THE task of modeling data traffic in networks is as old as the first commercial telephony systems. In the recent past in mobile telephone networks the focus has moved from voice to packetswitched services. The new cellular mobile networks of the third generation (UMTS) and the evolved second generation (GPRS) offer the subscriber the possibility of staying online everywhere and at any time. The design and dimensioning is well known for circuit switched voice systems, but not for mobile packet-switched systems. The terms user expectation, grade of service and so on need to be defined. To find these parameters it is important to have an accurate traffic model that delivers good traffic estimates. In this thesis we carried out measurements in a live 3G core network of an Austrian operator, in order to find appropriate models that can serve as ...

Digital visual communication has been increasingly adopted as an efficient new medium in a variety of different fields; multi-media computers, digital televisions, telecommunications, etc. Exchange of visual information between remote sites requires that digital video is encoded by compressing the amount of data and transmitting it through specified network connections. The compression and transmission of digital video is an amalgamation of statistical data coding processes, which aims at efficient exchange of visual information without technical barriers due to different standards, services, media, etc. It is associated with a series of different disciplines of digital signal processing, each of which can be applied independently. It includes a few different technical principles; distortion, rate theory, prediction techniques and control theory. The MPEG (Moving Picture Experts Group) video compression standard is based on this paradigm, thus, it contains a variety of different coding ...

Automatic face recognition has attracted a lot of attention not only because of the large number of practical applications where human identification is needed but also due to the technical challenges involved in this problem: large variability in facial appearance, non-linearity of face manifolds and high dimensionality are some the most critical handicaps. In order to deal with the above mentioned challenges, there are two possible strategies: the first is to construct a “good” feature space in which the manifolds become simpler (more linear and more convex). This scheme usually comprises two levels of processing: (1) normalize images geometrically and photometrically and (2) extract features that are stable with respect to these variations (such as those based on Gabor filters). The second strategy is to use classification structures that are able to deal with non-linearities and to generalize properly. To ...

Distributed source coding is a technique that allows to compress several correlated sources, without any cooperation between the encoders, and without rate loss provided that the decoding is joint. Motivated by this principle, distributed video coding has emerged, exploiting the correlation between the consecutive video frames, tremendously simplifying the encoder, and leaving the task of exploiting the correlation to the decoder. The first part of our contributions in this thesis presents the asymmetric coding of binary sources that are not uniform. We analyze the coding of non-uniform Bernoulli sources, and that of hidden Markov sources. For both sources, we first show that exploiting the distribution at the decoder clearly increases the decoding capabilities of a given channel code. For the binary symmetric channel modeling the correlation between the sources, we propose a tool to estimate its parameter, thanks to an ...

Everyday sounds in real-world environments are a powerful source of information by which humans can interact with their environments. Humans can infer what is happening around them by listening to everyday sounds. At the same time, it is a challenging task for a computer algorithm in a smart device to automatically recognise, understand, and interpret everyday sounds. Sound event detection (SED) is the process of transcribing an audio recording into sound event tags with onset and offset time values. This involves classification and segmentation of sound events in the given audio recording. SED has numerous applications in everyday life which include security and surveillance, automation, healthcare monitoring, multimedia information retrieval, and assisted living technologies. SED is to everyday sounds what automatic speech recognition (ASR) is to speech and automatic music transcription (AMT) is to music. The fundamental questions in designing ...

Probabilistic graphical models are a powerful concept for modeling high-dimensional distributions. Besides modeling distributions, probabilistic graphical models also provide an elegant framework for performing statistical inference; because of the high-dimensional nature, however, one must often use approximate methods for this purpose. Belief propagation performs approximate inference, is efficient, and looks back on a long success-story. Yet, in most cases, belief propagation lacks any performance and convergence guarantees. Many realistic problems are presented by graphical models with loops, however, in which case belief propagation is neither guaranteed to provide accurate estimates nor that it converges at all. This thesis investigates how the model parameters influence the performance of belief propagation. We are particularly interested in their influence on (i) the number of fixed points, (ii) the convergence properties, and (iii) the approximation quality. For this purpose, we take a different perspective ...

We consider Bayesian networks (BNs) with discriminatively optimized parameters and structures, i.e. BNs that are optimized to maximize a kind of probabilistic margin. These maximum margin Bayesian networks (MM BNs) are inspired by support vector machines (SVMs) that aim to separate samples from different classes by a large margin in some feature space. MM BNs achieve classification performance on par with BNs optimized according to other discriminative criteria, e.g. maximum conditional likelihood. Furthermore, in several applications, they achieve classification performance comparable to that of both, linear and kernelized, SVMs. In the literature, two definitions of MM BNs with respect to their parameters are available. We analyze these definitions in terms of asymptotic consistency, extend these definitions by a generative regularizer and analyze properties of MM BNs with respect to reduced-precision implementations. We start by analyzing the asymptotic consistency of MM ...

The fabrication of miniature electronics and sensors has encouraged the creation of a wide range of wireless enabled devices designed to be worn on the human body. This has led to the prominence of so-called wearable communications, which have emerged to satisfy the demand for wireless connectivity between these devices and with external networks. The work in this thesis has focused on the characterization of the composite fading (i.e combined multipath and shadowing) observed in wearable communications channels. It has also investigated the mitigation of the deleterious effects of both of these propagation phenomena in wearable communications. In order to accurately characterize the behaviour of the composite fading signal observed in wearable communications channels, new fading models such as F, $\kappa$-$\mu$ / inverse gamma and $\eta$-$\mu$ / inverse gamma composite fading models, have been proposed. The generality and utility of ...

In a communication system it results undoubtedly of great interest to compress the information generated by the data sources to its most elementary representation, so that the amount of power necessary for reliable communications can be reduced. It is often the case that the redundancy shown by a wide variety of information sources can be modelled by taking into account the probabilistic dependance among consecutive source symbols rather than the probabilistic distribution of a single symbol. These sources are commonly referred to as single or multiterminal sources "with memory" being the memory, in this latter case, the existing temporal correlation among the consecutive symbol vectors generated by the multiterminal source. It is well known that, when the source has memory, the average amount of information per source symbol is given by the entropy rate, which is lower than its entropy ...

are involved until the development of Sequential Monte Carlo techniques which are also known as the particle filters. In particle filtering, the problem is expressed in terms of state-space equations where the linearity and Gaussianity requirements of the Kalman filtering are generalized. Therefore, we need information about the functional form of the state variations. In this thesis, we bring a general solution for the cases where these variations are unknown and the process distributions cannot be expressed by any closed form probability density function. Here, we propose a novel modeling scheme which is as unified as possible to cover all these problems. Therefore we study the performance analysis of our unifying particle filtering methodology on non-stationary Alpha Stable process modeling. It is well known that the probability density functions of these processes cannot be expressed in closed form, except for ...

The thesis focuses on statistical approaches to model, mitigate, and prevent distributed network attacks. When dealing with distributed network attacks (and, more in general, with cyber-security problems), three fundamental phases/issues emerge distinctly. The first issue concerns the threat propagation across the network, which entails an "avalanche" effect, with the number of infected nodes increasing exponentially as time elapses. The second issue regards the design of proper mitigation strategies (e.g., threat detection, attacker's identification) aimed at containing the propagation phenomenon. Finally (and this is the third issue), it is also desirable to act on the system infrastructure to grant a conservative design by adding some controlled degree of redundancy, in order to face those cases where the attacker has not been yet defeated. The contributions of the present thesis address the aforementioned relevant issues, namely, propagation, mitigation and prevention of distributed ...

During the last decades, information is being gathered and processed at an explosive rate. This fact gives rise to a very important issue, that is, how to effectively and precisely describe the information content of a given source signal or an ensemble of source signals, such that it can be stored, processed or transmitted by taking into consideration the limitations and capabilities of the several digital devices. One of the fundamental principles of signal processing for decades is the Nyquist-Shannon sampling theorem, which states that the minimum number of samples needed to reconstruct a signal without error is dictated by its bandwidth. However, there are many cases in our everyday life in which sampling at the Nyquist rate results in too many data and thus, demanding an increased processing power, as well as storage requirements. A mathematical theory that emerged ...

Wyner-Ziv (WZ) video coding is a particular case of distributed video coding, a new video coding paradigm based on the Slepian-Wolf and Wyner-Ziv theorems which mainly exploit the source correlation at the decoder and not only at the encoder as in predictive video coding. Therefore, this new coding paradigm may provide a flexible allocation of complexity between the encoder and the decoder and in-built channel error robustness, interesting features for emerging applications such as low-power video surveillance and visual sensor networks among others. Although some progress has been made in the last eight years, the rate-distortion performance of WZ video coding is still far from the maximum performance attained with predictive video coding. The WZ video coding compression efficiency depends critically on the capability to model the correlation noise between the original information at the encoder and its estimation generated ...

This thesis deals with estimation of states and parameters in nonlinear and non-Gaussian dynamic systems. Sequential Monte Carlo methods are mainly used to this end. These methods rely on models of the underlying system, motivating some developments of the model concept. One of the main reasons for the interest in nonlinear estimation is that problems of this kind arise naturally in many important applications. Several applications of nonlinear estimation are studied. The models most commonly used for estimation are based on stochastic difference equations, referred to as state-space models. This thesis is mainly concerned with models of this kind. However, there will be a brief digression from this, in the treatment of the mathematically more intricate differential-algebraic equations. Here, the purpose is to write these equations in a form suitable for statistical signal processing. The nonlinear state estimation problem is ...

Cellular Neural Networks are characterized by simplicity of operation. The network consists of a large number of nonlinear processing units; called cells; that are equally spread in the space. Each cell has a simple function (sequence of multiply-add followed by a single discrimination) that takes an element of a topographic map and then interacts with all cells within a specified sphere of interest through direct connections. Due to their intrinsic parallel computing power, CNNs have attracted the attention of a wide variety of scientists in, e.g., the fields of image and video processing, robotics and higher brain functions. Simplicity of operation together with the local connectivity gives CNNs first-hand advantages for tiled VLSI implementations with very high speed and complexity. The first VLSI implementation has been based on analogue technology but was small and suffered from parasitic capacitances and resistances ...