&lt;title&gt;Statistical evaluation and modeling of Internet dial-up traffic&lt;/title&gt;

Faerber, J. E.; Bodamer, Stefan; Charzinski, Joachim

doi:10.1117/12.360361

Cited by 15 publications

(8 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In reality, it is possible that the distributions of the call holding time and inter-call time may be radically different from the exponential distribution (see, e.g. References [19,20]). Comparison between the always-on system and the system with CAC in terms of the probability of a poor-quality call and the probability of blocking.…”

Section: Numerical Studiesmentioning

confidence: 99%

See 1 more Smart Citation

Network dimensioning at the call level for the always‐on network

Nananukul

2004

Int J Communication

View full text Add to dashboard Cite

SUMMARYIt is becoming common for the network to provide always-on access services, where subscribers are guaranteed that their call requests will never be blocked. This paper studies the call-level link dimensioning for the always-on network with single-class traffic. The call-level QoS requirement is expressed in terms of the probability of a poor-quality call, which is the probability that a call experiences packet-level QoS violation at any time during its duration, as opposed to the probability of blocking in the network with call admission control (CAC). The system is modelled as the M/M/infinite system with finite population and an analytic expression for the probability of a poor-quality call is derived based on performability analysis. The effects of the call-level traffic characteristics on the required link resources are studied. It is also shown that the call-level link dimensioning for the always-on network needs more link resources than the network with CAC, and the call-level link dimensioning based on the analytic expression can be used to conservatively dimension the always-on network with arbitrarily distributed call holding time and inter-call time. The paper also studies the problem of estimating the call-level traffic characteristics when the knowledge of call boundaries is not available.

show abstract

Section: Numerical Studiesmentioning

confidence: 99%

Section: Analysis Of the Probability Of A Poor-quality Callmentioning

confidence: 99%

Network dimensioning at the call level for the always‐on network

Nananukul

2004

Int J Communication

View full text Add to dashboard Cite

show abstract

“…In particular, [3] analyzed the holding times and the call interarrival times of Internet access traffic at an ISDN central and [5] studied the influence of access speed on the Internet user behavior.…”

Section: /1mentioning

confidence: 99%

“…Internet traffic characterization has been the subject of several works [1,3,4,5] and is addressed on a permanent basis by organizations such as CAIDA [6]. In particular, [3] analyzed the holding times and the call interarrival times of Internet access traffic at an ISDN central and [5] studied the influence of access speed on the Internet user behavior.…”

Section: /1mentioning

confidence: 99%

Cluster Analysis of Internet Users Based on Hourly Traffic Utilization

Oliveira

Valadas

Pacheco

et al. 2007

IEICE Transactions on Communications

View full text Add to dashboard Cite

Internet access traffic follows hourly patterns that depend on various factors, such as the periods users stay in the access point (e.g. at home or in the office) or their preferences for applications. The clustering of Internet users may provide important information for traffic engineering and tariffing. For example, it can be used to set up service differentiation according to hourly behavior, resource optimization based on multi-hour routing and definition of tariffs that promote Internet access in low busy hours.In this work, we identify patterns of similar behavior by grouping Internet users of two distinct Portuguese ISPs, one using a CATV access network and the other an ADSL one and offering distinct traffic contracts. The grouping of the users is based on their traffic utilization measured every half-hour. Cluster analysis is used to identify the relevant Internet usage profiles, with the partitioning around medoids and Ward's method being the preferred clustering methods. For the two data sets, these clustering methods lead to 3 clusters with similar hourly traffic utilization profiles. The cluster structure is validated through discriminant analysis.Having identified the clusters, the type of applications used as well as the flow duration and transfer rate are analyzed for each cluster resulting in coherent outcomes.

show abstract

“…The main difference between the measured IP traffic at a dial-in modem/ISDN link and UMTS traffic constitutes of the different bandwidth classes of individual users. Färber, Bodamer, and Charzinski used measured trace data to derive a traffic model for Internet dial-in traffic [6]. They focused on the dial-in behavior of modem and ISDN users by fitting general distributions (e.g., Weibull and Lognormal) to the session interarrival-times and the holding times.…”

Section: Introductionmentioning

confidence: 99%

Traffic modeling and characterization for UMTS networks

Klemm

Lindemann²,

Lohmann³

GLOBECOM'01. IEEE Global Telecommunications Conference (Cat. No.01CH37270)

View full text Add to dashboard Cite

In this paper, we present a synthetic traffic model for the Universal Mobile Telecommunication Systems (UMTS) based on measured trace data. The analysis and scaling process of the measured trace data with respect to different bandwidth classes constitutes the basic concept of the UMTS traffic characterization. Furthermore, we introduce an aggregated traffic model for UMTS networks that is analytically tractable.The key idea of this aggregated traffic model lies in customizing the batch Markovian arrival process (BMAP) such that different packet sizes of IP packets are represented by rewards (i.e., batch sizes of arrivals) of the BMAP. The effectiveness of the customized BMAP for modeling UMTS traffic is illustrated using the synthetic traffic model previously presented.

show abstract

<title>Statistical evaluation and modeling of Internet dial-up traffic</title>

Cited by 15 publications

References 5 publications

Network dimensioning at the call level for the always‐on network

Network dimensioning at the call level for the always‐on network

Cluster Analysis of Internet Users Based on Hourly Traffic Utilization

Traffic modeling and characterization for UMTS networks

Contact Info

Product

Resources

About