B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communications," IEEE Trans. Commun.,, Vol. 36, No. 7, July 1988, pp. 834-843.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....traffic substitutes is promising, using processes with more parameters should be further investigated. This is important for extending the approach to match other measures of the queueing process, such as the mean queue length. Since this particular direction is well understood by now (e.g. see [23], 30] combining these approaches should be feasible. Another direction for future research is the application of our work on a network wide basis, possibly in conjunction with routing that is based on link metrics. Acknowledgement: The authors are grateful to Frank Kelly and Vasilios Siris for ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Robbins, "Performance models of statistical multiplexing in packet video communications, " IEEE Transactions on Communications, vol. 36, no. 7, pp. 834-- 844, July 1988.

.... Theta. This proposition is more general than Proposition 3.1. It applies to any traffic specification Theta, rather than just (Xmin; Xave; I; Smax) For example, Theta can be the (oe; ae) model as proposed in [19] or even stochastic models like Markov Modulated Poisson Process (MMPP) models [59]. As will be discussed in Chapter 4, this property of completely reconstructing the traffic pattern allows us to extend local statistical performance bounds to end to end statistical performance bounds. From Propositions 3.1 and 3.2, we can see that both types of regulators enforce the traffic ....

....bounds in a network environment. 72 In the literature, many models have been proposed for video or audio traffic sources. Among the most popular ones are the on off model for voice sources [8, 9] and more sophisticated models based on Markov or other renewal processes for video sources [59]. A good survey for the probabilistic models for voice and video sources is presented in [65] There are two important limitation to such traffic models. First, in an integrated services network, traffic sources are heterogeneous and will not in general conform to one model. If the traffic source ....

[Article contains additional citation context not shown here]

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J.D. Robbins. Performance models of statistical multiplexing in packet video communications. IEEE Transaction on Communication, 36(7):834--844, July 1988.

.... a series of focused research efforts in the mid to late eighties as attested by events such as the First International Packet Video Workshop, which took place at Columbia University in May 1987, and the appearance of papers that have since been widely referenced in the packet video literature [1, 2, 3]. From the very beginning it was recognized that successful deployment of packet video would require advances in both compression and networking technologies, along with a great deal of interaction and collaboration between the two communities. Thus, the Packet Video Workshop aimed at attracting ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. Robbins, "Performance models of statistical multiplexing in packet video communications," IEEE Trans. on Comm., vol. 36, pp. 834-843, July 1988.

....will probably require these control and management functions to remain software procedures for a while. The traffic associated with these functions is typically lower in volume by several orders of magnitude than the user to user traffic. A video connection sends about 10,000 packets per second [MASK88]. Distributed computing applications typically send no more than a few dozen packets per second [AKRP87] Therefore, in contrast with switching capacity, communication node designers will not have the same incentive to increase processing capacity within the node. The PARIS network prototype ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. Robbins, Performance Models of Statistical Multiplexing in Packet Video Communications, IEEE Transactions on Communications, VOL. 36, NO. 7, July 1988, pp. 834-844.

....a talk spurt and the subsequent silent interval are set to 3 sec, which means voice activity factor is 50 . Video Traffic In the literature, video traffic models ranging from classical models based on Poisson arrival processes to sophisticated models like autoregressive processes Markov chains [13] and self similar models [14] are used. Most of these models were derived from the data of MPEG video clips, and MPEG streams generally require a high data rate to sustain the quality of playback; therefore their suitability for the wireless medium is questionable. For these reasons, we chose to ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, J.D. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communications," IEEE Trans. Comm., vol. 36, pp. 834-843, 1988.

....intervals are omitted from some simulation values in order to avoid obscuring the information being presented. V. CONVERGENCE TO GAUSSIAN In recent years a number of researchers have investigated the usefulness of Gaussian processes in representing a variety of traffic types [20] 21] 22] [23], 24] 25] Analytic expressions have been developed for the queueing performance of both LRD and non LRD Gaussian processes [20] 25] The existence of such expressions makes the Gaussian process an attractive model, where it is applicable. In this section we will show one reason why the ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. Robbins, "Performance models of statistical multiplexing in packet video communications, " IEEE Transactions on Communications, vol. 36, pp. 834-- 844, July 1988.

....models are needed. In particular, given that the most of the network traffic will be constituted by MPEG video traffic, its statistical characterization and an accurate model of it is mandatory. For VBR video traffic various proposals have been made. The first Markov based model was proposed in [3] for slow motion video sequences and buffer performance was calculated through fluid flow approximation. In the same paper another approach, based on the first order autoregressive process theory, was used to generate video traffic for a simulative approach. Then other models [4 5] were proposed ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson and J. D. Robbins, "Performance models of statistical multiplexing in packet video communication", IEEE Transaction On Communications, Vol. 36, pp. 834-44, July 1988.

....the amount of data to be transmitted. The cells are distributed uniformly over one frame time (1 30st of a second) 2. The AAL buffers the cells and transmits them at the highest possible speed. In this case, cells are transmitted in a burst, in which the generation rate is constant. Maglaris [21] models video as a bit stream with discrete levels of bit rates, describing this as a superposition of 20 identical on off sources. This results in an MMPP with 21 different states, of which state 0 has bit rate 0, and state 20 denotes the peak rate. This reflects the properties of the first ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, J.P. Robbins, "Performance Models of statistical multiplexing in packet video communications", IEEE Transactions on Communications 36(7), pp.834-844, 1988.

....process at the time scale where overflows occur. In large systems, modeling should be performed using only the first and second order characteristics of the input process (diffusion approximation) This shows that our method agrees with the established ones for the cases of heavy multiplexing (see [9, 5, 6]) 3.4 The Case of B = 0 We consider bufferless links where we substitute real traffic using Gaussian sources. The overflow probability of the link depends only on steady state characteristics of the input process. The time parameter of bufferless systems is always (independently of the input ....

....traffic substitutes is promising, using processes with more parameters should be further investigated. This is important for extending the approach to match other measures of the queuing process, such as the mean queue length. Since this particular direction is well understood by now (e.g. see [9, 8]) combining these approaches should be feasible. An important application of the fast calculation procedure of the previous section is the on line calculation of the buffer overflow probability and the effective bandwidths of the input streams. The idea is to solve system (15) on line. This can ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Robbins. Performance models of statistical multiplexing in packet video communications. IEEE Trans. on Comm., 36(7):834--844, July 1988.

....should be, since in the new model, c and m i will also double. 3 Traffic Source Models In this section we describe some simple and powerful models, which can be used to characterize two real time traffic sources: voice and videotelephone. An Autoregressive Markov model Maglaris, et al. see [11]) use the following continuous state, autoregressive, discrete time Markov process as a model for videotelephone sources. It is very simple to implement in simulation experiments. Let X n represent the bit rate of a videotelephone source during the nth frame (assuming that the coder uses the ....

....in c=N , and requiring c NE also leads to (5) 5 Voice source: The model can be used for a voice source with ADPCM modulation and speech activity detection mechanisms. Ibrahim, et al. 8] suggest parameters of a = 64Kb s, 1= 650ms and 1= 352ms. Videotelephone source: Maglaris, et al. [11] use the superposition of 20 on off Markov fluid sources, or video minisources , to model a single videotelephone source. For each video minisource, they use the following parameters: a = 945Kb s, 1= 1:273s, and 1= 0:321s. These give values X n 2 [0; 18:9] Mb s, m = 3:81 Mb s, and fl(0) ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Robbins. Performance models of statistical multiplexing in packet video communications. IEEE Transactions on Communications, 36(7):834, July 1988.

....profiles is quite general and can accurately represent a wide range of bursty network traffic, originating from various applications. Plain Markov modulated rate processes (featuring exponentially distributed state sojourns) have been successfully employed for representing voice [4,10] video [6,7,11,12] and data [3,13] traffic. The additional generality offered by the semiMarkovian models, namely the ability of specifying general probability distribution functions for the state durations, permits accurate modeling of traffic subject to shaping (e.g. leaky bucket based regulation) or throttle ....

....the number of states is not excessive and transition probabilities values are not near zero. 3.7. A special case of semi Markovian models: superposition of on off sources Many real world traffic sources can be modeled as the superposition of a number of exponential on off sources (see, e.g. [11,12]) Furthermore, many QoS theory fundamentals are based on traffic load of this form. Consequently, it is important to equip the traffic generator with the ability of generating such loads efficiently. However, it is quite expensive to allocate one source module on a traffic generation tool to a ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Robbins, "Performance models of statistical multiplexing in packet video communications", IEEE Trans. Commun., vol. 36, pp. 834-844, 1988.

....) n w is a sequence of independent Gaussian random variables and a and b are constants. The key parameters of the model are: 8781 . 0 = a , 1108 . 0 = b and 572 . 0 ) w E . From the measurement, 52 . 0 ) l E bits pixel give fairly accurate approximation of the bit rate as detailed in [8]. Due to the high throughput nature of video traffic, the compressed IP header can be neglected in the simulation campaign [9] C. Data Service and HTTP Traffic The data service refers to the applications like WWW, Email or FTP. As HTTP traffic is the dominating data service [6] it is modelled ....

Basil Maglaris, Dimitris Anastassiou, Prodip Sen, Gunnar Karlsson and John D. Robbins, Performance Models of Statistical Multiplexing in Packet Video Communications, IEEE Trans. Com., Vol. 36, No. 7, Jul., 1988

....frames alternate. Accurate character ization of these compressed data streams is essential for real time transport of such data over ATM networks. Numerous models have previously been proposed for VBR video under various compression schemes [26] 31] 32] 33] 34] 35] 36] 37] [38], 39] 40] 41] 42] 43] 44] 45] Since the VBR behavior of a video stream strongly depends on the compression technique used, many of these models do not characterize MPEG coded video which is now widely accepted as a standard for transmission and storage of video data in many ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J.D. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communications," IEEE Trans. Comm., vol. 36, no. 7, pp. 834-844, July 1988.

....based on transmission requests preceded by each packet transmission. The reason for using such control schemes is that conventional access control protocols do not consider stochastic properties of each traffic. It is reported that the bit rate of voice or video traffic is correlated [7] [8]. If access protocols consider such a stochastic property of voice or video traffic, it might be able to estimate the instantaneous number of simultaneously transmitted packets. It would bring us the possibility of improvement of the performance. In this paper, we propose an access control ....

....received packets. But it can t know the number of voice packets that will be generated in the future slots. This is because the length of talkspurt is a random variable. B. Video Traffic Model Bit rate of a single video user during a video frame is modeled as a first order autoregressive process [8]. The generation process of the video packets is illustrated in Figure 2. The bit rate #### in the #th video frame is expressed as by ##### # # ## (1) where ##### # # is sequence of Gaussian random variables with mean # and standard deviation # # . The length of video frame is # # . Thus, the ....

B. Maglaris, et al, "Performance models of statistical multiplexing in packet video communications", IEEE Trans. Commun., vol. 36, no. 7, pp. 834--843, July 1988.

....of 1.004s and 1.587s respectively. This corresponds to a 38.53 talk spurt cycle, as recommended by ITU T specification for conversational speech. The peak transmission rate is 64kbps and the average is 24.8kbps. The model that is used for the Videoconference sources is the one proposed in [2]. The average transmission rate is 3.9Mbps and the peak transmission rate is 10.575Mbps. The characteristics of the transmission rate can be approximated either by a Continuous State Autoregressive Markov Model or by a superposition of 20 ON OFF Markov sources where each of them has a peak ....

Basil Maglaris, Dimitris Anastassiou, Prodip Sen, Gunnar Karlsson, John Robbins. Performance Models of Statistical Multiplexing in Packet Video Communications. IEEE Transactions on communications, July 1989

....[55, 7] Although it is also possible to estimate the effective bandwidth on line using the realtime traffic measurements [8] it is very hard to monitor the source conformance to its effective bandwidth in real time. Two traffic models for video phone sessions were proposed by Maglaris et al. [32]: the continuous state autoregressive Markov model, and the discrete state continuous time Markov process. The first model can be used in simulations, but it does not lead to simple results of queuing analysis. The second model can be used to analyze performance of a statistical multiplexer. This ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communications", IEEE Transactions on Communications, Vol. 36, No. 7, pp. 834-844, July 1988.

....black box approaches typically treat the measurement as a time series. They focus on capturing the statistical characteristics (particularly autocorrelation and marginal distribution) of empirical data to model network traffic, based on various approaches such as Markov process, ARIMA, TES etc. [24, 33, 48, 42, 32, 18, 23, 32, 40]. Although being able to reproduce the measured traffic correctly, these approaches generally ignore the underlying network structure and hence provide little or no insight about the observed characteristics of measured traffic and its underlying causes. On the other hand, structural modeling, ....

B. Maglaris, D. Anastassiou, G. Karlsson P. Sen, and J. D. Robbins. Performance models of statistical multiplexing in packet video communications. IEEE Trans. on Comm., 36(7):834--844, 1988.

....characterize the output of a particular coder used in a particular context, or oblige the coder to make its output conform to predefined parameters by means of rate control. A large number of studies have been performed on the characterization of coder output (e.g. 8] 10] 15] 16] 24] [26], and [29] For certain applications such as videoconferencing, it may be possible to characterize the output succinctly in terms of a small number of parameters such as the first two moments of the per frame bit rate and the coefficient of an assumed exponential autocorrelation function [17] ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Robbins, "Performance models of statistical multiplexing in packet video communications," IEEE Trans. Commun., vol. 36, pp. 834--844, July 1988.

....to model variations in the input cell arrival process on a variety of different time scales. These time scales may represent time intervals (in network time slots) over which logical information blocks are delivered to the network. For instance, consider a Variable Bit Rate (VBR) video coder, [7], 8] 9] Bits generated over a frame form a logical block of information which could provide a basis for the description of the generated traffic as well as the desirable QoS. Notice that if the VBR video coder prebuffers and uniformly distributes cells in each frame, then the cell activity over ....

....frame at slice boundaries. Note that the absence of frame level pre buffering has been shown [10] 11] 12] 13] to give rise to cell arrival autocorrelation functions which are pseudo periodic, as opposed to the monotonic autocorrelation functions arising from frame level pre buffering schemes [7]. Packetized voice provides yet another example of a common traffic source time scale when one considers the duration of time required to generate enough bits to form a cell. The resulting time scale is then representative of the minimum cell inter arrival time associated with the source. When ....

[Article contains additional citation context not shown here]

B. Maglaris, D. Anastassiou, S. Prodip, G. Karlsson, and J. Robbins. Performance models of statistical multiplexing in packet video communications. IEEE Trans. Commun., 36, July 1988.

....and hopefully realistic, packet arrival processes. Such studies can be found in [1] 6] and the references therein, for input traffic models which are capable of capturing the burst level correlations present in packetized voice. The multiplexing of variable bit rate (VBR) sources is considered in [7] and [8] An important characteristic of high speed networks, which is not explicitly considered in any of the previously mentioned studies, is the potentially large disparity in the relative speeds of the output link and the input source. This disparity, or speed up factor commonly appearing ....

B. Maglaris, D. Anastassiou, S. Prodip, G. Karlsson, J. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communications ", IEEE Trans. Commun., Vol. 36, July 1988.

....: stavraka emba.uvm.edu The magnitude of these time scales, which increases with the link capacity C, can significantly affect the impact that a traffic source has on the network. Consider a Variable Bit Rate (VBR) video coder which pre buffers and uniformly distributes cells in each frame, [1] [3] In general, such a source produces a cell stream in which the number of cells generated in consecutive frames is highly correlated. However, as far as network performance is concerned, the impact of inter frame autocorrelations may become less significant as the link speed, and consequently ....

....traffic is proposed based on an 8 bin histogram of the average bit rates over a frame. Cell arrivals over a frame are then considered to be Poisson with a constant intensity selected from the 8 bin histogram. This traffic model implies pre buffering of the VBR traffic at the frame level as in [1] and [2] When a number of VBR video streams are multiplexed, a cumulative average bit rate over a frame is derived through the convolution of the individual histograms and a Poisson model for the cell arrivals over a frame is considered with intensity selected from the cumulative histogram. The ....

B. Maglaris, D. Anastassiou, S. Prodip, G. Karlsson, J. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communications", IEEE Trans. Commun. , Vol. 36, July 1988.

....process at the time scale where overflows occur. In large systems, modeling should be performed using only the first and second order characteristics of the inputprocess (diffusion approximation) This shows that our method agrees with the established ones for the cases of heavy multiplexing (see [10], 11] 1] D. The Case of B = 0 We consider bufferless links where Gaussian sources are substituted for real traffic. The overflow probability of the link depends only on steady state characteristics of the input process. The time parameter of bufferless systems is always (independently of ....

....traffic substitutes is promising, using processes with more parameters should be further investigated. This is important for extending the approach to match other measures of the queueing process, such as the mean queue length. Since this particular direction is well understood by now (e.g. see [10], 14] combining these approaches should be feasible. An important application of the fast calculation procedure of the previous section is the on line calculation of the effective bandwidths of the input streams and the buffer overflow probability. The idea is to solve system (12) on line. ....

Basil Maglaris, Dimitris Anastassiou, Prodip Sen, Gunnar Karlsson, and John D. Robbins, "Performance models of statistical multiplexing in packet video communications," IEEE Trans. on Comm., vol. 36, no. 7, pp. 834-- 844, July 1988.

....itself. For our study in this paper, we focus on the case when the aggregate tra#c can be characterized by a stationary Gaussian process. Recently, Gaussian processes have received significant attention as good models for the arrival process to a high speed multiplexer [3] 18] 19] 20] 21] [22], 23] There are many reasons for this. Due to the huge link capacity of high speed networks, hundreds or even thousands of network applications are likely to be served by a network multiplexer. Also, when a large number of sources are multiplexed, characterizing the input process with ....

....that have been used in other papers in the literature (e.g. 20] 36] A. Gaussian Processes We begin by considering the simple case when the input is a Gaussian Autoregressive (AR) process with autocovariance C # (l) 258 0. 9 l (note that AR processes have been used to model VBR video [22]) In Fig. 2 one can see that the simulation and MVA Loss result in a close match over the entire range of bu#ers tested. The next example, in Fig. 3, covers a scenario of multitime scale correlated tra#c. Note that multiple time scale correlated tra#c is expected to be generated in high speed ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communication," IEEE Transactions on Communications, vol. 36, pp. 834--843, July 1988. 12

....and the number of cells in a burst is geometrically distributed. During a burst, cells are transmitted at the peak rate. The characteristics of video traffic depend to a large extent on the encoding algorithm used. The video model used in this simulation study is one developed by Maglaris et al. [8] based on measurements. The bit rate of a single source during the nth frame is modeled by a first order autoregressive Markov process. The length of a burst is based on the number of frames per second and the cell rate during a burst is based on the number of cells (or bits) per frame. By varying ....

....burst is based on the number of frames per second and the cell rate during a burst is based on the number of cells (or bits) per frame. By varying these two values, traffic streams of various load levels can be obtained. This does not violate the characteristics of the model since Maglaris et al. [8] note that their 7 model assumptions hold for a wide variety of video sources. Although the burst length of a video source is expected to be fairly long (on the order of 40 ms) this length is infeasible for practical simulation lengths. A burst length of 1.7 ms was used for this study. Longer ....

Basil Maglaris, Dimitris Anastassiou, Prodip Sen, Gunnar Karlsson, and John D. Robbins. Performance models of statistical multiplexing in packet video communications. IEEE Transactions on Communications, 36(7):834--843, July 1988.

....Magda El Zarki Department of Electrical Engineering University of Pennsylvania Philadelphia PA 19104, U.S.A. TEL: 215) 898 9780 e mail: magda ee.upenn.edu FAX: 215) 573 2068 pancha ee.upenn.edu such as interframe coding combined with variable length codes[9] or conditional replenishment[6] [8] that are less efficient than current coding techniques. Other techniques such as subband coding[1] have demonstrated promise for only a few types of services. Another recent study, which provides insights into long run statistics for coded video sequences, used block based intraframe ....

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J.D. Robbins. Performance models of statistical multiplexing in packet video communications. IEEE Trans. on Comm., 36(7):834--844, July 1988.

....typically used to model traffic is the Markov modulated fluid model. In this model, the current state of the underlying Markov chain determines the flow (traffic) rate. While in state s k , traffic arrives at a constant rate l k . This model is a Markov modulated constant rate model and is used in [7, 8] to model VBR video sources. In [7] the continuous bit rate is quantized into a finite set of discrete levels and sampled at random Poisson points (i.e. intersample time is exponentially distributed) as shown in Fig. 4. The number of states in the Markov chain is equal to the number of ....

....Markov modulated fluid model. In this model, the current state of the underlying Markov chain determines the flow (traffic) rate. While in state s k , traffic arrives at a constant rate l k . This model is a Markov modulated constant rate model and is used in [7, 8] to model VBR video sources. In [7], the continuous bit rate is quantized into a finite set of discrete levels and sampled at random Poisson points (i.e. intersample time is exponentially distributed) as shown in Fig. 4. The number of states in the Markov chain is equal to the number of quantized levels. Since Markov processes ....

[Article contains additional citation context not shown here]

B. Maglaris et al., "Performance Models of Statistical Multiplexing in Packet Video Communications," IEEE Trans. Commun., vol. 36, July 1988.

....of MPEG 2) are generally invisible, even to experienced viewers [4] To determine a set of techniques appropriate for CAC and UPC, traffic models that accurately represent the statistical nature of very high speed bursty services are necessary. Previous studies on video models can be found in [5 11]. These studies have focussed on modeling one layer VBR traffic. Much less is known about the statistical characteristics of two layer video. In two layer coding algorithms, the base layer can be decoded independently to produce a lower quality picture, and should be transported at high priority ....

B. Maglaris, D. Anastassiou, P. Sen and G. Karlsson, "Performance models of statistical multiplexing in packet video communications," IEEE Trans. Comm., vol. 36, p834-843, 1988.

....and the number of cells in a burst is geometrically distributed. During a burst, cells are transmitted at the peak rate. The characteristics of video traffic depend to a large extent on the encoding algorithm used. The video model used in this simulation study is one developed by Maglaris et al. [10] based on measurements. The bit rate of a single source during the nth frame is modeled by a first order autoregressive Markov process. The length of a burst is based on the number of frames per second and the cell rate during a burst is based on the number of cells (or bits) per frame. By varying ....

....is based on the number of frames per second and the cell rate during a burst is based on the number of cells (or bits) per frame. By varying these two values, traffic streams of different load levels can 6 be obtained. This does not violate the characteristics of the model since Maglaris et al. [10] note that their model assumptions hold for a wide variety of video sources. Although the burst length of a video source is expected to be fairly long (on the order of 40 ms) this length is infeasible for practical simulation runs. A burst length of 1.7 ms was used for this study. Longer burst ....

Basil Maglaris, Dimitris Anastassiou, Prodip Sen, Gunnar Karlsson, and John D. Robbins. Performance models of statistical multiplexing in packet video communications. IEEE Transactions on Communications, 36(7):834--843, July 1988. 18

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communications," IEEE Trans. Commun.,, Vol. 36, No. 7, July 1988, pp. 834-843.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Robbins, "Performance models of statistical multiplexing in packet video communications," IEEE Trans. Commun., vol. 36, pp. 834844, Nov. 1988. 0.0001 0.001 0.01 0.1 Video packet loss rate The number of voice mobile calls Scheme Scheme II Modified DQRUMA/MC-CDMA Modified ideal DQRUMA/MC-CDMA

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson , and J. D. Robbins, "Performance models of statistical multiplexing in packet video communication, " IEEE Trans. Commun., vol. 36, pp. 834--843, 1988.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson and J. D. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communication," IEEE Trans. on Comm., Vol. 36, pp. 834-843, 1988.

No context found.

Maglaris, B, Anastassiou, D, Sen, P, Karlsson, G and Robbins, J. D. (1988) Performance models of statistical multiplexing in packet video communication.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson and J. D. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communication," IEEE Trans .on Comm., vol. 36, pp. 834-843, 1988.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson and J. D. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communication," IEEE Trans. on Communications., vol. 36, pp. 834-843, 1988.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson and J. D. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communication," IEEE Trans. Communications, vol. 36, pp. 834-843, 1988.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Robbins, "Performance models of statistical multiplexing in packet video communication, " IEEE Trans. Commun., vol. 36, pp. 834--843, July 1988.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson and J. D. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communication," IEEE Trans. Comm., vol. 36, pp. 834-843, 1988.

No context found.

B. Maglaris et al., "Performance models of statistical multiplexing in packet video communications," IEEE J. Select. Areas Commun., vol. 36, no. 7, pp. 834--844, Jul. 1988.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson and J. D. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communication," IEEE Trans.on Comm., vol. 36, pp. 834843, 1988. 23

No context found.

B. Maglaris, D.Anastassiou, P.Sen, G. Karlsson and J. Robins, "Performance models of statistical multiplexing in packet video communications," IEEE Trans. Commun., vol. 36, No.7, pp.834-844, July 1988

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson and J.D. Robbins, Performance models of statistical multiplexing in packet video communications, IEEE Transactions on Communications 36(7) (July 1988) 834--844.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson and J.D. Robbins, "Performance models of Statistical Multiplexing in Packet Video Communication," IEEE Trans. Comm., vol. 36, pp. 834-43, 1988.

No context found.

B. Maglaris, D. Anastassiou, G. Karlsson P. Sen, and J. D. Robbins. Performance models of statistical multiplexing in packet video communications. IEEE Trans. on Comm., 36(7):834--844, 1988.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Robbins. Performance models of statistical multiplexing in packet video communications. IEEE Transactions on Communications,36( July 1988.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Robbins. Performance models of statistical multiplexing in packet video communications. IEEE Transactions on Communications, 36(7):834--844, July 1988.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Roberts. Performance models of statistical multiplexing in packet video communications. IEEE Trans. Communications, 36(7):-- July 1988.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Robbins. Performance models of statistical multiplexing in packet video communications. IEEE Transactions on Communications, 36(7):834--844, July 1988.

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. Robbins, "Performance Models of Statistical Multiplexing in Packet Video Communications", IEEE Trans. Commun., Vol. 36, pp. 834-844, 1988. 24

No context found.

B. Maglaris, D. Anastassiou, P. Sen, G. Karlsson, and J. D. Robbins, "Performance models of statistical multiplexing in packet video communications," IEEE Trans. on Communications, vol. 36, no. 7, pp. 834--844, July 1988.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC