Search

The control of communication networks critically relies on procedures capable of detecting unanticipated load changes. In this chapter we present an overview of such techniques, in a setting in which each connection consumes roughly the same amount of bandwidth (with VoIP as a leading example). For the situation of exponential holding times an explicit analysis can be performed in a large-deviations regime, leading to approximations of the test statistic of interest (and, in addition, to results for the transient of the M/M/∞ queue, which are of independent interest). This procedure being applicable to exponential holding times only, and also being numerically rather involved, we then develop an approximate procedure for general holding times. In this procedure we record the number of trunks occupied at equidistant points in time Δ,2Δ,…, where Δ is chosen sufficiently large to safely assume that the samples are independent; this procedure is backed by results on the transient of the M/G/∞ queue. The validity of the testing procedures is demonstrated through set of numerical experiments; it is also pointed out how diurnal patterns can be dealt with. An experiment with real data illustrates the proposed techniques.

In this Chapter we give an overview of statistical methods for anomaly detection (AD), thereby targeting an audience of practitioners with general knowledge of statistics. We focus on the applicability of the methods by stating and comparing the conditions in which they can be applied and by discussing the parameters that need to be set. © 2013 Springer-Verlag Berlin Heidelberg.

In this special issue,weintend to present recent advances in the growing research field of television (TV). While a few years ago the death of TV had been announced, the convergence of fixed and mobile networks and the ever-growing means of watching, sharing, and commenting TV content has instead created new vibrant research areas for smart, social, and converged TV. The papers included in this special section report on advances in this field with a combination of academic and industry research.

The introduction of third generation (3G) public wireless network infrastructures, such as the Universal Mobile Telecommunications System (UMTS), enable the development of innovative mobile services. For example, deploying m-health services which embed tele-monitoring and tele-treatment services become feasible with the role-out of 3G networks. These services allow healthcare professionals to monitor a mobile patient's vital signs and provide feedback to this patient anywhere and any time. The performance of m-health services perceived by end-users depends on the serviceableness of 3G networks to support these services. Hence, the performance of 3G networks is a critical factor for successful development of m-health services. In this paper, we present a methodology for measurements-based performance assessment of 3G networks that aim to support m-health services. This methodology has been applied to evaluate end-user perceived service performance, in relation to the performance (i.e. serviceableness) of a 3G network. In addition, we analyse the measurements with the purpose to improve the end-to-end delay characteristics of the tele-monitoring service as well as optimize the (derived) goodput behaviour of this 3G network. Our results show that the goodput behaviour is asymmetric and depends on a bearer assignment policy of the network. Based on our results we provide guidelines for the design of application protocols for m-health services and how these protocols deal with the changing performance behaviour of 3G networks.