Catálogo de publicaciones - libros

This paper contains the keynote address given at the Symposium by the Chairmain of the COST Action 285. It outlines The studies undertaken by the members of the Action with the objective to enhance existing modeling and simulation tools and to develop new ones for research in emerging multiservice telecommunication networks. The paper shows how the scope of COST Action 285 has been enriched by the contributions made at the symposium.

Low rate high latency data services will co-exist with high rate low latency real-time multimedia applications in the next generation networks. Increasing volume of multimedia flows in an environment with heterogeneity in bandwidth, propagation medium and statistical characteristics of traffic can be expected to generate time-varying demands on the quality of service (QoS) and network resources. In such non-stationary environment, dynamic resource reservation schemes operating in harmony with the variability in demand patterns may provide efficient mechanisms for resource utilization and guarantee QoS compliance. In this work we identify the challenges that need to be addressed in designing a three level core, distribution and edge (CDE) hierarchical network using time-varying resource allocation mechanisms. Learning, prediction and correction (LPC) architecture based upon integration of operational CDE network with online simulation proposed as a design alternative to contemporary networks.

We describe empirical approaches for multilayer traffic modeling — i.e., models that span several protocol layers — and for modeling multimedia traffic at various time scales. Multilayer traffic modeling is challenging, as one must deal with disparate traffic sources; control loops; the effects of network elements; cross-layer protocols; asymmetries in bandwidth, session lengths, and application behaviors; and an enormous number of potential confounding effects among the various factors. We summarize experiments that combine an analytical transport layer model (layer 4) with layer 1/2/3 components to investigate whether analytical multilayer traffic models might provide credible outcomes in (near) real time. Preliminary results suggest that such models can provide reasonable, steady-state, first-order approximations of behaviors that span several protocol layers. Multimedia traffic modeling is also challenging, as many types of multimedia traffic have characteristic statistical signatures induced by their encoders. Traffic analysts have proposed a number of feasible models for multimedia traffic, but it is not clear which is best. We summarize experiments using multiplicative SARIMA(s,p,d,q) models of MPEG-4 multimedia traces at various time scales. Preliminary results suggest that the seasonal effect induced by MPEG’s ‘group of pictures’ encoding is the dominant factor at time scales up to a few tens of seconds, while scene length predominates at longer time scales.

System Performance strongly depends on the incoming workload. To improve the user perceived perfomance, it is mandatory to analyze this observed workload with the aim to choose the most adequate system configuration and processing methods of the incoming requests. In this work, we will examine, whether some characterisctis, like heavy taildeness, temporal locality, and frequency of references are restricted to certain document types, or are present for all of them. We will see, that this analysis is helpful for further studies, especially for developing new approaches to improve the system performance (e.g. for caches)

The predictability of network traffic is a significant interest in many domains such as congestion control, admission control, and network management. An accurate traffic prediction model should have the ability to capture prominent traffic characteristics, such as long-range dependence (LRD) and self-similarity in the large time scale, multifractal in small time scale. In this paper we propose a new network traffic prediction model based on non-linear time series ARIMA/GARCH. This model combines linear time series ARIMA model with non-linear GARCH model. We provide a parameters estimation procedure for our proposed ARIMA/GARCH model. We then evaluate a scheme for our models’ prediction. We show that our model can capture prominent traffic characteristics, not only in large time scale but also in small time scale. Compare with existing FARIMA model, our model have better prediction accuracy.

Fluid models of IP networks have been recently proposed, to break the scalability barrier of traditional performance evaluation approaches, both simulative (e.g., ns-2) and analytical (e.g., queues and Markov chains). Fluid models adopt a deterministic description of the average source and network dynamics through a set of (coupled) ordinary differential equations that are solved numerically, obtaining estimates of the time-dependent behavior of the IP network. The most attractive property of fluid models resides in the fact that they are scalable, i.e., their complexity is independent of the number of TCP flows and of link capacities. In this paper we provide a theoretical investigation of the origins of the scalability of fluid models. We show that the set of differential equations defining the network dynamics under both drop-tail and AQM buffering exhibits a nice invariance property, that allows an equivalence relation to be established among different systems. The validity of the invariance property is verified in realistic network scenarios with ns-2 simulations.

In this paper we describe an approach for optimal dimensioning of multi-service lines. The solution presented is based on the method to convert recursions for global state probabilities of multi-service models into stable form. The suggested approach is numerically stable because it deals with normalised values of global state probabilities used for estimation of main stationary performance measures.

Current Network Management (NM) procedures are concerned primarily with monitoring aspects. The architectures are centralized and based on static managed objects. They do not provide real-time control capabilities. Within this framework, effective network management depends on the coordination of monitoring operations across various Element Management Systems. The coordination is handled for the most part, by engineers/operators through manual procedures at designated Network Operations Center (NOC). The Subject Matter Experts at the NOC also use intuition and heuristic data to occasionally fine-tune the network parameters in order to maintain the advertised service-level objectives and control the traffic flow. Today’s industry procedures, such as SNMP and TNM frameworks, presenta number of limitations for the complex and heterogeneous emerging telecommunication networks. The next generation networks are expected to inter-connect different access network technologies and architectures in a multi-vendor environment. The access technologies such as optical Ethernet, dark fiber, wireless LAN and fixed wireless access systems will provide better and more resilient alternatives for multimedia traffic than existing DSL and Cable access networks. The core networks serving metropolitan regions will similarly migrate to increasingly heterogeneous technologies and architectures across different software platforms .More efficient technologies, such as Resilient Packet Rings, possibly Terabit and Gigabit Routers, OXC switching nodes and multiple service protocol platforms will provide the framework for transporting heterogeneous multimedia traffic over Dense Wavelength Division Multiplexing core networks. A more efficient framework to facilitate network management and control in next generation networks is proposed. This framework employs a distributed architecture of autonomous and heterogeneous sensor entities. Thearchitecture facilitates peer-to-peer networking under the supervision of a novel Integrated Network Management System (INMS). This approach provides automated decision making, rapid deployment of network management “service” functions; real-time monitoring of fault, configuration, accounting, performance and security management functions; real-time provisioning; and real-time NM control activation and removal. The novel framework provides an integrated view of end-to-end managed network entities

Industry research tends to be product oriented. Simulations are expected to mimic the behavior of a specific product, a product that is often in the design phase. The product’s features, in turn, depend on ever changing market forecasts. At the same time, reliability is a must, since business decisions are based on simulation results. We look at the example of tool design for UMTS system level simulations to highlight the challenges of simulation tool development in the telecommunication industry and to discuss solution strategies.

Researchers performing simulations in the field of computer telecommunications are often faced with the time-consuming task of converting huge quantities of data to and from different formats. We examine some of the requirements of the telecommunications simulation community and propose an architecture for a general purpose archiver and converter for big quantities of simulation data to be released as free software.