Jan Potemans

Wavelet packet based multicarrier modulation

A recent but already popular technique for broadband communications is multicarrier modulation, which divides the channel into several orthogonal, overlapping subchannels. Multicarrier modulation is normally implemented using a Fourier transform to create and detect the different subcarriers. This can be efficiently implemented using an (I)FFT. This transform however has the drawback that it uses a rectangular window, which creates rather high sidelobes. This leads to rather high interference when the channel impairments cannot be fully compensated. Looking at other transforms leads us to the wavelet transform, and more specifically the wavelet packet transform, which has more flexibility and due to the time-overlap can have much lower sidelobes. This article studies the implementation of these wavelet packets, and the effects this implementation has on the requirements imposed in the design of usable wavelets. It is shown that the restrictions imposed by the perfect reconstruction requirement necessitate the use of biorthogonal wavelets. This however influences the performance. Moreover, the frequency behaviour of the wavelet packet transform is not straightforward which limits the practical use of this transform in a multicarrier system.

Project Driven Graduate Network Education

In this paper projects for engineering students are described. Three different practical subjects on networking are suggested: a point-to-point link, a VoIP implementation on a LAN and an Office network. These projects are set up to offer Master students in Electrical Engineering (Telecommunications option) of the University of Leuven (K.U.Leuven) a realistic practical background. The authors consider such projects as a necessity, supplementary to the theory. With those projects, the leap to industry after graduation becomes easier to take.

A tunable discrete traffic generator based on a hierarchical scheme of Bernoulli sources

In this paper we present an efficient method for the generation of a discrete series of traffic. The method is based on a superposition of hierarchical Bernoulli sources and generates non-Gaussian distributions. This way self-similar traffic can be generated with an arbitrary combination of Hurst parameter, variance and mean. The method is fully parameterized before generating the trace, which avoids the need for post-processing. Contrary to other methods, the method also allows calculating the trace point by point. The described hierarchical scheme offers a generic framework for fitting the variance-time behavior and can be combined with other stochastic sources.

A Qualitative Description of the Effect of Single Queues on Bin Counts

A large number of research articles are dedicated to queuing theory and queuing systems. Most of these articles employ a continuous representation of network traffic, in the form of timestamps or interarrival times. In this, there is a contradiction with more recent traffic models capable of capturing the multi-fractal nature of network traffic e.g. the Conservative Cascade model. These models often represent packet streams in a discrete way by calculating the bin count vector. Directly describing the effect of queuing systems on the variance-time behavior of this discrete representation of traffic is relatively unexplored terrain. This paper presents and analyzes some qualitative results on the altering of a bin count vector when passing it through a basic queuing system. The scenario is also extended with the influence of correlated background traffic.

Improving RED Performance during Handovers in Wireless IP Networks

The random early detection (RED) is a widely used active queue management (AQM) algorithm for congestion avoidance. It monitors the average queue length to detect incipient congestion and notifies the connections of congestion to adjust their sending rate. During a handover, all the active connections of the mobile node will be diverted to the new network and increase the total traffic there immediately. Because RED uses an exponential weighted moving average (EWMA) to calculate the queue length, it will be too slow to track this rapid change and fail to react correctly. In this paper we propose a dual RED algorithm to accommodate the coming handed-over connections quickly while keeping the link utilization high. The improvement is verified with simulations.

Evaluation of network traffic workload scaling techniques

Performance testing on a network that is loaded by realistic background traffic, is an important issue in the design of network applications. To obtain this background traffic, a packet stream can be captured on a real network. For testing the application under varying network loads, straightforward techniques can be applied to alter the originally captured traffic trace. An example thereof is faster replay of the packet stream. The drawback of these simple techniques is that the packet stream will be altered in many ways, and thus a simple multiplication of the load that is inflicted by the packet stream on a network cannot be guaranteed. This paper presents and analyzes some frequently used load scaling methods. Next to the computationally more demanding use of traffic models, a novel technique is introduced based on the rotation of the bin vector, that combines better accuracy with computational simplicity.

Design considerations of a modular testbed for mobile IP networks

In this paper we present a research testbed being developed at K.U.Leuven, Belgium, aiming at providing a flexible and convenient testing infrastructure for researchers working on mobility issues over IP networks. The testbed is built on general-purpose PCs, which makes it suitable for most laboratories. Its user-level modularity makes it easy for both developers and users to configure and extend even for unusual mobility scenarios and new mobility techniques. Movement is realized in a way that makes mobility experiment easy to be carried out and automated