Re: [OT] SR ARQ Markov Model
On Mon, May 07, 2001, MaD dUCK wrote:
> hey guys,
> sorry to be posting this absolutely off-topic... reason is that I am
> desperate (this is academic research, but I ain't cheatin' by asking
> you) and of all my mailing lists, you guys are simply the smartest.
>
> i am looking for a Markov Model of the selective-repeat ARQ used in
> TCP. but i can't find one anywhere. problem is that it's a little more
> than simple M/M/1 but noone in this world seems to have ever modeled
> it and published the stuff on the web... and i can't believe that so i
> am asking you for pointers... anything?
>
> thanks!
Martin,
Attached is the output of running a search on 'markov model' and 'tcp'
on INSPEC. It found 23 journal articles, and the attachment lists
references and abstracts for each. INSPEC, of course, is available at
'http://firstsearch.oclc.org', but you generally need your academic
institution to set up access, which is not readily available to
general public, I believe.
The only tangential experience I've personally had with markov models
was when I was working on cluster algorithms in monte carlo
simulations (2D Ising Model, if you care). I've generally found Kurt
Binder's books very good in this field, so thought I might mention 'A
Guide to Monte Carlo Simulations in Statistical Physics', if you're
not already familiar with it. But then again, I'm not sure how much
this has on Markov specifically (as Markov models are more
mathematics, than physics, IMHO).
Anyway, hope this helps and good luck with the research.
Take care,
Daniel
--
Daniel A. Freedman
Laboratory for Atomic and Solid State Physics
Department of Physics
Cornell University
Database: INSPEC
Query: (kw: markov and kw: model) and kw: tcp
Database: INSPEC
Copyright: Copyright 2001, IEE
Accession No: 6900029 INSPEC Abstract No: B2001-05-6150M-108
C2001-05-5640-071
Author(s): Barakat, C.; Altman, E.
Editor: Crowcroft, J.; Roberts, J.; Smirnov, M.I.
Corp Source: Inst. Nat. de Recherche en Inf. et Autom., Sophia
Antipolis, France
Title: A Markovian model for TCP analysis in a
Differentiated Services network
Source: Quality of Future Internet Services. First COST 263
International Workshop, QofIS 2000. Proceedings (Lecture Notes in
Computer Science Vol.1922) p. xi+368, 55-67
Publication: Germany
ISBN: 3540410767
Language: English
Publisher: Springer-Verlag; Berlin, Germany
Sponsor: GMD FOKUS; Deutsche Telekom T-Nova; Cisco Syst.;
Nokia; ERCIM; IST Program, Commission of the Eur. Communities
Conference: Quality of Future Internet Services. First COST 263
International Workshop, QofIS 2000
Conference Loc: Berlin, Germany; 25-26 Sept. 2000
Year: 2000
Treatment: P Practical; T Theoretical or Mathematical
Record Type: Conference Paper
Abstract: Many schemes have been proposed to support TCP
traffic in a Differentiated Services network. We present an analytical
model to study the performance of these schemes. The model is based on
a Markovian fluid approach. We provide first a general version of the
model, then we specify it to the different proposed schemes. For each
scheme, we calculate the throughput achieved by a TCP connection. We
compare then their service differentiation capacity under different
subscription levels, different reservations, and different round-trip
times (18 Refs.)
Descriptor(s): Internet; Markov processes; quality of service;
telecommunication traffic; transport protocols
Identifier: Markovian model; TCP analysis; Differentiated
Services network; TCP traffic; scheme performance; Markovian fluid
approach; throughput; service differentiation; quality of service;
Internet
Class Code: B6150M (Protocols); B6210L (Computer
communications); B0240J; C5640 (Protocols); C5620W (Other computer
networks); C1140J
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 2001, IEE
Accession No: 6894210 INSPEC Abstract No: B2001-05-6250F-374
Author(s): Yong Bai; Ogielski, A.T.; Gang Wu
Corp Source: Wireless Inf. Network Lab., Rutgers Univ.,
Piscataway, NJ, USA
Title: TCP over IS-707
Source: 10th International Symposium on Personal, Indoor
and Mobile Radio Communications (PIMRC'99). Proceedings p. 3 vol.
xxiii+1560, 1253-7 vol.3
Publication: Japan
Language: English
Publisher: Osaka Univ; Odsaka, Japan
Conference: Proceedings of PIMRC'99: International Symposium on
Personal and Indoor Mobile Radio Communications
Conference Loc: Osaka, Japan; 12-15 Sept. 1999
Year: 1999
Treatment: P Practical
Record Type: Conference Paper
Abstract: Mobile wireless Internet access services face a
problem of mitigating the time-varying, correlated frame losses in
fading radio channels. The losses may be compounded by detrimental
interactions between separate error recovery mechanisms in the link and
transport protocol layers. In this work we investigate by simulations
the performance of TCP Reno over IS-707, a radio link protocol (RLP)
standard for spread spectrum digital cellular systems, and study the
interactions of these protocols in the high loss regime. We show the
improvement of TCP goodput by a fast RLP frame error recovery scheme
based on using the IS-707 idle frames. We generalize the problem of
characterizing the performance of user applications accessing the
Internet over wireless data services by investigating the distributions
of TCP goodput at multiple time scales, that give more insight than
long-time averages. Results are presented for a range of frame error
rates and normalized Doppler frequencies using a standard Markov model
of a fading radio channel (12 Refs.)
Descriptor(s): cellular radio; digital radio; error correction;
error statistics; fading channels; Internet; Markov processes; radio
access networks; spread spectrum communication; telecommunication
standards; telecommunication traffic; transport protocols
Identifier: TCP Reno over IS-707; mobile wireless Internet
access services; time-varying correlated frame losses; fading radio
channels; radio link protocol standard; transport protocol layer;
spread spectrum digital cellular systems; TCP goodput; frame error
recovery scheme; IS-707 idle frames; wireless data services; frame
error rates; normalized Doppler frequencies; standard Markov model
Class Code: B6250F (Mobile radio systems); B6210L (Computer
communications); B6250B; B6150M (Protocols); B0240J
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 2001, IEE
Accession No: 6873286 INSPEC Abstract No: B2001-04-6150M-094
C2001-04-5640-064
Author(s): Marsan, M.A.; Casetti, C.; Gaeta, R.; Meo, M.
Editor: Obaidat, M.S.; Davoli, F.; Marsan, M.A.
Corp Source: Dipartimento di Elettronica, Politecnico di Torino,
Italy
Title: An approximate GSPN model for the accurate
performance analysis of correlated TCP connections
Source: Proceedings of the 2000 Symposium on Performance
Evaluation of Computer and Telecommunication Systems p. xii+570, 154-62
Publication: USA
ISBN: 1565552067
Language: English
Publisher: SCS; San Diego, CA, USA
Sponsor: SCS
Conference: Proceedings of SPECTS 2000. SCS Symposium on
Performance Evaluation of Computer and Telecommunication Systems
Conference Loc: Vancouver, BC, Canada; 16-20 July 2000
Year: 2000
Treatment: A Application; P Practical
Record Type: Conference Paper
Abstract: A number of analytical models of the TCP behavior
have been presented in the recent literature; most of them consider the
interaction among TCP connections sharing one bottleneck link, but
assume that these connections behave independently of each other. In
this paper we develop a new model to describe the behavior of a number
of TCP connections that share a bottleneck link, considering the
effects of the synchronization among connections. The model is based on
Markovian assumptions and on two GSPN (Generalized Stochastic Petri
Net) descriptions of the system. From the two GSPN descriptions, using
a fixed point algorithm, several interesting performance metrics
referring to TCP connections can be derived. A validation of our
modeling approach is provided through extensive comparisons of the
model's results with the output of the LBL simulator, "ns version 2".
Comparisons show that our model succeeds in providing an accurate
representation of the behavior of TCP connections under several
different settings (10 Refs.)
Descriptor(s): Internet; Markov processes; performance evaluation;
Petri nets; stochastic processes; synchronisation; transport protocols
Identifier: approximate GSPN model; performance analysis;
correlated TCP connections; analytical models; synchronization;
Markovian assumptions; generalized stochastic Petri net; performance
metrics
Class Code: B6150M (Protocols); B6210L (Computer
communications); B0240J; C5640 (Protocols); C5620W (Other computer
networks); C7210N; C1140J; C5470 (Performance evaluation and testing);
C5670 (Network performance)
------------------------------------------------------------------------
Database: INSPEC
Ownership: FirstSearch indicates your institution owns the
item. Libraries that Own Item: 155
Copyright: Copyright 2000, IEE
Accession No: 6749615 INSPEC Abstract No: C2000-12-7210N-073
Author(s): Marsan, M.A.; Casetti, C.; Gaeta, R.; Meo, M.
Corp Source: Dept. of Electron., Politecnico di Torino, Italy
Title: Performance analysis of TCP connections sharing a
congested Internet link?
Source: Performance Evaluation, vol.42, no.2-3 p. 109-27
ISSN: 0166-5316
CODEN: PEEVD9
Publication: Netherlands
Language: English
Publisher: Elsevier
Year: Oct. 2000
Copyright No: 0166-5316/2000/$20.00
Treatment: P Practical; T Theoretical or Mathematical
Record Type: Journal Paper
Abstract: Many Internet channels are today heavily loaded
with traffic, most of which is due to TCP connections supporting
applications, such as web browsing. We develop a new model to describe
the behavior of a number of TCP connections that share a bottleneck
link, considering the effects of the synchronization among connections.
The model is based on Markovian assumptions and on two GSPN
(generalized stochastic Petri net) descriptions of the system. From the
two GSPN descriptions, using a fixed point algorithm, several
interesting performance metrics referring to the TCP connections can be
derived. A validation of our modeling approach is provided through
extensive comparisons of the model's results with the output of the LBL
simulator. Comparisons show that our model succeeds in providing an
accurate representation of the behavior of TCP connections under
several different settings (13 Refs.)
Descriptor(s): Internet; Markov processes; Petri nets; software
performance evaluation; synchronisation; transport protocols
Identifier: TCP connections; Internet link; performance
analysis; synchronization; Markovian model; stochastic Petri net
Class Code: C7210N; C5640 (Protocols); C6110B (Software
engineering techniques); C1160 (Combinatorial mathematics); C1140J
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 2000, IEE
Accession No: 6702773 INSPEC Abstract No: B2000-10-6150M-065
C2000-10-5640-045
Author(s): Abouzeid, A.A.; Azizoglu, M.; Roy, S.
Editor: Banerjee, P.; Prasanna, V.K.; Sinha, B.P.
Corp Source: Dept. of Electr. Eng., Washington Univ., Seattle,
WA, USA
Title: Stochastic modeling of TCP/IP over random loss
channels
Source: High Performance Computing - HiPC'99. 6th
International Conference. Proceedings. (Lecture Notes in Computer
Science Vol.1745) p. xxii+412, 309-14
Publication: Germany
ISBN: 3540669078
Language: English
Publisher: Springer-Verlag; Berlin, Germany
Conference: Proceedings of 6th International Conference on High
Performance Computing (HiPC'99) - Mobile Computing for this Millenium
Conference Loc: Calcutta, India; 17-20 Dec. 1999
Year: 1999
Treatment: T Theoretical or Mathematical
Record Type: Conference Paper
Abstract: An analytical framework for modeling the
performance of a single TCP session in the presence of random packet
loss is presented that is based on a semi-Markov model for the window
size evolution. The model predicts the throughput for LAN/WAN (low and
high bandwidth-delay products) with good accuracy, as compared against
simulation results with the ns simulator. Generally, higher speed
channels are found be more vulnerable to random loss than slower
channels, especially for moderate to high loss rates (10 Refs.)
Descriptor(s): local area networks; Markov processes; modelling;
simulation; software performance evaluation; telecommunication
channels; telecommunication traffic; transport protocols; wide area
networks
Identifier: stochastic modeling; TCP/IP; random loss channels;
performance modeling; random packet loss; semi-Markov model; window
size evolution; throughput; LAN; WAN; bandwidth-delay products; ns
simulator; channel speed; loss rates
Class Code: B6150M (Protocols); B0240J; B6210L (Computer
communications); C5640 (Protocols); C1140J; C5620L (Local area
networks); C5620W (Other computer networks)
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 2000, IEE
Accession No: 6690260 INSPEC Abstract No: B2000-10-6150M-018
Author(s): Dongjie Huang; Shi, J.J.
Corp Source: Ericsson Inc., Richardson, TX, USA
Title: TCP with adaptive radio link
Source: Virginia Tech's Ninth Symposium on Wireless
Personal Communications. Proceedings p. 290, 211-19
Publication: USA
Language: English
Publisher: Virginia Tech; Blacksburg, VA, USA
Sponsor: Virginia Tech.; Army Res. Office; AT&T Corp.;
Bellsouth Cellular; Comcast CellularCommun.; DATUM; et al
Conference: Proceedings of the 9th Ann Virginia Technical
Symposium on Wireless Personal Communication
Conference Loc: Blacksburg, VA, USA; 2-4 June 1999
Year: 1999
Treatment: T Theoretical or Mathematical
Record Type: Conference Paper
Abstract: The performance of TCP over wireless links could
degrade due to handoff, high bit error rate and long roundtrip delay on
the air interface. The conventional radio link protocol uses fixed
channel coding and ARQ to mitigate impairment over wireless channels.
Adaptive channel coding with the use of punctured convolutional codes
has been proposed for its ability to adapt to channel quality and
maintain a high level of system throughput. In this paper, we study TCP
over adaptive RLP based on estimated channel condition. We consider a
Markov channel model and evaluate the performance of TCP with adaptive
RLP in such an environment. We also evaluate the existing channel
quality measurement in wireless systems and propose a methodology to
update RLP channel coding schemes based on filtered channel
measurement. We use simulation to calculate the throughput of the TCP
system and determine the effectiveness of adaptive channel coding with
respect to different fading rates (10 Refs.)
Descriptor(s): adaptive codes; channel coding; convolutional
codes; data communication; error statistics; fading channels; Markov
processes; mobile radio; packet radio networks; parameter estimation;
radio links; telecommunication traffic; transport protocols
Identifier: adaptive radio link; TCP; adaptive RLP; radio link
protocol; channel estimation; Markov channel model; performance
evaluation; channel quality measurement; filtered channel measurement;
simulation; throughput; adaptive channel coding; fading rates;
punctured convolutional codes; bit error rate; roundtrip delay;
handoff; air interface
Class Code: B6150M (Protocols); B6250F (Mobile radio systems);
B0240J
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 2000, IEE
Accession No: 6678098 INSPEC Abstract No: B2000-09-6250-032
Author(s): Huang, D.; Shi, J.J.
Corp Source: Ericsson Inc., Richardson, TX, USA
Title: TCP over packet radio link with adaptive channel
coding
Source: VTC2000-Spring. 2000 IEEE 51st Vehicular Technology
Conference Proceedings (Cat. No.00CH37026) p. 3 vol. (lvi+2577), 790-4
vol.2
Publication: USA
ISBN: 0780357183
Language: English
Publisher: IEEE; Piscataway, NJ, USA
Conference: 2000 IEEE 51st Vehicular Technology Conference.
Proceedings. VTC2000-Springer
Conference Loc: Tokyo, Japan; 15-18 May 2000
Year: 2000
Copyright No: 0 7803 5718 3/2000/$10.00
Treatment: T Theoretical or Mathematical
Record Type: Conference Paper
Abstract: A wireless packet data communications system that
uses adaptive channel coding on the radio link is studied. A channel
coding adaptation mechanism based on recommendation from the receiver
is proposed. A traffic model that exhibits self-similarity is
considered for packet data applications. The radio channel is modeled
as a Rayleigh fading channel. The performance of TCP over such an
adaptive radio link is obtained through analysis and simulation. The
result indicates significant improvement in throughput and delay
performance for TCP over adaptive packet radio link. To gain insight to
dimensioning of wireless packet networks. TCP throughput and delay are
studied with respect to the number of users and file size. System
utilization is analyzed and theoretical bounds are derived and compared
with simulation results (11 Refs.)
Descriptor(s): adaptive codes; channel coding; data communication;
delays; fractals; Markov processes; packet radio networks; radio links;
radio receivers; telecommunication traffic; transport protocols
Identifier: adaptive channel coding; wireless packet data
communications system; receiver; traffic model; self-similarity; packet
data applications; Rayleigh fading channel; radio channel; adaptive
radio link; simulation; throughput performance; delay performance;
adaptive packet radio link; wireless packet network dimensioning; TCP
throughput; file size; simulation results; Markov fading channel model
Class Code: B6250 (Radio links and equipment); B6150M
(Protocols); B6120B (Codes); B0240J
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 2000, IEE
Accession No: 6559231 INSPEC Abstract No: B2000-05-6150M-123
C2000-05-5640-104
Author(s): Abouzeid, A.A.; Roy, S.; Azizoglu, M.
Corp Source: Dept. of Electr. Eng., Washington Univ., Seattle,
WA, USA
Title: Stochastic modeling of TCP over lossy links
Source: Proceedings IEEE INFOCOM 2000. Conference on
Computer Communications. Nineteenth Annual Joint Conference of the IEEE
Computer and Communications Societies (Cat. No.00CH37064) p. 3
vol.(xxvi+1826), 1724-33 vol.3
Publication: USA
ISBN: 0780358805
Language: English
Publisher: IEEE; Piscataway, NJ, USA
Sponsor: IEEE Comput. Soc.; IEEE Commun. Soc
Conference: Proceedings IEEE INFOCOM 2000. Conference on
Computer Communications. Nineteenth Annual Joint Conference of the IEEE
Computer and Communications Societies
Conference Loc: Tel Aviv, Israel; 26-30 March 2000
Year: 2000
Copyright No: 0 7803 5880 5/2000/$10.00
Treatment: T Theoretical or Mathematical
Record Type: Conference Paper
Abstract: An analytical framework for modeling the
performance of a single TCP session in the presence of random packet
loss is presented. A Markovian approach is developed that allows us to
study both memoryless channels (IID packet loss) and channels with
memory (correlated packet loss) modeled by a two-state continuous-time
Gilbert model. The analytical results are validated against results
using the ns simulator. It is shown that the model predicts throughput
for LAN/WAN (low and high bandwidth-delay products) with good accuracy.
Further, throughput for the IID loss model is found to be relatively
insensitive to the probability density function (PDF) of the loss
inter-arrival process. For channels with memory, we present an
empirically validated rule of thumb to categorize the channel
transition frequency (11 Refs.)
Descriptor(s): continuous time systems; correlation theory; local
area networks; Markov processes; memoryless systems; packet radio
networks; performance evaluation; probability; queueing theory;
telecommunication traffic; transport protocols; wide area networks
Identifier: stochastic modeling; TCP; lossy links; performance;
random packet loss; Markovian approach; memoryless channels; correlated
packet loss; two-state continuous-time Gilbert model; ns simulator;
throughput; LAN; WAN; IID packet loss; probability density function;
PDF; loss inter-arrival process; channel transition frequency
Class Code: B6150M (Protocols); B0240J; B6210L (Computer
communications); B0240C (Queueing theory); B6150J (Queueing systems);
C5640 (Protocols); C1140J; C5670 (Network performance); C5620L (Local
area networks); C5620W (Other computer networks); C1140C (Queueing
theory)
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 2000, IEE
Accession No: 6559028 INSPEC Abstract No: B2000-05-6150M-091
Author(s): Zorzi, M.; Rao, R.R.
Corp Source: Dipt. di Ingegneria, Ferrara Univ., Italy
Title: Is TCP energy efficient?
Source: 1999 IEEE International Workshop on Mobile
Multimedia Communications (MoMuC'99) (Cat. No.99EX384) p. xii+390,
198-201
Publication: USA
ISBN: 0780359046
Language: English
Publisher: IEEE; Piscataway, NJ, USA
Sponsor: IEEE Commun. Soc. Tech. Committees on Comput.
Commun., Multimedia Commun., Internet, & Personal Commun.; ACM SIGCOMM,
ACM SIGMOBILE; Center for Wireless Commun., Univ. California, San
Diego; Center for Telecommun. Res., Coloumbia Univ
Conference: 1999 IEEE International Workshop on Mobile
Multimedia Communications
Conference Loc: San Diego, CA, USA; 15-17 Nov. 1999
Year: 1999
Copyright No: 0 7803 5904 6/99/$10.00
Treatment: T Theoretical or Mathematical
Record Type: Conference Paper
Abstract: We analyze the energy consumption performance of
various versions of TCP, for bulk data transfer in an environment where
channel errors are correlated. We focus on a single wireless TCP
connection and model the packet loss or error process as a first-order
Markov chain. We compute the throughput and energy performance of
various versions of TCP. The main findings of this study are that (1)
error correlations significantly affect the energy performance of TCP
consistent with analogous conclusions for throughput; and (2) the
congestion control mechanism implemented by TCP does a good job at
saving energy as well, by backing off and idling during error bursts.
An interesting conclusion is that, unlike throughput, the energy
efficiency metric may be very sensitive to the TCP version used and to
the choice of the protocol parameters (13 Refs.)
Descriptor(s): correlation methods; data communication; energy
conservation; Markov processes; packet radio networks;
telecommunication congestion control; transport protocols
Identifier: energy consumption performance; bulk data transfer;
correlated channel errors; wireless TCP connection; packet loss; error
process; first-order Markov chain; throughput; energy performance;
congestion control mechanism; error bursts; energy efficiency metric;
protocol parameters; wireless communications
Class Code: B6150M (Protocols); B0240J
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 2000, IEE
Accession No: 6551997 INSPEC Abstract No: B2000-05-6150M-046
C2000-05-5640-037
Author(s): Casetti, C.; Meo, M.
Corp Source: Dipt. di Elettronica, Politecnico di Torino, Italy
Title: A new approach to model the stationary behavior of
TCP connections
Source: Proceedings IEEE INFOCOM 2000. Conference on
Computer Communications. Nineteenth Annual Joint Conference of the IEEE
Computer and Communications Societies (Cat. No.00CH37064) p. 3
vol.(xxvi+1826), 367-75 vol.1
Publication: USA
ISBN: 0780358805
Language: English
Publisher: IEEE; Piscataway, NJ, USA
Sponsor: IEEE Comput. Soc.; IEEE Commun. Soc
Conference: Proceedings IEEE INFOCOM 2000. Conference on
Computer Communications. Nineteenth Annual Joint Conference of the IEEE
Computer and Communications Societies
Conference Loc: Tel Aviv, Israel; 26-30 March 2000
Year: 2000
Copyright No: 0 7803 5880 5/2000/$10.00
Treatment: T Theoretical or Mathematical
Record Type: Conference Paper
Abstract: In this paper, we outline a methodology that can be
applied to model the behavior of TCP flows. The proposed methodology
stems from a Markovian model of a single TCP source, and eventually
considers the superposition and interaction of several such sources
using standard queueing analysis techniques. Our approach allows the
evaluation of such performance indices as throughput, queueing delay
and packet loss of TCP flows. The results obtained through our model
are validated by means of simulation, under several topology and
traffic settings (10 Refs.)
Descriptor(s): delays; Markov processes; network topology;
queueing theory; telecommunication traffic; transport protocols
Identifier: stationary behavior; TCP connections; TCP flows;
Markovian model; queueing analysis; performance evaluation; throughput;
queueing delay; packet loss; topology; traffic; simulation
Class Code: B6150M (Protocols); B0240J; B0240C (Queueing
theory); B6150J (Queueing systems); C5640 (Protocols); C1140J; C1140C
(Queueing theory); C5670 (Network performance)
------------------------------------------------------------------------
Database: INSPEC
Ownership: Check the catalogs in your library. Libraries that
Own Item: 29
Copyright: Copyright 2000, IEE
Accession No: 6524163 INSPEC Abstract No: B2000-04-6150-003
C2000-04-7410F-081
Author(s): Ajmone Marsan, M.; De Souza e Silva, E.; Lo Cigno,
R.; Meo, M.
Corp Source: Dipartimento di Elettronica, Politecnico di Torino,
Italy
Title: A Markovian model for TCP over ATM
Source: Telecommunication Systems - Modeling, Analysis,
Design and Management, vol.12, no.4 p. 341-68
ISSN: 1018-4864
CODEN: TESYEV
Publication: Netherlands
Language: English
Publisher: Baltzer
Year: 1999
Treatment: A Application; T Theoretical or Mathematical
Record Type: Journal Paper
Abstract: This paper focuses on the development of an
approximate Markov chain model and a detailed simulation model for the
performance analysis of TCP connections supporting long file transfers
within high-speed ATM networks. The Markov model construction is first
illustrated, emphasizing the simplifying assumptions introduced to
maintain an acceptable complexity in the performance study. Then the
model validation is described, comparing the analytical results with
those produced by a very detailed simulation setup in which the
officially distributed BSD 4.3 TCP-reno code was adapted to run on top
of a software tool for the simulation of ATM networks. Numerical
results indicate that the analytical model is able to capture the
overall behavior of the system with respect to the performance measures
we consider, with a cost which is orders of magnitude less than the one
necessary to obtain reliable estimates from the detailed simulation
model (16 Refs.)
Descriptor(s): asynchronous transfer mode; broadband networks;
continuous time systems; Markov processes; software tools;
telecommunication computing; transport protocols
Identifier: Markovian model; TCP over ATM; approximate Markov
chain model; detailed simulation model; performance analysis; TCP
connections; long file transfers; high-speed ATM networks; model
validation; BSD 4.3 TCP-reno code; software tool; ATM networks;
performance measures; continuous time Markov chain model
Class Code: B6150 (Communication system theory); B6150M
(Protocols); B0240J; C7410F (Communications computing); C1140J; C5640
(Protocols)
------------------------------------------------------------------------
Database: INSPEC
Ownership: Check the catalogs in your library. Libraries that
Own Item: 94
Copyright: Copyright 1999, IEE
Accession No: 6394886 INSPEC Abstract No: B1999-12-6135C-072
C1999-12-5260D-053
Author(s): Le Leannec, F.; Toutain, F.; Guillemot, C.
Corp Source: INRIA, IRISA, Rennes, France
Title: Packet loss resilient MPEG-4 compliant video coding
for the Internet
Source: Signal Processing: Image Communication, vol.15,
no.1-2 p. 35-56
ISSN: 0923-5965
CODEN: SPICEF
Publication: Netherlands
Language: English
Publisher: Elsevier
Year: Sept. 1999
Copyright No: 0923-5965/99/$20.00
Treatment: P Practical; T Theoretical or Mathematical
Record Type: Journal Paper
Abstract: Targeting multimedia communications over the
Internet, this paper describes a set of complementary techniques in the
direction of both improved packet loss resiliency of video-compressed
streams and efficient usage of available network resources. Aiming
first at a best trade-off between compression efficiency and packer
loss resiliency, a procedure for adapting the video coding modes to
varying network characteristics is introduced. The coding mode
selection is based on a rate-distortion procedure with global
distortion metrics incorporating channel characteristics under the form
of a two-state Markov model. This procedure has been incorporated in an
MPEG-4 video encoder. It has been observed that, in error-free
environments, the channel adaptive mode selection technique does not
bring any penalty in terms of compression while allowing a significant
gain with respect to simple conditional replenishment. On the other
hand, under the same loss conditions, it is shown that this procedure
significantly improves the encoder's performance with respect to the
original MPEG-4 encoder, to approach the robustness of conditional
replenishment mechanisms. This intrinsic robustification of the encoder
allows minimization of the effects of packet losses on the visual
quality of the received video; however, it does not avoid losses. A
rate-based flow control mechanism is then developed and introduced into
the encoder, in order to match the bandwidth requirements of the source
to the bandwidth available over the path of the connection, for both
social' and individual' benefits. The control mechanism developed
combines an RTT-based control loop allowing early reaction to
congestion and a TCP-friendly rate prediction model getting into play
under lossy conditions. This hybrid control mechanism allows full rate
control and smooth rate variations together with high responsiveness.
The introduction of the rate control in the MPEG-4 compliant encoder
allows a stable PSNR and visual quality while decreasing significantly
the source throughput (36 Refs.)
Descriptor(s): Internet; Markov processes; multimedia
communication; packet switching; prediction theory; rate distortion
theory; telecommunication congestion control; video coding
Identifier: packet loss resilient coding; MPEG-4 compliant
video coding; Internet; multimedia communications; complementary
techniques; video-compressed streams; compression efficiency; video
coding modes; varying network characteristics; rate-distortion
procedure; global distortion metrics; channel characteristics;
two-state Markov model; conditional replenishment mechanisms; visual
quality; received video; rate-based flow control mechanism; RTT-based
control loop; TCP-friendly rate prediction model; hybrid control
mechanism; rate control; MPEG-4 compliant encoder; source throughput
Class Code: B6135C; B6210L (Computer communications); B6210R
(Multimedia communications); B0240J; C5260D; C5620W (Other computer
networks); C1140J
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 1999, IEE
Accession No: 6345946 INSPEC Abstract No: B1999-10-6150C-022
Author(s): Xiaowei Yang
Corp Source: Lab. for Comput. Sci., MIT, Cambridge, MA, USA
Title: A model for window based flow control in
packet-switched networks
Source: IEEE INFOCOM '99. Conference on Computer
Communications. Proceedings. Eighteenth Annual Joint Conference of the
IEEE Computer and Communications Societies. The Future is Now (Cat.
No.99CH36320) p. 3 vol. xxv+1583, 423-30 vol.2
Publication: USA
ISBN: 0780354176
Language: English
Publisher: IEEE; Piscataway, NJ, USA
Sponsor: IEEE Comput. Soc.; IEEE Commun. Soc
Conference: Proceedings of INFOCOM'99: Conference on Computer
Communications
Conference Loc: New York, NY, USA; 21-25 March 1999
Year: 1999
Copyright No: 0 7803 5417 6/99/$10.00
Treatment: T Theoretical or Mathematical
Record Type: Conference Paper
Abstract: Networks have increased rapidly both in scale and
speed. Problems related to the control and management are of increasing
interest. The average throughput and end-to-end delay of a network flow
are important design factors. However, there is no satisfactory tool to
obtain such parameters. The traditional packet-by-packet event driven
simulation is slow when the network speed is high. The time driven
simulation faces the difficulty of choosing the right time interval
when simulating packet-switched networks. As the Transmission Control
Protocol (TCP) is the most widely used transport layer protocol, and it
uses a window based flow control mechanism, classic queuing theories
involving Markov chain assumptions are not applicable. This paper
describes a model for window based flow control packet-switched
networks. The model attempts to provide a way to obtain the steady
state results for large and high speed networks using TCP. We discuss
in detail the construction, implementation and application of the
model. This paper also compares the results obtained from the model
with those from the packet-by-packet event driven simulation. The
comparison shows the model is correctly modeling the networks (9 Refs.)
Descriptor(s): packet switching; telecommunication congestion
control; telecommunication network management; telecommunication
network routing; transport protocols
Identifier: packet-switched networks; window based flow
control; network management; average throughput; end-to-end delay;
network flow; network speed; Transmission Control Protocol; transport
layer protocol; TCP; steady state results; packet-by-packet event
driven simulation; random early detection routers
Class Code: B6150C (Communication switching); B6210C (Network
management); B6150M (Protocols); B6150P (Communication network design
and planning)
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 1999, IEE
Accession No: 6300358 INSPEC Abstract No: B1999-08-6150M-084
Author(s): Joshi, A.S.; Umesh, M.N.; Kumar, A.; Mukhopadhyay,
T.; Natesh, K.; Sen, S.; Arunachalam, A.
Corp Source: Performance Modeling Group, Silicon Autom. Syst.
(India) Pvt. Ltd., Bangalore, India
Title: Performance evaluation of TCP over radio link
protocol in TIA/EIA/IS-99 environment
Source: 1999 IEEE International Conference on Personal
Wireless Communications (Cat. No.99TH8366) p. xiv+507, 216-20
Publication: USA
ISBN: 0780349121
Language: English
Publisher: IEEE; Piscataway, NJ, USA
Sponsor: IEEE/AES/COM/LEOS India Council Chapter; Univ.
Victoria; IEEE Region 10; IEEE India Council; IETE (India); IEEE
Commun. Soc.; Dept. Electron.; Gov. India; Delft Univ. Technol.;
Malaviya Regional Eng. Coll
Conference: 1999 IEEE International Conference on Personal
Wireless Communications
Conference Loc: Jaipur, India; 17-19 Feb. 1999
Year: 1999
Copyright No: 0 7803 4912 1/99/$10.00
Treatment: T Theoretical or Mathematical
Record Type: Conference Paper
Abstract: This paper presents the work done to study the
effects of the radio link protocol (RLP) layer on the performance of
TCP over wireless links. In particular we consider RLP as described in
TIA/EIA/IS-99 with variations in some of its parameters. We concentrate
on the sensitivity study of TCP throughput and delay, when the
parameters of RLP such as segmentation depth and maximum number of
retrials varied. Also we consider the effect of maximum TCP advertised
window size on the abovementioned TCP performance parameters. Previous
work in this area have provided some upper bounds for the performance
measures. The results presented are on the basis of detailed simulation
of the RLP retransmission mechanism. Network Simulator has been used as
the simulation platform for our study. We use a two-state Markov model
to capture the mobility characteristics of the channel. Some results
for high-speed links are also presented (6 Refs.)
Descriptor(s): delays; Markov processes; mobile radio; radio
links; telecommunication traffic; transport protocols
Identifier: performance evaluation; TCP; radio link protocol;
TIA/EIA/IS-99; sensitivity study; throughput; delay; maximum advertised
window size; RLP retransmission mechanism; Network Simulator; two-state
Markov model; mobility characteristics
Class Code: B6150M (Protocols); B6250F (Mobile radio systems);
B0240J
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 1999, IEE
Accession No: 6275743 INSPEC Abstract No: B1999-07-6250F-213
Author(s): An Mei Chen; Rao, R.R.
Corp Source: Dept. of Electr. & Comput. Eng., California Univ.,
San Diego, La Jolla, CA, USA
Title: On tractable wireless channel models
Source: Ninth IEEE International Symposium on Personal,
Indoor and Mobile Radio Communications (Cat. No.98TH8361) p. 3 vol.
1574, 825-30 vol.2
Publication: USA
ISBN: 0780348729
Language: English
Publisher: IEEE; New York, NY, USA
Sponsor: Worcester Polytech. Inst.; Nokia; Bell Atlantic
Mobile; CWINS; Analog Devices; IEEE; IEEE Commun. Soc.; IEE; IEICE; GTE
Lab
Conference: Proceedings of Ninth International Symposium on
Personal, Indoor, and Mobile Radio Communications (PIMRC'98)
Conference Loc: Boston, MA, USA; 8-11 Sept. 1998
Year: 1998
Copyright No: 0 7803 4872 9/98/$10.00
Treatment: T Theoretical or Mathematical
Record Type: Conference Paper
Abstract: We present techniques to capture the behavior of
realistic channels with mathematically tractable models, We examine
different state-space aggregation techniques to reduce a large number
of states of Markov chain to a fewer number of states, The property of
strong and weak lumpability is discussed and the stochastic bounding
techniques are examined. These techniques are applied to three
different types of previously published channel models for mobile VHF,
wireless indoor, and Rayleigh fading channels. Results show that our
stochastic bounding technique can produce a simpler yet still accurate
upper bound the for original channel model. We find that the bounds
perform well when the higher-layer error control protocols such as
stop-and-go and TCP are considered (15 Refs.)
Descriptor(s): indoor radio; land mobile radio; Markov processes;
protocols; Rayleigh channels
Identifier: tractable wireless channel models; realistic
channels; state-space aggregation techniques; Markov chain;
lumpability; stochastic bounding techniques; mobile VHF; wireless
indoor; Rayleigh fading channels; higher-layer error control protocols;
stop-and-go; TCP
Class Code: B6250F (Mobile radio systems); B6150M (Protocols);
B0240J
------------------------------------------------------------------------
Database: INSPEC
Ownership: FirstSearch indicates your institution owns the
item. Libraries that Own Item: 301
Copyright: Copyright 1999, IEE
Accession No: 6274798 INSPEC Abstract No: C1999-07-5670-006
Author(s): Ost, A.; Haverkort, B.R.
Corp Source: Lab. for Distributed Syst., Tech. Hochschule
Aachen, Germany
Title: Analysis of windowing mechanisms with
infinite-state stochastic Petri nets
Source: Performance Evaluation Review, vol.26, no.2 p. 38-46
ISSN: 0163-5999
CODEN: PEREDN
Publication: USA
Language: English
Publisher: ACM
Year: Aug. 1998
Treatment: T Theoretical or Mathematical
Record Type: Journal Paper
Abstract: In this paper we present a performance evaluation
of windowing mechanisms in world-wide web applications. Previously,
such mechanisms have been studied by means of measurements only,
however, given suitable tool support, we show that such evaluations can
also be performed conveniently using infinite-state stochastic Petri
nets. We briefly present this class of stochastic Petri nets as well as
the approach for solving the underlying infinite-state Markov chain
using matrix-geometric methods. We then present a model of the TCP
slow-start congestion avoidance mechanism, subject to a (recently
published) typical worldwide web workload. The model is parameterized
using measurement data for a national connection and an overseas
connection. Our study shows how the maximum congestion window size, the
connection release timeout and the packet loss probability influence
the expected number of buffered segments at the server, the connection
setup rate and the connection time (19 Refs.)
Descriptor(s): distributed processing; information resources;
Internet; performance evaluation; Petri nets
Identifier: windowing mechanisms; infinite-state stochastic
Petri nets; world-wide web; TCP; congestion avoidance mechanism
Class Code: C5670 (Network performance); C1160 (Combinatorial
mathematics); C6150N (Distributed systems software); C5620W (Other
computer networks)
------------------------------------------------------------------------
Database: INSPEC
Ownership: Check the catalogs in your library. Libraries that
Own Item: 73
Copyright: Copyright 1999, IEE
Accession No: 6227841 INSPEC Abstract No: B1999-06-6210L-002
C1999-06-5620W-002
Author(s): Ost, A.; Haverkort, B.R.
Corp Source: Lab. for Distributed Syst., Tech. Hochschule
Aachen, Germany
Title: Analysis of windowing mechanisms with
infinite-state stochastic Petri nets
Source: Proc. SPIE - Int. Soc. Opt. Eng. (USA), Proceedings
of the SPIE - The International Society for Optical Engineering,
vol.3530 p. 178-89
ISSN: 0277-786X
CODEN: PSISDG
Publication: USA
Language: English
Publisher: SPIE-Int. Soc. Opt. Eng
Sponsor: SPIE
Conference: Performance and Control of Network Systems II
Conference Loc: Boston, MA, USA; 2-4 Nov. 1998
Year: 1998
Copyright No: 0277-786X/98/$10.00
Treatment: T Theoretical or Mathematical
Record Type: Conference Paper; Journal Paper
Abstract: In this paper we present a performance evaluation
of windowing mechanisms in WWW applications. Previously, such
mechanisms have been studied by means of measurements only, however,
given suitable tool support, we show that such evaluations can also be
performed conveniently using infinite-state stochastic Petri nets. We
briefly present this class of stochastic Petri nets as well as the
approach for solving the underlying infinite-state Markov chain using
matrix-geometric methods. We then present a model of the TCP slow-start
congestion avoidance mechanisms, subject to a typical WWW workload. The
model is parameterized using measurement data for a national connection
and an overseas connection. Our study shows how the maximum congestion
window size, the connection release timeout and the packet loss
probability influence the expected number of buffered segments at the
server, the connection setup rate and the connection time. Furthermore,
the crucial effect of correctly modeling the bursty nature of the
system workload is illustrated by investigating several arrival models
(18 Refs..)
Descriptor(s): information resources; Internet; Markov processes;
Petri nets; telecommunication congestion control; transport protocols
Identifier: windowing mechanisms; infinite-state stochastic
Petri nets; performance evaluation; infinite-state Markov chain;
matrix-geometric methods; TCP slow-start congestion avoidance
mechanisms; national connection; overseas connection; maximum
congestion window size; connection release timeout; packet loss
probability; buffered segments; connection setup rate; connection time;
bursty traffic; system workload; arrival models
Class Code: B6210L (Computer communications); B0250
(Combinatorial mathematics); B0240J; B6150M (Protocols); C5620W (Other
computer networks); C1160 (Combinatorial mathematics); C1140J; C5640
(Protocols)
------------------------------------------------------------------------
Database: INSPEC
Ownership: Check the catalogs in your library. Libraries that
Own Item: 40
Copyright: Copyright 1998, IEE
Accession No: 6043521 INSPEC Abstract No: B9811-6150M-049
C9811-5640-040
Author(s): Kumar, A.; Holtzman, J.
Corp Source: Dept. of Electr. Commun. Eng., Indian Inst. of
Sci., Bangalore, India
Title: Comparative performance analysis of versions of TCP
in a local network with a mobile radio link
Source: Sadhana, vol.23, pt.1 p. 113-29
ISSN: 0256-2499
CODEN: SAPSER
Publication: India
Language: English
Publisher: Indian Acad. Sci
Year: Feb. 1998
Treatment: T Theoretical or Mathematical
Record Type: Journal Paper
Abstract: The scenario is that a bulk data transfer is being
performed over a TCP connection, from a host on a local area network
(LAN) to a mobile host attached to the LAN by a radio link. In an
earlier work we had assumed that packet losses in a TCP connection over
a radio link are statistically independent. In this paper, we extend
this analysis to a Rayleigh fading link, which we model by a two-state
Markov model. The bulk throughputs of TCP-OldTahoe and TCP-Tahoe are
compared with and without fading, for various average signal-to-noise
ratios. We also study the performance with a link protocol on the
wireless link, and study the effect of varying the link packet size,
the number of link packet attempts, and the vehicle speed. For the
parameters of the BSD UNIX implementation, over a 1.5 Mbit/s wireless
link, we find that, with fading, a signal-to-noise ratio of at least 30
dB is required to get reasonable throughput with TCP Tahoe or OldTahoe;
this corresponds to at least 100 times more power than is needed
without fading. For fixed signal-to-noise ratios, as the vehicle speed
varies there are roughly 3 regions of performance: at very low speeds
(pedestrian speeds) the throughput is very good; at low vehicular
speeds the throughput deteriorates, and again becomes very good at
higher vehicle speeds. The speeds corresponding to the various regions
depend on the parameters of the link protocol (20 Refs.)
Descriptor(s): data communication; fading; Markov processes;
mobile radio; performance evaluation; radio links; Rayleigh channels;
transport protocols; wireless LAN
Identifier: performance analysis; mobile radio link; local area
network; LAN; bulk data transfer; mobile host; TCP connection; Rayleigh
fading; two-state Markov model; TCP-OldTahoe; TCP-Tahoe; bulk
throughputs; signal-to-noise ratios; link protocol; packet size;
vehicle speed; BSD UNIX implementation; 1.5 Mbit/s
Class Code: B6150M (Protocols); B6250F (Mobile radio systems);
B6210L (Computer communications); C5640 (Protocols); C5670 (Network
performance); C5620L (Local area networks)
Numerical Ind: bit rate 1.5 E+06 bit/s
------------------------------------------------------------------------
Database: INSPEC
Ownership: FirstSearch indicates your institution owns the
item. Libraries that Own Item: 589
Copyright: Copyright 1998, IEE
Accession No: 6015398 INSPEC Abstract No: B9810-6150M-056
C9810-5640-049
Author(s): Kumar, A.
Corp Source: Wireless Inf. Network Lab., Rutgers Univ.,
Piscataway, NJ, USA
Title: Comparative performance analysis of versions of TCP
in a local network with a lossy link
Source: IEEE/ACM Transactions on Networking, vol.6, no.4 p.
485-98
ISSN: 1063-6692
CODEN: IEANEP
Publication: USA
Language: English
Publisher: IEEE; ACM
Year: Aug. 1998
Copyright No: 1063-6692/98/$10.00
Treatment: T Theoretical or Mathematical
Record Type: Journal Paper
Abstract: We use a stochastic model to study the throughput
performance of various transport control protocol (TCP) versions (Tahoe
(including its older version that we call OldTahoe), Reno, and NewReno)
in the presence of random losses on a wireless link in a local network.
We model the cyclic evolution of TCP, each cycle starting at the epoch
at which recovery starts from the losses in the previous cycle. TCP
throughput is computed as the reward rate in a certain Markov
renewal-reward process. Our model allows us to study the performance
implications of various protocol features, such as fast retransmit and
fast recovery. We show the impact of coarse timeouts. In the local
network environment the key issue is to avoid a coarse timeout after a
loss occurs. We show the effect of reducing the number of duplicate
acknowledgements (ACKs) for triggering a fast retransmit. A large
coarse timeout granularity seriously affects the performance of TCP,
and the various protocol versions differ in their ability to avoid a
coarse timeout when random loss occurs; we quantify these differences.
We show that, for large packet-loss probabilities, TCP-Reno performs no
better, or worse, than TCP-Tahoe. TCP-NewReno is a considerable
improvement over TCP-Tahoe, and reducing the fast-retransmit threshold
from three to one yields a large gain in throughput; this is similar to
one of the modifications in the TCP-Vegas proposal. We explain some of
these observations in terms of the variation of fast-recovery
probabilities with packet-loss probability. The results of our analysis
compare well with a simulation that uses actual TCP code (15 Refs.)
Descriptor(s): land mobile radio; Markov processes; transport
protocols; wireless LAN
Identifier: local network; lossy link; stochastic model;
throughput performance; transport control protocol; Tahoe; OldTahoe;
Reno; NewReno; random losses; wireless link; cyclic evolution;
recovery; reward rate; Markov renewal-reward process; fast retransmit;
fast recovery; coarse timeouts; duplicate acknowledgements; packet-loss
probabilities; TCP-Reno; TCP-Tahoe; TCP-NewReno; TCP-Vegas; TCP-OldTahoe
Class Code: B6150M (Protocols); B6210L (Computer
communications); B6250F (Mobile radio systems); B0240Z (Other topics in
statistics); C5640 (Protocols); C5620L (Local area networks); C5670
(Network performance); C1140Z (Other topics in statistics)
------------------------------------------------------------------------
Database: INSPEC
Ownership: FirstSearch indicates your institution owns the
item. Libraries that Own Item: 548
Copyright: Copyright 1998, IEE
Accession No: 5988401 INSPEC Abstract No: B9809-6150J-007
Author(s): Altman, E.; Basar, T.
Corp Source: Inst. Nat. de Recherche en Inf. et Autom., Sophia
Antipolis, France
Title: Multiuser rate-based flow control
Source: IEEE Transactions on Communications, vol.46, no.7
p. 940-9
ISSN: 0090-6778
CODEN: IECMBT
Publication: USA
Language: English
Publisher: IEEE
Year: July 1998
Copyright No: 0090-6778/98/$10.00
Treatment: T Theoretical or Mathematical
Record Type: Journal Paper
Abstract: Flow and congestion control allow the users of a
telecommunication network to regulate the traffic that they send into
the network in accordance with the quality of service that they
require. Flow control may be performed by the network, as is the case
in asynchronous transfer mode (ATM) networks (the available bit rate
(ABR) transfer capacity), or by the users themselves, as is the case in
the Internet [transmission control protocol/Internet protocol
(TCP/IP)]. We study both situations using optimal control and dynamic
game techniques. The first situation leads to the formulation of a
dynamic team problem, while the second one leads to a dynamic
noncooperative game, for which we establish the existence and
uniqueness of a linear Nash equilibrium and obtain a characterization
of the corresponding equilibrium policies along with the performance
costs. We further show that when the users update their policies in a
greedy manner, not knowing a priori the utilities of the other players,
the sequence of policies thus generated converges to the Nash
equilibrium. Finally, we study an extension of the model that
accommodates multiple traffic types for each user, with the switching
from one type of traffic to another being governed by a Markov jump
process. Presentation of some numerical results complements this study
(30 Refs.)
Descriptor(s): asynchronous transfer mode; convergence of
numerical methods; decentralised control; game theory; Markov
processes; optimal control; queueing theory; telecommunication
congestion control; telecommunication networks; telecommunication
traffic; transport protocols
Identifier: multiuser rate-based flow control; congestion
control; telecommunication network; traffic regulation; quality of
service; asynchronous transfer mode; ATM networks; available bit rate;
ABR transfer capacity; Internet; transmission control protocol/Internet
protocol; TCP/IP; optimal control; dynamic game; dynamic team problem;
dynamic noncooperative game; linear Nash equilibrium; performance
costs; multiple traffic types; Markov jump process; convergence
analysis; bottleneck queue; greedy decentralised algorithms
Class Code: B6150J (Queueing systems); B0240C (Queueing
theory); B0240E (Game theory); B6150M (Protocols); B0240Z (Other topics
in statistics); B6150C (Communication switching)
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 1998, IEE
Accession No: 5932201 INSPEC Abstract No: B9807-6210L-067
C9807-5620L-020
Author(s): Nguyen, G.T.; Katz, R.H.; Noble, B.;
Satyanarayanan, M.
Editor: Charnes, J.M.; Morrice, D.J.; Brunner, D.T.; Swain,
J.J.
Corp Source: Div. of Comput. Sci., California Univ., Berkeley,
CA, USA
Title: A trace-based approach for modeling wireless
channel behavior
Source: 1996 Winter Simulation Conference Proceedings p.
xxxi+1527, 597-604
Publication: USA
ISBN: 0780333837
Language: English
Publisher: SCS Int; San Diego, CA, USA
Sponsor: American Statistical Assoc.; ACM; INFORMS; IEEE;
IIE; NIST; SCS
Conference: Proceedings of 1996 Winter Simulation Conference
Proceedings
Conference Loc: Coronado, CA, USA; 8-11 Dec. 1996
Year: 1996
Treatment: A Application; P Practical
Record Type: Conference Paper
Abstract: The loss behavior of wireless networks has become
the focus of many recent research efforts. Although it is generally
agreed that wireless communications experience higher error rates than
wireline, the nature of these lossy links is not fully understood. This
paper describes an effort to characterize the loss behavior of the AT&T
WaveLAN, a popular in-building wireless interface. Using a trace-based
approach, packet loss information is recorded, analyzed, and validated.
Our results indicate that WaveLAN experiences an average packet error
rate of 2 to 3 percent. Further analysis reveals that these errors are
not independent, making it hard to model them with a simple two-state
Markov chain. We derive another model based on the distributions of the
error and error-free length of the packet streams. For validation, we
modulate both the error models and the traces in a simulator.
Trace-driven simulations yield an average TCP throughput of about 5
percent less than simulations using our best error model (13 Refs.)
Descriptor(s): discrete event simulation; Markov processes;
transport protocols; wireless LAN
Identifier: trace-based approach; wireless channel behavior
modelling; loss behavior; AT&T WaveLAN; packet loss information;
two-state Markov chain; trace-driven simulations; TCP throughput
Class Code: B6210L (Computer communications); B6150M
(Protocols); B0240Z (Other topics in statistics); C5620L (Local area
networks); C5640 (Protocols); C6185 (Simulation techniques); C1140Z
(Other topics in statistics)
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 1997, IEE
Accession No: 5564987 INSPEC Abstract No: B9706-6140C-125
C9706-5260B-074
Author(s): Davis, G.M.; Danskin, J.N.; Xiyong Song
Corp Source: Dartmouth Coll., Hanover, NH, USA
Title: Joint source and channel coding for Internet image
transmission
Source: Proceedings. International Conference on Image
Processing (Cat. No.96CH35919) p. 3 vol. (xlviii+1029+1067+1073), 21-4
vol.1
Publication: USA
ISBN: 0780332598
Language: English
Publisher: IEEE; New York, NY, USA
Sponsor: IEEE Signal Process. Soc
Conference: Proceedings of 3rd IEEE International Conference on
Image Processing
Conference Loc: Lausanne, Switzerland; 16-19 Sept. 1996
Year: 1996
Copyright No: 0 7803 3258 X/96/$5.00
Treatment: T Theoretical or Mathematical
Record Type: Conference Paper
Abstract: Images are usually transmitted across the Internet
using a lossless protocol such as TCP/IP. Lossless protocols require
retransmission of lost packets, which substantially increases
transmission time. We introduce a fast lossy Internet image
transmission scheme (FLIIT) for compressed images which uses forward
error correction to eliminate retransmission delays. FLIIT couples an
embedded quantization scheme with optimized shielding of important
portions of the image. We describe a joint source and channel coding
algorithm for images that minimizes the expected distortion resulting
both from network packet losses and from inexact quantization of
transform coefficients. The algorithm adaptively allocates quantizer
resolution bits and redundancy bits to image subbands to control
quantization errors and expected packet transmission losses. Our packet
loss model incorporates the effects of burst losses using a Markov
packet loss model with empirically derived parameters. Our preliminary
results show a dramatic reduction in the amount of forward error
correction needed to achieve a given expected fidelity as compared to a
fixed-rate channel coding scheme (16 Refs.)
Descriptor(s): adaptive signal processing; channel coding; forward
error correction; image coding; Internet; Markov processes; packet
switching; source coding; transform coding; transport protocols; visual
communication
Identifier: compressed images; image coding; joint
source-channel coding; lossless protocols; TCP/IP; Markov packet loss
model; transmission time; fast lossy Internet image transmission;
forward error correction; retransmission delay; embedded quantization;
optimized image shielding; distortion; network packet losses; transform
coefficient; adaptive algorithm; quantizer resolution bits; quantizer
redundancy bits; quantization error control; packet transmission
losses; burst losses
Class Code: B6140C (Optical information, image and video signal
processing); B6120B (Codes); B6210L (Computer communications); B6150M
(Protocols); B6150C (Communication switching); C5260B (Computer vision
and image processing techniques); C5620W (Other computer networks);
C5640 (Protocols)
------------------------------------------------------------------------
Database: INSPEC
Copyright: Copyright 1995, IEE
Accession No: 5113173 INSPEC Abstract No: B9512-6150M-091
C9512-5640-070
Author(s): Holmstrom, T.; Korner, U.
Editor: Norros, I.; Virtamo, J.
Corp Source: Lund Inst. of Technol., Sweden
Title: Two algorithms for enhancing the performance of
end-to-end transport protocols on top of ATM
Source: Twelfth Nordic Teletraffic Seminar NTS12 (VTT
Symposium 154) p. 494, 325-38
Publication: Finland
Language: English
Publisher: Tech. Res. Centre Finland; Espoo, Finland
Conference: Twelfth Nordic Teletraffic Seminar NTS12 (VTT
Symposium 154)
Conference Loc: Espoo, Finland; 22-24 Aug. 1995
Year: 1995
Treatment: P Practical; T Theoretical or Mathematical
Record Type: Conference Paper
Abstract: Using "classical" end-to-end protocols, like TCP
and TP4, in ATM networks often means a degraded performance. This is
due to the lack of retransmission at link by link level that is a
characteristic of ATM networks. In this paper we report on
investigations of two algorithms that could help to somewhat eliminate
the negative effects. The algorithms were initially designed for a LAN
emulation environment but here they are applied to nodes in an
intermediate ATM-network. The packet discard algorithm, PDA aims at
releasing valuable bandwidth by means of discarding cells belonging to
packets where one cell has already been lost. These cells would
ultimately have been discarded by the receiver. The early packet
discard algorithm, EPDA monitors a buffer level, when the level reaches
a predetermined threshold, the algorithm starts discarding cells
belonging to new packets. It continues to do so until the buffer level
drops below a lower threshold. A simple analytical time continuous
Markov model is presented for the EPDA. Results from discrete time
simulations are also presented. The results show that the EPDA enhances
the performance in some cases but also that the algorithm gives worse
performance in other cases, due to low buffer space utilisation. For
the PDA, results from discrete time simulations are shown. The results
show that the algorithm gives better performance in all cases (4 Refs.)
Descriptor(s): asynchronous transfer mode; buffer storage;
discrete time systems; Markov processes; packet switching; transport
protocols
Identifier: end-to-end transport protocols; link by link level;
ATM networks; intermediate ATM-network; early packet discard algorithm;
buffer level; predetermined threshold; lower threshold; analytical time
continuous Markov model; discrete time simulations; buffer space
utilisation
Class Code: B6150M (Protocols); B6150C (Communication
switching); B0240Z (Other topics in statistics); C5640 (Protocols);
C1140Z (Other topics in statistics)
------------------------------------------------------------------------
FirstSearch(r) Copyright (c) 1992-2001 OCLC as to electronic
presentation and platform. All Rights Reserved.
Reply to: