taken from nine sites connected in a full-mesh configuration. We measure the quality of the routes on a hourly basis by transmitting a pre-recorded call between a pair of sites. We repeat the procedure for all nine sites during the one hour interval. Based on the obtained jitter, delay and loss values as defined in RFC 1889 (RTP) we conclude that the VoIP quality is acceptable for all but one of the nine sites we tested. We also conclude that VoIP quality has improved marginally since we last conducted a similar study in 1998.
Conference Paper
As the Internet evolves into a ubiquitous communication infrastructure and provides various services including telephony, it has to stand up to the toll quality standards set by traditional telephone companies. Our objective is to assess to what extent today's Internet meets this expectation. Our assessment is based on delay and loss measurements taken over wide-area backbone networks, considers realistic VoIP scenarios and uses quality measures appropriate for voice. Our findings indicate that although voice services can be adequately provided by some ISPs, a significant number of paths lead to poor performance even for excellent VoIP end-systems. This makes a strong case for special handling of voice traffic on those paths. Even on the good paths, rare loss events can occasionally cause perceptible degradation of voice quality. Finally, the appropriate choice of the playout buffer scheme for each path was found to be of critical importance for the perceived quality.
Article
Today's powerful computers and networks present the opportunity for video across the Internet right to the desktop. However, Internet video often suffers from packet loss and jitter, degrading the user's perceived quality of the video. The effects of packet loss on perceptual quality are well-understood, but to date there have not been careful user studies measuring the impact of jitter on perceptual quality. The major contributions of this work are carefully designed experiments that measure and compare the impact of both jitter and packet loss on perceptual quality of packet video. We find that jitter degrades perceptual quality nearly as much as does packet loss, and that perceptual quality degrades sharply even with low levels of jitter or packet loss as compared to perceptual quality for perfect video. 1 Introduction The power of today's computers and the connectivity of today's networks present the opportunity of packet video over the Internet and to the desktop. Video on the de...
A framework for Quality of Service analysis of IP-based video networks
  • V Chandrashekar
V. Chandrashekar, " A framework for Quality of Service analysis of IP-based video networks ", Masters Thesis, North Carolina State University, 2003.
Infrastructure of audiovisual services-Systems and terminal equipment for audiovisual services
  • Itu-T Recommendation
ITU-T Recommendation H.323, " Infrastructure of audiovisual services-Systems and terminal equipment for audiovisual services ", 1999.