The Second International Conference on Internet Monitoring and Protection

ICIMP 2007

July 1-6, 2007 - Silicon Valley, USA


Tutorials

T1. Is the IMS Service Platform a Solution for Next Generation Network Providers to Be More Than Bit Carriers?
by Antonio Cuevas

T2. Call-level Multi-Rate Teletraffic Loss Models
by Michael D. Logothetis and Ioannis D. Moscholios

T3. Assessing and Hacking Network Security
by Radu State

T4. Architectures On-Demand for Any Domain Using Stable Software Patterns
by Mohamed Fayad

 

Details:

T1. Is the IMS Service Platform a Solution for Next Generation Network Providers to Be More Than Bit Carriers?

Abstract:

Telecommunications are undergoing a major revolution. On the one side there is a migration to a universal IP network, supporting any access technology, fixed or wireless and any service. On the other side, there is a convergence of the different business models and a coexistence of all of them.

IMS is a key part in this revolution. It is a service platform enabling thesemi-walled garden business model cheered by network operators. It targets"traditional" telecom. operators' businesses, like voice calls. It is based on open IETF standards and uses IPv6. As such, it should be accessible byany mean, fixed, mobile, wired, wireless. Because of, this IMS seems to be the service platform for next generation networks.  IMS has attracted stronginterest both by the scientific and industry communities. Still few IMSskilled professionals exist.

This tutorial will:

  • Present IMS architecture and building blocks
  • Present IMS interfaces and protocols, "internal" and external.
  • Present IMS philosophy; bias towards business models
  • Be biased to Next Generation Networks and how IMS suits them

This tutorial will not: Enter into details at "development level"

T2. Call-level Multi-Rate Teletraffic Loss Models

Call-level multi-rate teletraffic loss models aim at assessing the call-level QoS of IP based networks with resource reservation capabilities but also for the emerging and future all-optical core networks, such as MPλS/GMPLS, or 3G wireless networks (e.g. UMTS) networks.  This assessment is important for the bandwidth allocation among service-classes (guaranteeing QoS), the avoidance of too costly over-dimensioning of the network and the prevention, through traffic engineering mechanisms, of excessive throughput degradation. Despite of its importance, the call-level performance modelling and QoS assessment is a challenge in the highly heterogeneous environment of modern telecom networks, due to the presence of elastic traffic, or complicated call arrival process. The key call-level performance index is the Call Blocking Probabilities (CBP), while the efficient CBP calculation is a sine qua non of a teletraffic model, in order to cope with the high bandwidth capacities of network links. We present efficient call-level teletraffic loss models leading to recurrent CBP calculations in multirate systems by taking into account that calls (and consequently the models) can be distinguished as follows.

  • According to the call arrival process, into:
    • Random calls – Random traffic (infinite number of traffic sources).
    • Quasi-random calls – Quasi-random traffic (finite number of traffic sources)
    • Batch Poisson arrivals (infinite number of traffic sources). Calls from different service-classes arriving in batches, while batches arriving randomly.
  • According to bandwidth requirements upon arrival, into:
    • Calls with fixed bandwidth requirements.
    • Calls with several, alternative, contingency (“elastic”) bandwidth requirements.
  • According to their behavior while in service, into:
    • Calls with constant use of their assigned bandwidth (constant-bit-rate) during their lifetime.
    • Calls that, although constantly use the assigned bandwidth, can tolerate bandwidth compression/expansion while in service.
    • Calls that alternate between transmission periods (ON) and idle periods (OFF).

This tutorial will enable engineers and telecom/computer network managers, even those who are not experts in teletraffic to learn and put the latest teletraffic models into practice. Researchers (postgraduate students) will learn how teletraffic models are built and they will be motivated for creating models satisfying new specifications, or for applying the existing teletraffic models to new networking technologies. No special prerequisite knowledge of audience is expected rather than basic teletraffic theory, that is, notion of traffic load and properties, call origination processes, holding (service) time distribution and elementary analysis of Markovian loss systems.

T3. Assessing and Hacking Network Security

(to be announced)

T4. Architectures On-Demand for Any Domain Using Stable Software Patterns

(to be announced)

 
 

Copyright (c) 2006, IARIA