Mihaylova L.,University of Sheffield |
Carmi A.Y.,Ben - Gurion University of the Negev |
Septier F.,Telecom Lille 1 |
Gning A.,University College London |
And 2 more authors.
Digital Signal Processing: A Review Journal | Year: 2014
This work presents the current state-of-the-art in techniques for tracking a number of objects moving in a coordinated and interacting fashion. Groups are structured objects characterized with particular motion patterns. The group can be comprised of a small number of interacting objects (e.g. pedestrians, sport players, convoy of cars) or of hundreds or thousands of components such as crowds of people. The group object tracking is closely linked with extended object tracking but at the same time has particular features which differentiate it from extended objects. Extended objects, such as in maritime surveillance, are characterized by their kinematic states and their size or volume. Both group and extended objects give rise to a varying number of measurements and require trajectory maintenance. An emphasis is given here to sequential Monte Carlo (SMC) methods and their variants. Methods for small groups and for large groups are presented, including Markov Chain Monte Carlo (MCMC) methods, the random matrices approach and Random Finite Set Statistics methods. Efficient real-time implementations are discussed which are able to deal with the high dimensionality and provide high accuracy. Future trends and avenues are traced. © 2013 Elsevier Inc.
Zhang Z.,Telecom Lille 1 |
Nait-Abdesselam F.,Lille University of Science and Technology
Wireless Networks | Year: 2010
As one of the backup measures of intrusion prevention techniques, intrusion detection plays a paramount role in the second defense line of computer networks. Intrusion detection in wireless mesh networks (WMNs) is especially challenging and requires particular design concerns due to their special infrastructure and communication mode. In this paper, we propose a novel anomaly detection system, termed RADAR, to detect and handle anomalous mesh nodes in wireless mesh networks. Specifically, reputation is introduced to characterize and quantify a node's behavior in terms of fine-grained performance metrics of interest. The dual-core detection engine of RADAR then explores spatio-temporal property of such behavior to manifest the deviation between that of normal and anomalous nodes. Although the current RADAR prototype is only implemented with routing protocols, the design architecture allows it to be easily extended to cross-layer anomaly detection where anomalous events occur at different layers and can be resulted by either intentional intrusion or accidental network failure. The simulation results demonstrate that RADAR can achieve high detection accuracy, low computational complexity, and low false positive rate. © 2010 Springer Science+Business Media, LLC.
Septier F.,Telecom Lille 1 |
Peters G.W.,University College London
IEEE Journal on Selected Topics in Signal Processing | Year: 2016
Nonlinear non-Gaussian state-space models arise in numerous applications in statistics and signal processing. In this context, one of the most successful and popular approximation techniques is the Sequential Monte Carlo (SMC) algorithm, also known as particle filtering. Nevertheless, this method tends to be inefficient when applied to high dimensional problems. In this paper, we focus on another class of sequential inference methods, namely the Sequential Markov Chain Monte Carlo (SMCMC) techniques, which represent a promising alternative to SMC methods. After providing a unifying framework for the class of SMCMC approaches, we propose novel efficient strategies based on the principle of Langevin diffusion and Hamiltonian dynamics in order to cope with the increasing number of high-dimensional applications. Simulation results show that the proposed algorithms achieve significantly better performance compared to existing algorithms. © 2015 IEEE.
Wang S.,Xidian University |
Zhang Z.,Telecom Lille 1 |
Computers and Security | Year: 2013
The increasing complexity of today's computer systems, together with the rapid emergence of novel vulnerabilities, make security hardening a formidable challenge for security administrators. Although a large variety of tools and techniques are available for vulnerability analysis, the majority work at system or network level without explicit association with human and organizational factors. This article presents a middleware approach to bridge the gap between system-level vulnerabilities and organization-level security metrics, ultimately contributing to cost-benefit security hardening. In particular, our approach systematically integrates attack graph, a commonly used effective approach to representing and analyzing network vulnerabilities, and Hidden Markov Model (HMM) together, for exploring the probabilistic relation between system observations and states. More specifically, we modify and apply dependency attack graph to represent network assets and vulnerabilities (observations), which are then fed to HMM for estimating attack states, whereas their transitions are driven by a set of predefined cost factors associated with potential attacks and countermeasures. A heuristic searching algorithm is employed to automatically infer the optimal security hardening through cost-benefit analysis. We use a synthetic network scenario to illustrate our approach and evaluate its performance through a set of simulations. © 2012 Elsevier Ltd. All rights reserved.
Maalej A.,Lille University of Science and Technology |
Maalej A.,Telecom Lille 1 |
Amor B.B.,Lille University of Science and Technology |
Amor B.B.,Telecom Lille 1 |
And 4 more authors.
Pattern Recognition | Year: 2011
In this paper we address the problem of 3D facial expression recognition. We propose a local geometric shape analysis of facial surfaces coupled with machine learning techniques for expression classification. A computation of the length of the geodesic path between corresponding patches, using a Riemannian framework, in a shape space provides a quantitative information about their similarities. These measures are then used as inputs to several classification methods. The experimental results demonstrate the effectiveness of the proposed approach. Using multiboosting and support vector machines (SVM) classifiers, we achieved 98.81% and 97.75% recognition average rates, respectively, for recognition of the six prototypical facial expressions on BU-3DFE database. A comparative study using the same experimental setting shows that the suggested approach outperforms previous work. © 2011 Elsevier Ltd. All rights reserved.