No description available.

No description available.

No description available.

No description available.

No description available.

Steganography and covert-communications represent a great and real threat today more than ever due to the evolution of modern communications. This doctoral work proposes defenses against such covert-communication techniques in two threatening but underdeveloped domains. Indeed, this work focuses on the novel problem of visual sensor network steganalysis but also proposes one of the first solutions against video steganography. The first part of the dissertation looks at covert-communications in videos. The contribution of this study resides in the combination of image processing using motion vector interpolation and non-traditional detection theory to obtain better results in identifying the presence of embedded messages in videos compared to what existing still-image steganalytic solutions would offer. The proposed algorithm called MoViSteg utilizes the specifics of video, as a whole and not as a series of images, to decide on the occurrence of steganography. Contrary to other solutions, MoViSteg is a video-specific algorithm, and not a repetitive still-image steganalysis, and allows for detection of embedding in partially corrupted sequences. This dissertation also lays the foundation for the novel study of visual sensor network steganalysis. We develop three different steganalytic solutions to the problem of covert-communications in visual sensor networks. Because of the inadequacy of the existing steganalytic solutions present in the current research literature, we introduce the novel concept of preventative steganalysis, which aims at discouraging potential steganographic attacks. We propose a set of solutions with active and passive warden scenarii using the material made available by the network. To quantify the efficiency of the preventative steganalysis, a new measure for evaluating the risk of steganography is proposed: the embedding potential which relies on the uncertainty of the image’s pixel values prone to corruption.

Wireless Sensor Networks (WSN) is an important technology that can be used to provide new data sets for many applications ranging from healthcare monitoring to military surveillance. Due to the increasing popularity of WSNs, user demands have evolved as well. To achieve the end-to-end goals and requirements of the applications, managing the resources of the network becomes a critical task. Cognitive networking techniques for resource management have been proposed in recent years to provide performance gains over traditional design methodologies. However, even though several tools have been considered in cognitive network design, they show limitations in their adaptability, complexity, and their ability to consider multiple conflicting goals. Thus, this thesis proposes novel cognitive solutions for WSNs that include a reasoning machine and a learning protocol. Weighted Cognitive Maps (WCM) and Q-Learning are identified as suitable tools for addressing the aforementioned challenges and designing the cognitive solutions due to their ability to consider conflicting objectives with low complexity. WCM is a mathematical tool that has powerful inference capabilities. Thus, they are used to design a reasoning machine for WSNs. Two case studies are proposed in this thesis that illustrate the capabilities of WCMs and their flexibility in supporting different application requirements and network types. In addition, an elaborate theoretical model based on Markov Chains (MC) is proposed to analyze the operation of the WCM system. Extensive computer simulations and analytical results show the ability of the WCM system to achieve the end-to-end goals of the network and find compromises between conflicting constraints. On the other hand, Q-Learning is a well known reinforcement learning algorithm that is used to evaluate the actions taken by an agent over time. Thus, it is used to design a learning protocol that improves the performance of the WCM system. Furthermore, to ensure that the learning protocol operates efficiently, methods for improving the learning speed and achieving distributed learning across multiple nodes are proposed as well. Extensive computer simulations show that the learning protocol improves the performance of the WCM system in several metrics. / Thesis (Ph.D, Electrical & Computer Engineering) -- Queen's University, 2013-02-04 16:24:55.385

No description available.

The subject of this thesis is the development of parametric methods for the calibration of array shape errors. Two physical scenarios are considered, the online calibration (self-calibration) using far-field sources and the offline calibration using near-field sources. The maximum likelihood (ML) estimators are employed to estimate the errors. However, the well-known computational complexity in objective function optimization for the ML estimators demands effective and efficient optimization algorithms. A novel space-alternating generalized expectation-maximization (SAGE)-based algorithm is developed to optimize the objective function of the conditional maximum likelihood (CML) estimator for the far-field online calibration. Through data augmentation, joint direction of arrival (DOA) estimation and array calibration can be carried out by a computationally simple search procedure. Numerical experiments show that the proposed method outperforms the existing method for closely located signal sources and is robust to large shape errors. In addition, the accuracy of the proposed procedure attains the Cram´er-Rao bound (CRB). A global optimization algorithm, particle swarm optimization (PSO) is employed to optimize the objective function of the unconditional maximum likelihood (UML) estimator for the farfield online calibration and the near-field offline calibration. A new technique, decaying diagonal loading (DDL) is proposed to enhance the performance of PSO at high signal-to-noise ratio (SNR) by dynamically lowering it, based on the counter-intuitive observation that the global optimum of the UML objective function is more prominent at lower SNR. Numerical simulations demonstrate that the UML estimator optimized by PSO with DDL is optimally accurate, robust to large shape errors, and free of the initialization problem. In addition, the DDL technique is applicable to a wide range of array processing problems where the UML estimator is employed and can be coupled with different global optimization algorithms.

This Thesis describes research aimed towards the realisation of an embedded ultrasonic monitoring system for operation in a smart structure environment. Alien fibres embedded within carbon fibre reinforced composite plates could be utilised to guide ultrasound to strategic release points for the interrogation of the test sample, with minimal structural degradation. In particular, an array of fibre waveguides could be utilised to deliver periodic stresses to a plate-type structure to excite Lamb wave propagation. With such a system several problems need to be addressed. Firstly, identification of an appropriate mode of propagation to transport the acoustic energy along the waveguide and meet the requirements of the system is required. Theoretical analysis of elastic wave propagation in cylinders is provided to establish the characteristics of the various supported modes. Subsequently, the longitudinal (axisymmetric) modes are highlighted as being the most appropriate for the proposed system. Ideally, such a mode should be singly generated to provide increased control on the coupling of ultrasound from the waveguide to the desired Lamb wave mode. Therefore, the transducer-fibre interface utilised for efficient single mode generation in the fibre waveguides is essential. This Thesis investigates a novel, cheap, simple, and robust coupling strategy, which employs a conical polymer bond to behave as a mechanical transformer efficiently coupling the lateral displacements of a piezoceramic transducer into longitudinal displacements in the waveguide. Theoretical and experimental analysis of this technique are described and design guidelines detailed. The Thesis then goes on to provide experimental and theoretical evidence of the fibre waveguide's ability to transport acoustic energy to and from plate-type structures for subsequent Lamb wave generation and detection. Initially, a surface mount strategy is adopted, whereby the ends of several thin cylindrical waveguides are bonded to the surface of aluminium plates to behave as a line array. Generation and detection of the fundamental symmetric (So) and asymmetric (A0) Lamb wave modes is demonstrated. Following this, the Thesis details considerations concerning the realisation of an embedded fibre waveguide monitoring system. Here, the choice of waveguide material and geometry is crucial to ensure the guided ultrasound does not leak into the structure at undesirable locations and that the structural integrity of the test sample is maintainable. Consequently, various waveguide configurations are investigated to establish an arrangement capable of meeting the system requirements. An embedded fibre waveguide Lamb wave system is then presented in a hard-setting polymer plate confirming the basic design methodology. Finally, So Lamb wave generation and detection in a carbon fibre reinforced composite plate is demonstrated utilising embedded acoustic waveguides sensors.