T. Turletti, S.F. Parisis, and J. Bolot, "Experiments with a layered transmission scheme over Internet", INRIA Research Report No 3296, Nov. 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

.... bit stream that allows decoding at multiple rates [134, 85, 68, 129] Under layered multicast, di#erent layers of a scalable video are carried in di#erent multicast groups so that receivers can individually subscribe or unsubscribe to the appropriate multicast groups to achieve rate control [84, 114, 135, 140, 139, 76]. Thus, the more layers to which a receiver subscribes, the higher the quality of the received video. Generally, producing an embedded bit stream for scalable compression results in lower compression e#ciency than its non scalable counterpart at the same bit rates [27] Nevertheless, layered ....

....techniques may be e#ective for a single multicast session, there are no existing mechanisms to coordinate probes across di#erent sessions, leading to potentially unfair allocation of bandwidth. A more recently proposed scheme, equation based rate control (ERC) does not involve probing [139, 129]. Instead, the available bandwidth is computed directly at the receivers based on measured quantities such as packet loss rate. One characteristic of ERC is its ability to allocate bandwidth fairly among competing sessions [77] For example, two simultaneous video multicast sessions which share ....

[Article contains additional citation context not shown here]

T. Turletti, S. Parisis, and J. Bolot. Experiments with a layered transmission scheme over the Internet. Technical report, INRIA, http://www.inria.fr/RRRT/RR-3296.html, November 1997.

....data of a session into some layers, and sends them on separate multicast groups. Each receiver decides the number of layers to sub scribe based on its available bandwidth [2] That is, receivers adapt to network conditions by adding and dropping layers (i.e. joining and leaving multicast groups) [3, 4, 5, 6, 7]. However, these methods do not consider any routing protocol. There are two types of routing protocol in multicast. The former is dense mode routing protocols like DVMRP [8] MOSPF [9] and PIM DM, in which the distribution tree is Shortest Path Tree rooted at a sender node. The latter is ....

T. hrletti, S.F. Parisis, and J.-C. Bolot, "Experiments with a Layered Transmission Scheme over the Internet" . INRIA Research Report No. 3296, Nov. 1997.

....mechanism. A.1 Parameter Estimation Implementing a throughput model requires knowledge of the connection s MTU, RTT and loss rate. The MTU can be either set to a fixed value, for instance the standard minimum value of 576 bytes defined for TCP, or determined using an MTU discovery algorithm [20]. In our experiments, we used fixed length packets which allow for a constant MTU, set to 576 bytes. The RTT estimation is performed by keeping a recent average value. Following TCP s RTT smoothing strategy, we implemented an exponential filter with a constant 0.9 smoothing factor [21] This ....

....allows each receiver to determine its bottleneck rate. With the source advertising the layers it generates along with their current rates on a multicast control channel, each receiver is free to join to the layers that collectively fulfill its bandwidth capacity, similarly to the approach taken in [20]. Rate adaptation may further be implemented in several ways. For those applications which involve a limited number of participants (for instance up to ten receivers) we envision that the number of layers equates the number of participants, so that each one gets a dedicated enhancement layer. ....

T. Turletti, S. Fosse-Parisis, and J.C. Bolot. Experiments with a layered transmission scheme over the internet. Technical Report RR-3296, INRIA Sophia-Antipolis, 1998.

....complicated by the substantial heterogeneity in client transmission paths with respect to delay, bandwidth, and loss characteristics. Multi rate congestion control schemes have been proposed that roughly provide clients with transmission rates commensurate to their capabilities [12] 47] 50] [74], 75] 72] 7] 45] 70] Error control mechanisms are required to mitigate the effects of packet loss. Since real time streaming is delay constrained, the feasibility of retransmission based recovery mechanisms is debatable. In general, a mix of sender and receiver based mechanisms are ....

....their reception rate in response to changes in RTT, resulting in unfairness to TCP flows. A layered transmission scheme similar to RLM, in which join experiments are replaced by explicit bandwidth estimation at the receivers, using a TCP throughput function, was first proposed by Turletti et al. [74] for audio streaming. A similar protocol was proposed by Tan and Zakhor [71] A more elaborate protocol was recently proposed by Sisalem and Wolisz [70] where the server dynamically tunes the rate of the layers using feedback received from the clients. These schemes typically use the Real time ....

T. Turletti, S. Parisis, and J. Bolot. Experiments with a Layered Transmission Scheme over the Internet. INRIA Tech. Report 3296, November 1997.

....describe the collected throughput traces. 2 TCP Friendly Internet Bandwidth To compare the two streaming schemes, adding dropping layers and switching versions, we need to obtain TCP friendly traces. Several TCP friendly rate adjustment protocols have recently been developed by researchers [6, 12, 13, 8, 11]. The rate adjustment protocols reported in the literature have different limitations, depending on the TCP characterization on which they are based, and achieve TCP friendliness only under specific types of scenarios. For example, some of the proposed protocols are not TCP friendly at loss rates ....

....on which they are based, and achieve TCP friendliness only under specific types of scenarios. For example, some of the proposed protocols are not TCP friendly at loss rates higher than 5 [8, 11] Some of the protocols are specific to multicast applications and rely on data layering [12, 13]. We do not use any of the TCP friendly rate adjustment schemes in the literature to generate TCP friendly traces. Instead, it is natural to suppose that the perceived rate fluctuations of a TCP friendly scheme exhibit similar behavior as the fluctuations of TCP throughput over medium (seconds) ....

T. Turletti, S. Parisis, and J. Bolot. Experiments with a layered transmission scheme over the Internet. Technical Report RR-3296, INRIA, France, 1997.

....gateway transcodes an input video stream encoded using some scheme into a video stream encoded using another scheme with lower bandwidth, and forwards this stream down to the congested link which is a part of a multicast tree. More efficient approach is layered multicast with layered source coding[4, 5, 6, 7, 8, 9, 10]. In layered multicast, a sender of a session encodes an original stream into a base layer and several enhancement layers. A base layer has the highest priority and higher enhancement layer has lower priority. Then, a sender transmits each layer on a separate IP multicast group. Receivers join ....

....rate control must assure fair bandwidth allocation among competitive sessions at a congested link. Note that fairness in this paper is defined as max min fairness for layered multicast[6, 11] In general, rate control for layered multicast is classified into two classes; receiver driven scheme[4, 5, 6, 7, 8, 9] and network supported scheme[10] In receiver driven rate control, each receiver makes decisions to add or drop an enhancement layer. On the other hand, in network supported rate control, some of internal nodes such as routers or switches adopt priority dropping mechanism which drops lower ....

T. Turletti, S.F. Parisis, and J.-C. Bolot. Experiments with a Layered Transmission Scheme over the Internet. 1NRIA Research Report No. 3296, Nov. 1997.

....is to calculate the TCP friendly bandwidth at the receivers, and subscribe to the corresponding layers. Receivers simply adjust their subscription level to the rate given by the equation that models the throughput of an equivalent TCP session (The Layered Transmission Scheme (LTS) proposed in [8] and the TCP Friendly Transport Protocol (TFRP) proposed in [9] 2.2 Hybrid Source Receiver Mechanisms Source driven mechanisms try to accommodate all receivers with a single rate stream. The source determines this rate, by aggregating receiver feedbacks values. There is a trade off, in ....

T. Turletti, S. Parisis, and J. Bolot, "Experiments with a layered transmission scheme over the internet," Tech. Rep., Technical report RR-3296, INRIA, France, Nov. 1997.

.... Moreover, the fast and low cost operations for layer stream setup and termination have been supported in advanced video streaming standards, such as the MPEG 4 Delivery Multimedia Integration Framework (DIMF) Many scalable feedback algorithms have also been presented in the networking area [11]. It is a fact that a scalable feedback loop, such as RTCP, has been embedded in many streaming video systems. In this paper, we present a multi granular end to end adaptation framework for layered video transmission. In this framework, both the sender and the receivers perform adaptation. ....

T. Turletti, S. Parisis, and J. Bolot, "Experiments with a Layered Transmission Scheme over the Internet," Technical Report, INRIA, N'3296, November 1997.

....and a layered transmission scheme. For layered transmission over the best effort Internet, the key issues are where and how rate adaptation is performed. There have been a vast amount of proposals in the literature, which show that adaptation can be performed either on the receiver s side only [1,11], or on both the receiver s and the sender s sides [12,22] In a pure receiver based scheme, layers are mapped to different IP multicast groups. By joining corresponding groups, a receiver can obtain a certain level of video layers commensurate with its capacity [1] This distributed scheme ....

....overwhelming the background TCP traffic, HALM adopts the long term equivalent TCP throughput as the bandwidth requirement of a receiver, and performs layer rate allocation and layer adaptation accordingly. By incorporating additional information in the user feedback messages provided by RTP RTCP [11,30], it provides a general and efficient framework for layered video multicast that is friendly to TCP traffic, while the overhead is kept at a low level. The performance of HALM has been extensively studied through both simulation and statistical analysis under a variety of configurations. Our ....

[Article contains additional citation context not shown here]

T. Turletti, S. Parisis, and J. Bolot, "Experiments with a Layered Transmission Scheme over the Internet," Technical Report, INRIA, N'3296, November 1997.

....on equipment procured with NSF grant no. CDA 9624082. significant heterogeneity and may not achieve certain desirable fairness properties [18] In a multi rate approach, the sender transmits at several rates to different sets of receivers using either a replicated scheme [5] or a layering scheme [19, 15, 13, 20, 12, 22]. In the replicated scheme [5] receivers are partitioned into groups. The sender generates and sends separate data streams to different groups. The sending rate to a group is adjusted according to the capacities of receivers in the group, and is restricted to be in some fixed range. Note that ....

T. Turletti, S. F. Parisis, and J. Bolot. Experiments with a layered transmission scheme over the Internet. In Proceedings of IEEE INFOCOM '98, San Francisco, California, U.S.A., Mar. 1998.

.... many real time applications would find halving the sending rate of a flow to be too severe a response to a congestion indication, as it can noticeably reduce the flow s user perceived quality [25] In the past few years, many unicast congestion control schemes have been proposed and investigated [12, 16, 28, 23, 3, 18, 22, 25, 20, 29, 9]. The common objective of these studies is to find a good alternative to the congestion control scheme of TCP. Since the dominant Internet traffic is TCP based [26] it is important that new congestion control schemes be TCP friendly. By this, we mean that the sending rate of a non TCP flow should ....

.... mean that the sending rate of a non TCP flow should be approximately the same as that of a TCP flow under the same conditions of round trip time and packet loss [16] The congestion control schemes investigated can be divided into two categories: AIMD based [12, 23, 3, 22, 18] and formula based [16, 28, 25, 20, 29, 9]. Roughly speaking, AIMD based schemes emulate the increase by one and decrease to half window behavior of TCP. Formula based schemes use a stochastic model [16, 17, 19] to derive a formula that expresses the TCP sending rate as a function of packet loss rate, round trip time, and timeout. ....

T. Turletti, S. F. Parisis, and J. Bolot. Experiments with a layered transmission scheme over the Internet. In Proceedings of IEEE INFOCOM '98, San Francisco, California, U.S.A., Mar. 1998.

....paper is to evaluate these properties by analytically and experimentally studying the transient behaviors of several TCP friendly congestion control protocols. Proposed congestion control schemes in the literature fall into two major categories: AIMD based [13, 21, 6, 19, 27, 20] and formula based [16, 24, 22, 17, 10]. For our study, we select TCP [14] and GAIMD [27] as representatives of the AIMD based schemes. GAIMD generalizes TCP by parameterizing the congestion window increase value and decrease ratio. That is, in the congestion avoidance state, the window size is increased by per window of packets ....

Thierry Turletti, Sacha Fosse Parisis, and Jean-Chrysostome Bolot. Experiments with a layered transmission scheme over the Internet. Research Report No 3296, INRIA, November 1997.

....the TCP sending rate in congestion avoidance state is controlled by a congestion window which is halved for every window of data containing a packet drop, and increased by one packet per window of data acknowledged. Recently, many new congestion control protocols were proposed and investigated [4, 6, 13, 11, 1, 14, 7, 10, 12, 8, 15]. The objective of these new congestion protocols is to address the needs of new multimedia applications. We notice that, like TCP, many of these proposals are also based on the AIMD principle. Further, there is even a common belief that AIMD is optimal and is a necessary condition for a ....

Thierry Turletti, Sacha Fosse Parisis, and JeanChrysostome Bolot. Experiments with a layered transmission scheme over the Internet. Research Report No 3296, INRIA, November 1997.

....equation. However, the authors provide no details as to how loss rate and round trip times are determined. 5.1.5 LTS and TFRP Two similar congestion control protocols for the transmission of video streams are presented by Turletti et al. and Tan and Zakhor. The Layered Transmission Scheme (LTS) [29] and the TCP Friendly Transport Protocol TFRP [30] both refrain from join experiments to probe for available bandwidth, using instead the simple TCP equation 1 to adjust the rate. Receivers simply adjust their subscription level to the rate given by the equation. The necessary parameters of loss ....

T. Turletti, S. Parisis, and J. Bolot, "Experiments with a layered transmission scheme over the Internet," Tech. Rep. RR-3296, INRIA, France, Nov. 1997.

....which is discovered through join experiments that entail the addition of layers at well chosen times or detection intervals . A layered transmission scheme in which RLM s join experiments are replaced by explicit estimation of the TCP friendly transmission rate at each receiver is proposed in [58]. The multicast rate control mechanism computes round trip times and estimates packet loss rates at each receiver in order to compute the rate that would be used by an equivalent TCP connection. The authors note, however, that determining round trip times in a multicast environment is a difficult ....

....rapid fluctuations in quality by allowing the server to tradeoff short term improvement in quality for long term smoothing of quality. Our approach to the problem of video streaming focuses on unicast delivery of video and thus differs from the rate adaptation mechanisms developed in [29] and [58], which are applicable only in multicast settings. Our work on streaming video is complementary to the congestion control protocols developed in [34] and [48] as it allows the streaming application to utilize (fair share) available bandwidth efficiently, in a manner that minimizes quality ....

T. Turletti, S. Parisis, and J. Bolot. Experiments with a layered transmission scheme over the internet. Technical Report 3296, INRIA, Sophia Antipolis, France, 1997.

....for the underlying multicast routing protocol as join and leave decisions occur much more frequently. 5.1.3 LTS and TFRP Two similar congestion control protocols for the transmission of video streams are presented by Turletti et al. and Tan and Zakhor. The Layered Transmission Scheme (LTS) [24] and the TCP Friendly Transport Protocol TFRP [25] both refrain from join experiments to probe for available bandwidth, using instead the simple TCP equation 1 to adjust the rate. Receivers simply adjust their subscription level to the rate given by the equation. The necessary parameters of loss ....

T. Turletti, S. Parisis, and J. Bolot, "Experiments with a layered transmission scheme over the internet," Tech. Rep., Technical report RR-3296, INRIA, France, Nov. 1997.

.... as streaming multimedia would find halving the sending rate of a flow to be too severe a response to a congestion indication as it can noticeably reduce the flow s user perceived quality [22] In the last few years, many unicast congestion control protocols have been proposed and investigated [16, 24, 21, 6, 19, 22, 17, 25, 10, 28, 20]. Since the dominant Internet tra#c is TCP based [23] it is important that new congestion control protocols be TCP friendly. By this, we mean that the sending rate of a non TCP flow should be approximately the same as that of a TCP flow under the same conditions of round trip time and packet loss ....

....paper is to evaluate these properties by studying the transient behaviors of several congestion control protocols under three network environment changes. Proposed congestion control protocols in the literature fall into two major categories: AIMD based [21, 6, 19, 28, 20] and 2 formula based [16, 24, 22, 17, 10]. For our study, we select TCP [13] and GAIMD [28] as representatives of the first category. GAIMD generalizes TCP by parameterizing the congestion window increase value and decrease ratio. That is, in the congestion avoidance state, the window size is increased by # per window of packets ....

Thierry Turletti, Sacha Fosse Parisis, and Jean-Chrysostome Bolot. Experiments with a layered transmission scheme over the Internet. Research Report No 3296, INRIA, November 1997.

.... For instance, using waveform coded audio, lost ADUs lead to distortions caused by signal interruptions [10, 11] while the available data bandwidth is perceived by the amount of quantization noise (assuming that the data bandwidth is varied by adjusting the bit resolution of the audio signal [12]) The in uence of loss distribution can be captured by equation 4. While some single missing ADUs can be tolerated up to a certain degree, a gap of consecutive ADUs typically leads to a higher loss impact (assuming the same mean loss rate for both cases) depending on the coding scheme. Therefore, ....

T. Turletti, S. Fosse-Parisis, and J.-C. Bolot, \Experiments with a layered transmission scheme over the Internet," Research Report 3296, INRIA, November 1997.

.... embedded bit stream that allows decoding at multiple rates [12, 13, 14, 4] Under layered multicast, different layers of a scalable video are carried in di#erent multicast groups so that receivers can individually subscribe or unsubscribe to the appropriate multicast groups to achieve flow control [11, 9, 15, 16, 17, 18]; thus, the more layers a receiver subscribes to, the higher the quality of the received video. Generally, producing an embedded bit stream for scalable compression results in lower compression e#ciency than its non scalable counterpart at the same bit rates [19] Nevertheless, layered multicast ....

....techniques may be e#ective for a single multicast session, there are no existing mechanisms to coordinate probes across di#erent sessions, leading to potentially unfair allocation of bandwidth. A more recently proposed scheme, equation based rate control (ERC) does not involve probing [17, 4]. Instead, the available bandwidth is computed directly at the receivers based on measured quantities such as packet loss rate. One characteristic of ERC is its ability to allocate bandwidth fairly among competing sessions [40] For example, two simultaneous video multicast sessions which share ....

[Article contains additional citation context not shown here]

T. Turletti, S. Parisis, and J. Bolot, "Experiments with a layered transmission scheme over the Internet," Tech. Rep., INRIA, http://www.inria.fr/RRRT/RR-3296.html, Nov. 1997.

.... as streaming multimedia would find halving the sending rate of a flow to be too severe a response to a congestion indication as it can noticeably reduce the flow s user perceived quality [3] In the last few years, many unicast congestion control protocols have been proposed and investigated [4] [5], 6] 7] 8] 3] 9] 10] 11] 12] 13] Since the dominant Internet traffic is TCP based [14] it is important that new congestion control protocols be TCP friendly. By this, we mean that the sending rate of a non TCP flow should be approximately the same as that of a TCP flow under the ....

....is to evaluate these properties by studying the transient behaviors of several congestion control protocols under three network environment changes. Proposed congestion control protocols in the literature fall into two major categories: AIMD based [6] 7] 8] 12] 13] and formulabased [4] [5], 3] 9] 11] For our study, we select TCP [2] and GAIMD [12] as representatives of the first category. GAIMD generalizes TCP by parameterizing the congestion window increase value and decrease ratio. That is, in the congestion avoidance state, the window size is increased by # per window of ....

Thierry Turletti, Sacha Fosse Parisis, and Jean-Chrysostome Bolot, "Experiments with a layered transmission scheme over the Internet," Research Report No 3296, INRIA, Nov. 1997.

....applications in terrestrial networks. For example, in [13] a TCPlike scheme that does not perform retransmissions is proposed. The Streaming Control Protocol (SCP) is introduced in [6] SCP is a modified version of TCP [14] 15] that performs TCP Vegas like rate adjustment [23] In [19] and [28] transmission rate is adjusted based on TCP throughput model [11] 20] In [27] and [23] two rate adaptation protocols, namely LDA and RAP, are presented. Both of them perform flow control for real time streams by means of mechanisms very similar to those of TCP [14] Packet losses are the only ....

T. Turletti, S. F. Prisis, and J.-C. Bolot, "Experiments with a Layered Transmission Scheme over the Internet," Rapport de recherche 3296, INRIA, November 1997.

....own multicast group. The set of layers comprising a video stream constitute a session. Receivers of a session subscribe to as many layers as network conditions and receiver capabilities allow. Several schemes for layered multicast have been proposed, including [MJV96] WSS97] LPA98] VRC98] [TPB97]. Layered multicast protocols typically use a receiver driven approach in which the end systems decide which layers should be delivered. Protocols such as Receiver Driven Layered Multicast (RLM) protocol [MJV96] have been analyzed and evaluated by several researchers [MJV96] BBS98] GGHS99] ....

....to a larger number of layers will surrender them more quickly than a receiver with a smaller subscription set. Instability is not addressed, and may be significant given the join leave overheads of thin layers. The issue of fairness between RLM and TCP traffic has also been studied [VRC98] [TPB97]. The idea in [VRC98] is for receivers to use a join leave strategy for congestion control which mimics the behavior of TCP. This relies on making appropriate choices of layer bandwidths and the time delay between trying to increase subscription. In [TPB97] each receiver tries to determine the ....

[Article contains additional citation context not shown here]

T. Turletti, S. F. Parisis, and J-C. Bolot. Experiments with a layered transmission scheme over the Internet. Technical Report 3296, INRIA Sophia Antipolis, France, November 1997.

....fairness among flows with different RTTs. The rate based approaches, on the other hand, adjust the packet transmission rate (or interval) in an additive increase multiplicative decrease fashion as in TCP [13, 15, 21] or in compliance with the TCP throughput characterization analytically derived [11, 19, 22], whenever an acknowledgment is received or a packet loss is detected. The major advantage of rate based approaches is that the notion of rate is better suited for continuous media applications and the rate attainable can be controlled to be independent of RTT. The major drawback of rate based ....

....for continuous media applications and the rate attainable can be controlled to be independent of RTT. The major drawback of rate based approaches is that instantaneous and average rates must be on line measured. All the proposed rate based approaches focus on unicasts perhaps except for [19, 21, 22]. We will provide a more detailed summary of related work in Section 5. 2 The rest of the paper is organized as follows. In Section 2, we provide an overview of RACCOOM and its components. In Section 3, we give a detailed description of each of the RACCOOM components. In Section 4, we use ....

[Article contains additional citation context not shown here]

T. Turletti, S. Parisis and J. Bolot. Experiments with a Layered Transmission Scheme over the Internet Technical report RR-3296, INRIA, France, 1996.

....clients and depending upon the network bandwidth available to a particular client or its decoding capability, it receives data from the enhancement layers to enhance the quality of its presentation. Examples of layered encoding schemes for audio and video are discussed by McCanne [20] and Turletti [41]. TLTCP exploits the property of error resilience, since it does not guarantee reliable data delivery. It also attempts to adhere to the time lined nature of the continuous media data by keeping track of and discarding obsolete data. All the encoding schemes described above are error resilient ....

T. Turletti, S. F. Parisis, and J-C. Bolot. Experiments with a layered transmission scheme over the internet. Technical Report 3296, INRIA, Sophia Antipolis, France, November 1997.

....if the transmission window [t lead (s) t lag (s) becomes sufficiently large. The advantages of this approach are total compatibility with TCP, effective flow and congestion control, and a proven, stable protocol. An alternative approach is to use TCP friendly equation based congestion control [62, 63, 64, 65, 66, 67, 68, 69]. In this approach, the congestion control mechanism makes a local estimate of TCP s long term average transmission rate using an equation for the number of seconds between data units such as [68] T = R p 2ffl R =3 3(R 4oe R )ffl R (1 32ffl 2 R ) p 3ffl R =8; 16) where ffl R , R , ....

T. Turletti, S.F. Parisis, and J.-C. Bolot. Experiments with a layered transmission scheme over the Internet. Technical Report 3296, INRIA, Sophia Antipolis, France, November 1997.

....source quench messages to inform the involved sources of the network congestion state. The fairness achieved is a trade o between TCP like and Max Min fairness. 3.2. 2 Receiver Oriented Congestion Control This class includes those proposals using hierarchical coding for continuous streams as [66] and exponential layering bandwidth subscribing N layers = bandwidth first layer 2 (N 1) for continuous streams and bulk data [67] A receiver belonging to more less layer groups of the same session receives more less data or at a faster slower rate. The receivers compute the rate using packet ....

T. Turletti, S.F. Parisis, and J. Bolot. Experiments with a layered transmission scheme over the internet. In IEEE INFOCOM'98, February 1998.

....known delay and loss conditions. r TCP = 1:22 #M # # p l (1) with M as the maximum packet length, # as the round trip time of the connection and l as the average loss measured during the lifetime of the connection. Based on this estimation, Mahdavi et al. 14] and Turletti et al. [24] propose end to end, TCPfriendly congestion control schemes in which the end systems measure losses and delays in the network and restrict their transmission rates to the value estimated by Eq. 1. Floyd et al. 9] describe a mechanism in which routers identify connections that use more bandwidth ....

T. Turletti, S. F. Prisis, and J.-C. Bolot. Experiments with a layered transmission scheme over the Internet. Rapport de recherche 3296, INRIA, Nov. 1997.

No context found.

T. Turletti, S.F. Parisis, and J. Bolot, "Experiments with a layered transmission scheme over Internet", INRIA Research Report No 3296, Nov. 1997.

No context found.

T. Turletti, S.F. Parisis and J. Bolot, "Experiments with a Layered Transmission Scheme over the Internet," INRIA Research Report no 3296, Nov. 1997.

No context found.

T. Turletti, S.F. Parisis, and J. Bolot, "Experiments with a layered transmission scheme over Internet", INRIA Research Report No 3296, Nov. 1997.

No context found.

T. Turletti, S.F. Parisis and J. Bolot, "Experiments with a Layered Transmission Scheme over the Internet", INriA Research Report No 3296, Nov. 1997.

No context found.

T. Turletti, S. F. Parisis, J. Bolot, "Experiments with a Layered Transmission Scheme over the Internet", INRIA Research Report No 3296, Nov. 97.

No context found.

T. Turletti, S. Parisis, and J. Bolot, "Experiments With a Layered Transmission Scheme Over the Internet," INRIA, Tech. Rep. N'3296, Nov. 1997.

No context found.

Thierry Turletti, Sacha Fosse Parisis, and Jean-Chrysostome Bolot. Experiments with a layered transmission scheme over the Internet. Research Report No 3296, INRIA, November 1997.

No context found.

T. Turletti, S. Parisis, and J. Bolot, "Experiments with a layered transmission scheme over the Internet," INRIA, Sophia Antipolis, France, Tech. Rep. RR-3296, Nov. 1997.

No context found.

T. Turletti, S.F. Parisis, and J.-C. Bolot. Experiments with a layered transmission scheme over the Internet. Technical Report 3296, INRIA, Sophia Antipolis, France, November 1997.

No context found.

Thierry Turletti, Sacha Fosse Parisis, and Jean-Chrysostome Bolot. Experiments with a layered transmission scheme over the Internet. Research Report No 3296, INRIA, November 1997.

No context found.

T. Turletti, S. F. Parisis, and J.-C. Bolot, "Experiments with a layered transmission scheme over the Internet," INRIA Research Report 3296, November 1997.

No context found.

T. Turletti, S. F. Parisis, and J.-C. Bolot, "Experiments with a layered transmission scheme over the Internet," INRIA Research Report 3296, November 1997.

No context found.

T. Turletti, S. F. Parisis, and J.-C. Bolot, "Experiments with a layered transmission scheme over the Internet," INRIA Research Report 3296, November 1997.

No context found.

T. Turletti, S. Fosse-Parisis, and J. Bolot. Experiments with a layered transmission scheme over the internet. Research Report No. 3296, INRIA, France, Novemeber 1997.

No context found.

T. Turletti, S. F. Parisis, and J.-C. Bolot, "Experiments with a layered transmission scheme over the Internet," INRIA Research Report 3296, November 1997.

No context found.

T. Turletti, S. F. Parisis, and J.-C. Bolot, "Experiments with a layered transmission scheme over the Internet," INRIA Research Report 3296, November 1997.

No context found.

T. Turletti, S.F. Parisis, J-C. Bolot, "Experiments with a Layered Transmission Scheme over the Internet", in INRIA Research Report No 3296, Nov. 97.

No context found.

T. Turletti, S.F. Parisis, and J.-C. Bolot. Experiments with a Layered Transmission Scheme over the Internet. INRIA Research Report No. 3296, Nov. 1997.

No context found.

T. rletti, S.F. Parisis, and J.-C. Bolot, Experiments with a Layered Transmission Scheme over the Internet, INRIA Research Report No. 3296, Nov. 1997.

No context found.

T. Turletti, S.F. Parisis, and J. Bolot, "Experiments with a layered transmission scheme over the internet", IEEE INFOCOM'98, Feb. 1998.

No context found.

T. Turletti, S. F. Parisis, and J.-C. Bolot, "Experiments with a layered transmission scheme over the Internet," INRIA Research Report 3296, November 1997.

No context found.

T. Turletti, S. Fosse-Parisis, and J.-C. Bolot, #Experiments with a layered transmission scheme over the Internet, " Technical Report 3296, INRIA, November 1997.

No context found.

T. Turletti, S. F. Parisis, and J.-C. Bolot, "Experiments with a layered transmission scheme over the Internet," INRIA Research Report 3296 , November 1997.

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