J.M. Pitts

Chaotic maps for traffic modelling and queueing performance analysis

Abstract In this paper we present an overview of the progress made using chaotic maps to model individual and aggregated self-similar traffic streams and in particular their impact on queue performance. Our findings show that the asymptotic behaviour of the queue is a function only of the tail of the ON active periods, and that the Hurst parameter is not a good parameter to achieve traffic control: two different self-similar traffic traces can have the same Hurst parameter but have a very different effect on the queue statistics. These results are part of a framework for developing chaotic control of networks.

Applying emergent self-organizing behavior for the coordination of 4G networks using complexity metrics

Future fourth-generation (4G) wireless networks are expected to have an ad hoc, dynamic structure with cheap, ubiquitous, low-powered nodes that are autoconfigurable and flexible. Controlling such a network means coping with uncertainty, not only of traffic demand, but also in the network structure. Because of this, a new approach to the control and coordination of 4G networks will be needed, one that replaces centralized with highly decentralized control. One promising approach is to view networks as self-organizing systems comprising simple interacting nodes that rely on emergent behavior to provide network-wide coordination. However, such networks are often difficult to predict or manipulate, due to their distributed nature. This paper describes the use of an entropy-based complexity metric to investigate and manipulate the behavior of such self-organizing systems in mobile networks. We introduce a self-organizing algorithm for cell dimensioning, and apply the complexity metric to extract information on network-wide behavior. We then introduce a framework for using the metric to manipulate emergent self-organizing behavior in 4G networks.

A Hybrid Technique for Accelerated Simulation of ATM Networks and Network Elements

Conventional simulation of cell- or packet-switched networks involves the use of discrete event simulators that model each individual cell through the network, typically called cell-level simulation. Each cells arrival at, or departure from, a network element is represented by an event. However, statistical considerations are such that very large numbers of cells have to be simulated to guarantee the accuracy of the results. This has always caused very long simulation times, often amounting to many hours of real time just to simulate a few minutes of simulated time. In this article we describe a novel methodology for accelerating simulation studies in cell-based communication networks, e.g., ATM, by using a hybrid analytical/simulation combination. The methodology uses a mathematical technique to seperate foreground traffic from background traffic, and focuses on accelerating cell by cell simulation.

Monitoring real-time applications events in net-centric enterprise systems to ensure high quality of experience

Faced with intense competition, network service providers, supported by their respective Network Operations Centers (NOCs), must ensure the best possible Quality of Service (QoS) and corresponding Quality of Experience (QoE) for end-users or face the loss of business. QoE represents the perception of quality experienced by end-users of a real-time system, such as VoIP. The QoE challenge becomes significant when real-time applications running on net-centric enterprise systems, driven by methodologies such as Service Oriented Architecture (SOA), encounter issues that require time-sensitive problem determination and resolution. Optimizing end-user QoE becomes daunting as system complexity grows to span multiple applications and functions across different administrative areas of responsibility. Available tools fall short in their support of enterprise QoE monitoring because they examine only portions of the available data within their respective purview resulting in a fractionalized picture of the true state of the enterprise system. This paper addresses the above limitation by presenting an approach for aggregating observed real-time applications data across the enterprise, thereby permitting timely and effective interpretation and response to real-time application events. This approach uses a proven framework, reference architecture, and QoE metrics categories to produce simulation results for the collection, correlation and analysis of application information on net-centric enterprise systems. Results are presented for several VoIP scenarios including a Denial of Service (DoS) event that causes noticeable applications delay. These results represent measurements for over one billion packets derived from a large- scale simulation running on the UK national supercomputing service, HECToR. The paper concludes by highlighting key benefits to the service providers business from using this approach.

Overflow probability in shared cell switched buffers

The ubiquity of the cell switching technique is a clear indication of its continued importance for the future of telecommunications worldwide. The performance of shared buffers in cell switches provides substantial improvement over switches with separate output buffers. The conventional approach to shared buffer analysis is numerical convolution; however, previous work provides an analytical solution based on the Chernoff bound. In this paper we develop a new closed-form expression for the cell loss probability in the shared buffer that offers superior accuracy, enhanced simplicity, and better generality.

Scalable voice over internet protocol service-level agreement guarantees in converged transmission control protocol/ internet protocol networks

Increasing complexity of the recent introduced quality of service (QoS) provisioning tools in packet-switched networks becomes a major obstacle for Internet service providers to practically deploy and manage these mechanisms to guarantee a range of transparent real-time service-level agreements. The authors revisit the existing Internet protocol (IP) QoS approaches and present a cost-effective solution to provide voice over Internet protocol (VoIP) service guarantees in converged IP networks by combining open shortest-path first traffic engineering with new insights for the configuration of standard active queue management random early detection mechanism. Extensive simulation studies are carried out, and the service quality is explicitly quantified using international telecommunication union E-model for a range of scenarios representing different types of network uncertainties. The direct representation of VoIP service quality well demonstrate that such a simple, coordinated approach, in keeping with the Internet paradigm, can achieve increased load for a given quality level and greater resilience under degraded network conditions.

Analysing the transition between unsaturated and saturated operating conditions in 802.11 network scenarios

The aim of this paper is an investigation of 802.11 scenarios, and in particular to evaluate the nature of the transition between unsaturated and saturated operating conditions. Most existing analyses do not adequately account for the extra idle state, and typically assume saturated behaviour even for unsaturated conditions. This paper introduces an extension to the Bianchi Markov chain that deals with the accurate evaluation of contention behaviour across the load range. Results compare favourably with simulation, and indicate the presence of hysteresis in both throughput and the access time.

Guaranteeing Enterprise VoIP QoS with Novel Approach to DiffServ AF Configuration

To satisfy the low delay, low jitter performance requirements of real-time traffic such as VoIP, DiffServ EF class using priority scheduling is normally recommended. This requires limits on admissible load, configured to meet the most stringent QoS in the real-time traffic mix. To handle multiple realtime service types with heterogeneous QoS, we propose a novel alternative: DiffServ AF classes with RED queue management. Recent queue theoretic advances have demonstrated REDs ability to control the delay distribution of inelastic traffic under congested conditions. This significantly reduces late delivery of packets at the cost of (probabilistically) dropping a greater proportion in the network. We investigated the load-quality trade-off in a DiffServ domain with both hop-based and link weight optimized OSPF routing. End-to-end delay, jitter and loss were measured for two AF classes carrying VoIP across all source-destination paths in order to compare the effects of tail- drop and RED queue management. We used the ITU-T E-model to express how these performance measures affect voice quality; results demonstrate that our novel AF configuration enables the network to carry more traffic for a given quality level, and to degrade more gracefully under severe congestion.

Analysis, simulation and measurement in large-scale packet networks

The authors review analysis, simulation and measurement techniques, the three fundamental methods for performance evaluation in packet networks, looking at whats known, whats new and some outstanding issues. In trying to avoid re-reviewing material which has already been well summarised elsewhere, the authors concentrate on areas that are relatively new or possibly less generally well appreciated. So, under analysis, the focus is on models for network topologies and connectivity, and on wireless access. In the simulation section the focus is on techniques for scalable simulation for large-scale packet networks. Compared to the other two areas measurement is relatively new anyway, and more time is spent on motivation, techniques and some recently discovered limitations.

Economic evaluation of a mature ATM network

The economic evaluation of a mature ATM network is presented. The authors discuss the issues involved and present two examples. Four customer categories are taken into account, namely residential, small business, medium business, and large business customers. The aim is to introduce a simple meaningful model, which incorporates the most important factors that influence the economic viability of an asynchronous transfer mode (ATM) network, and to present the limits and limitations of the area of economic evaluation. The revenue, cost, and net income (of the network operator) are calculated for two different scenarios, that is with and without residential customers. >