Spatial and temporal characterisation of ground deformation recorded by geodetic techniques

A critical point in the analysis of ground displacements time series is the development of data driven methods that allow the different sources that generate the observed displacements to be discerned and characterised. A widely used multivariate statistical technique is the Principal Component Analysis (PCA), which allows reducing the dimensionality of the data space maintaining most of the variance of the dataset explained. Anyway, PCA does not perform well in finding the solution to the so-called Blind Source Separation (BSS) problem, i.e. in recovering and separating the original sources that generated the observed data. This is mainly due to the assumptions on which PCA relies: it looks for a new Euclidean space where the projected data are uncorrelated. The Independent Component Analysis (ICA) is a popular technique adopted to approach this problem. However, the independence condition is not easy to impose, and it is often necessary to introduce some approximations. To work around this problem, I use a variational bayesian ICA (vbICA) method, which models the probability density function (pdf) of each source signal using a mix of Gaussian distributions. This technique allows for more flexibility in the description of the pdf of the sources, giving a more reliable estimate of them. Here I present the application of the vbICA technique to GPS position time series. First, I use vbICA on synthetic data that simulate a seismic cycle (interseismic + coseismic + postseismic + seasonal + noise) and a volcanic source, and I study the ability of the algorithm to recover the original (known) sources of deformation. Secondly, I apply vbICA to different tectonically active scenarios, such as the 2009 L'Aquila (central Italy) earthquake, the 2012 Emilia (northern Italy) seismic sequence, and the 2006 Guerrero (Mexico) Slow Slip Event (SSE).

Performance analysis using equivariant kernel density estimator in nonlinear mixture

This paper investigates the performance analysis of separation of mutually independent sources in nonlinear models. The nonlinear mapping constituted by an unsupervised linear mixture is followed by an unknown and invertible nonlinear distortion, are found in many signal processing cases. Generally, blind separation of sources from their nonlinear mixtures is rather difficult. We propose using a kernel density estimator incorporated with equivariant gradient analysis to separate the sources with nonlinear distortion. The kernel density estimator parameters of which are iteratively updated to minimize the output independence expressed as a mutual information criterion. The equivariant gradient algorithm has the form of nonlinear decorrelation to perform the convergence analysis. Experiments are proposed to illustrate these results.

Investigating the performance analysis of EASI algorithm and EKENS algorithm in nonlinear model

This paper investigates the performance of EASI algorithm and the proposed EKENS algorithm for linear and nonlinear mixtures. The proposed EKENS algorithm is based on the modified equivariant algorithm and kernel density estimation. Theory and characteristic of both the algorithms are discussed for blind source separation model. The separation structure of nonlinear mixtures is based on a nonlinear stage followed by a linear stage. Simulations with artificial and natural data demonstrate the feasibility and good performance of the proposed EKENS algorithm.

Exploiting sensitivity of nonorthogonal joint diagonalisation as a security mechanism in steganography

We have recently proposed the framework of independent blind source separation as an advantageous approach to steganography. Amongst the several characteristics noted was a sensitivity to message reconstruction due to small perturbations in the sources. This characteristic is not common in most other approaches to steganography. In this paper we discuss how this sensitivity relates the joint diagonalisation inside the independent component approach, and reliance on exact knowledge of secret information, and how it can be used as an additional and inherent security mechanism against malicious attack to discovery of the hidden messages. The paper therefore provides an enhanced mechanism that can be used for e-document forensic analysis and can be applied to different dimensionality digital data media. In this paper we use a low dimensional example of biomedical time series as might occur in the electronic patient health record, where protection of the private patient information is paramount.

Wireless Network Analytics for Next Generation Spectrum Awareness

The importance of networks, in their broad sense, is rapidly and massively growing in modern-day society thanks to unprecedented communication capabilities offered by technology. In this context, the radio spectrum will be a primary resource to be preserved and not wasted. Therefore, the need for intelligent and automatic systems for in-depth spectrum analysis and monitoring will pave the way for a new set of opportunities and potential challenges. This thesis proposes a novel framework for automatic spectrum patrolling and the extraction of wireless network analytics. It aims to enhance the physical layer security of next generation wireless networks through the extraction and the analysis of dedicated analytical features. The framework consists of a spectrum sensing phase, carried out by a patrol composed of numerous radio-frequency (RF) sensing devices, followed by the extraction of a set of wireless network analytics. The methodology developed is blind, allowing spectrum sensing and analytics extraction of a network whose key features (i.e., number of nodes, physical layer signals, medium access protocol (MAC) and routing protocols) are unknown. Because of the wireless medium, over-the-air signals captured by the sensors are mixed; therefore, blind source separation (BSS) and measurement association are used to estimate the number of sources and separate the traffic patterns. After the separation, we put together a set of methodologies for extracting useful features of the wireless network, i.e., its logical topology, the application-level traffic patterns generated by the nodes, and their position. The whole framework is validated on an ad-hoc wireless network accounting for MAC protocol, packet collisions, nodes mobility, the spatial density of sensors, and channel impairments, such as path-loss, shadowing, and noise. The numerical results obtained by extensive and exhaustive simulations show that the proposed framework is consistent and can achieve the required performance.

Blind separation of auditory event-related brain responses into independent components

Averaged event-related potential (ERP) data recorded from the human scalp reveal electroencephalographic (EEG) activity that is reliably time-locked and phase-locked to experimental events. We report here the application of a method based on information theory that decomposes one or more ERPs recorded at multiple scalp sensors into a sum of components with fixed scalp distributions and sparsely activated, maximally independent time courses. Independent component analysis (ICA) decomposes ERP data into a number of components equal to the number of sensors. The derived components have distinct but not necessarily orthogonal scalp projections. Unlike dipole-fitting methods, the algorithm does not model the locations of their generators in the head. Unlike methods that remove second-order correlations, such as principal component analysis (PCA), ICA also minimizes higher-order dependencies. Applied to detected—and undetected—target ERPs from an auditory vigilance experiment, the algorithm derived ten components that decomposed each of the major response peaks into one or more ICA components with relatively simple scalp distributions. Three of these components were active only when the subject detected the targets, three other components only when the target went undetected, and one in both cases. Three additional components accounted for the steady-state brain response to a 39-Hz background click train. Major features of the decomposition proved robust across sessions and changes in sensor number and placement. This method of ERP analysis can be used to compare responses from multiple stimuli, task conditions, and subject states.

Comparing lock-in ranges and transient responses of second- and third-order phase-locked loops in master-slave clock distribution networks

The distribution of clock signals throughout the nodes of a network is essential for several applications. in control and communication with the phase-locked loop (PLL) being the component for electronic synchronization process. In systems with master-slave (MS) strategies, the PLLs are the slave nodes responsible for providing reliable clocks in all nodes of the network. As PLLs have nonlinear phase detection, double-frequency terms appear and filtering becomes necessary. Imperfections in filtering process cause oscillations around the synchronous state worsening the performance of the clock distribution process. The behavior of one-way master-slave (OWMS) clock distribution networks is studied and performances of first- and second-order filter processes are compared, concerning lock-in ranges and responses to perturbations of the synchronous state. (c) 2007 Elsevier GmbH. All rights reserved.

Truncation errors in finite difference models for solute transport equation with first-order reaction

The truncation errors associated with finite difference solutions of the advection-dispersion equation with first-order reaction are formulated from a Taylor analysis. The error expressions are based on a general form of the corresponding difference equation and a temporally and spatially weighted parametric approach is used for differentiating among the various finite difference schemes. The numerical truncation errors are defined using Peclet and Courant numbers and a new Sink/Source dimensionless number. It is shown that all of the finite difference schemes suffer from truncation errors. Tn particular it is shown that the Crank-Nicolson approximation scheme does not have second order accuracy for this case. The effects of these truncation errors on the solution of an advection-dispersion equation with a first order reaction term are demonstrated by comparison with an analytical solution. The results show that these errors are not negligible and that correcting the finite difference scheme for them results in a more accurate solution. (C) 1999 Elsevier Science B.V. All rights reserved.

Municipal solid waste management system: decision support through systems analysis

Thesis submitted to the Universidade Nova de Lisboa, Faculdade de Ciências e Tecnologia for the degree of Doctor of Philosophy in Environmental Engineering

On separation of a degenerate differential operator in Hilbert space

A coercive estimate for a solution of a degenerate second order di fferential equation is installed, and its applications to spectral problems for the corresponding dif ferential operator is demonstrated. The suffi cient conditions for existence of the solutions of one class of the nonlinear second order diff erential equations on the real axis are obtained.

A Fast Gradient Approximation for Nonlinear Blind Signal Processing

When dealing with nonlinear blind processing algorithms (deconvolution or post-nonlinear source separation), complex mathematical estimations must be done giving as a result very slow algorithms. This is the case, for example, in speech processing, spike signals deconvolution or microarray data analysis. In this paper, we propose a simple method to reduce computational time for the inversion of Wiener systems or the separation of post-nonlinear mixtures, by using a linear approximation in a minimum mutual information algorithm. Simulation results demonstrate that linear spline interpolation is fast and accurate, obtaining very good results (similar to those obtained without approximation) while computational time is dramatically decreased. On the other hand, cubic spline interpolation also obtains similar good results, but due to its intrinsic complexity, the global algorithm is much more slow and hence not useful for our purpose.

A Canonical Genetic Algorithm for Blind Inversion of Linear Channels

It is well known the relationship between source separation and blind deconvolution: If a filtered version of an unknown i.i.d. signal is observed, temporal independence between samples can be used to retrieve the original signal, in the same manner as spatial independence is used for source separation. In this paper we propose the use of a Genetic Algorithm (GA) to blindly invert linear channels. The use of GA is justified in the case of small number of samples, where other gradient-like methods fails because of poor estimation of statistics.

Post-Nonlinear Mixtures and Beyond

Although sources in general nonlinear mixturm arc not separable iising only statistical independence, a special and realistic case of nonlinear mixtnres, the post nonlinear (PNL) mixture is separable choosing a suited separating system. Then, a natural approach is based on the estimation of tho separating Bystem parameters by minimizing an indcpendence criterion, like estimated mwce mutual information. This class of methods requires higher (than 2) order statistics, and cannot separate Gaarsian sources. However, use of [weak) prior, like source temporal correlation or nonstationarity, leads to other source separation Jgw rithms, which are able to separate Gaussian sourra, and can even, for a few of them, works with second-order statistics. Recently, modeling time correlated s011rces by Markov models, we propose vcry efficient algorithms hmed on minimization of the conditional mutual information. Currently, using the prior of temporally correlated sources, we investigate the fesihility of inverting PNL mixtures with non-bijectiw non-liacarities, like quadratic functions. In this paper, we review the main ICA and BSS results for riunlinear mixtures, present PNL models and algorithms, and finish with advanced resutts using temporally correlated snu~sm

Tapping the innovation potential of corporate engagement with stakeholders

Summary This dissertation explores how stakeholder dialogue influences corporate processes, and speculates about the potential of this phenomenon - particularly with actors, like non-governmental organizations (NGOs) and other representatives of civil society, which have received growing attention against a backdrop of increasing globalisation and which have often been cast in an adversarial light by firms - as a source of teaming and a spark for innovation in the firm. The study is set within the context of the introduction of genetically-modified organisms (GMOs) in Europe. Its significance lies in the fact that scientific developments and new technologies are being generated at an unprecedented rate in an era where civil society is becoming more informed, more reflexive, and more active in facilitating or blocking such new developments, which could have the potential to trigger widespread changes in economies, attitudes, and lifestyles, and address global problems like poverty, hunger, climate change, and environmental degradation. In the 1990s, companies using biotechnology to develop and offer novel products began to experience increasing pressure from civil society to disclose information about the risks associated with the use of biotechnology and GMOs, in particular. Although no harmful effects for humans or the environment have been factually demonstrated even to date (2008), this technology remains highly-contested and its introduction in Europe catalysed major companies to invest significant financial and human resources in stakeholder dialogue. A relatively new phenomenon at the time, with little theoretical backing, dialogue was seen to reflect a move towards greater engagement with stakeholders, commonly defined as those "individuals or groups with which. business interacts who have a 'stake', or vested interest in the firm" (Carroll, 1993:22) with whom firms are seen to be inextricably embedded (Andriof & Waddock, 2002). Regarding the organisation of this dissertation, Chapter 1 (Introduction) describes the context of the study, elaborates its significance for academics and business practitioners as an empirical work embedded in a sector at the heart of the debate on corporate social responsibility (CSR). Chapter 2 (Literature Review) traces the roots and evolution of CSR, drawing on Stakeholder Theory, Institutional Theory, Resource Dependence Theory, and Organisational Learning to establish what has already been developed in the literature regarding the stakeholder concept, motivations for engagement with stakeholders, the corporate response to external constituencies, and outcomes for the firm in terms of organisational learning and change. I used this review of the literature to guide my inquiry and to develop the key constructs through which I viewed the empirical data that was gathered. In this respect, concepts related to how the firm views itself (as a victim, follower, leader), how stakeholders are viewed (as a source of pressure and/or threat; as an asset: current and future), corporate responses (in the form of buffering, bridging, boundary redefinition), and types of organisational teaming (single-loop, double-loop, triple-loop) and change (first order, second order, third order) were particularly important in building the key constructs of the conceptual model that emerged from the analysis of the data. Chapter 3 (Methodology) describes the methodology that was used to conduct the study, affirms the appropriateness of the case study method in addressing the research question, and describes the procedures for collecting and analysing the data. Data collection took place in two phases -extending from August 1999 to October 2000, and from May to December 2001, which functioned as `snapshots' in time of the three companies under study. The data was systematically analysed and coded using ATLAS/ti, a qualitative data analysis tool, which enabled me to sort, organise, and reduce the data into a manageable form. Chapter 4 (Data Analysis) contains the three cases that were developed (anonymised as Pioneer, Helvetica, and Viking). Each case is presented in its entirety (constituting a `within case' analysis), followed by a 'cross-case' analysis, backed up by extensive verbatim evidence. Chapter 5 presents the research findings, outlines the study's limitations, describes managerial implications, and offers suggestions for where more research could elaborate the conceptual model developed through this study, as well as suggestions for additional research in areas where managerial implications were outlined. References and Appendices are included at the end. This dissertation results in the construction and description of a conceptual model, grounded in the empirical data and tied to existing literature, which portrays a set of elements and relationships deemed important for understanding the impact of stakeholder engagement for firms in terms of organisational learning and change. This model suggests that corporate perceptions about the nature of stakeholder influence the perceived value of stakeholder contributions. When stakeholders are primarily viewed as a source of pressure or threat, firms tend to adopt a reactive/defensive posture in an effort to manage stakeholders and protect the firm from sources of outside pressure -behaviour consistent with Resource Dependence Theory, which suggests that firms try to get control over extemal threats by focussing on the relevant stakeholders on whom they depend for critical resources, and try to reverse the control potentially exerted by extemal constituencies by trying to influence and manipulate these valuable stakeholders. In situations where stakeholders are viewed as a current strategic asset, firms tend to adopt a proactive/offensive posture in an effort to tap stakeholder contributions and connect the organisation to its environment - behaviour consistent with Institutional Theory, which suggests that firms try to ensure the continuing license to operate by internalising external expectations. In instances where stakeholders are viewed as a source of future value, firms tend to adopt an interactive/innovative posture in an effort to reduce or widen the embedded system and bring stakeholders into systems of innovation and feedback -behaviour consistent with the literature on Organisational Learning, which suggests that firms can learn how to optimize their performance as they develop systems and structures that are more adaptable and responsive to change The conceptual model moreover suggests that the perceived value of stakeholder contribution drives corporate aims for engagement, which can be usefully categorised as dialogue intentions spanning a continuum running from low-level to high-level to very-high level. This study suggests that activities aimed at disarming critical stakeholders (`manipulation') providing guidance and correcting misinformation (`education'), being transparent about corporate activities and policies (`information'), alleviating stakeholder concerns (`placation'), and accessing stakeholder opinion ('consultation') represent low-level dialogue intentions and are experienced by stakeholders as asymmetrical, persuasive, compliance-gaining activities that are not in line with `true' dialogue. This study also finds evidence that activities aimed at redistributing power ('partnership'), involving stakeholders in internal corporate processes (`participation'), and demonstrating corporate responsibility (`stewardship') reflect high-level dialogue intentions. This study additionally finds evidence that building and sustaining high-quality, trusted relationships which can meaningfully influence organisational policies incline a firm towards the type of interactive, proactive processes that underpin the development of sustainable corporate strategies. Dialogue intentions are related to type of corporate response: low-level intentions can lead to buffering strategies; high-level intentions can underpin bridging strategies; very high-level intentions can incline a firm towards boundary redefinition. The nature of corporate response (which encapsulates a firm's posture towards stakeholders, demonstrated by the level of dialogue intention and the firm's strategy for dealing with stakeholders) favours the type of learning and change experienced by the organisation. This study indicates that buffering strategies, where the firm attempts to protect itself against external influences and cant' out its existing strategy, typically lead to single-loop learning, whereby the firm teams how to perform better within its existing paradigm and at most, improves the performance of the established system - an outcome associated with first-order change. Bridging responses, where the firm adapts organisational activities to meet external expectations, typically leads a firm to acquire new behavioural capacities characteristic of double-loop learning, whereby insights and understanding are uncovered that are fundamentally different from existing knowledge and where stakeholders are brought into problem-solving conversations that enable them to influence corporate decision-making to address shortcomings in the system - an outcome associated with second-order change. Boundary redefinition suggests that the firm engages in triple-loop learning, where the firm changes relations with stakeholders in profound ways, considers problems from a whole-system perspective, examining the deep structures that sustain the system, producing innovation to address chronic problems and develop new opportunities - an outcome associated with third-order change. This study supports earlier theoretical and empirical studies {e.g. Weick's (1979, 1985) work on self-enactment; Maitlis & Lawrence's (2007) and Maitlis' (2005) work and Weick et al's (2005) work on sensegiving and sensemaking in organisations; Brickson's (2005, 2007) and Scott & Lane's (2000) work on organisational identity orientation}, which indicate that corporate self-perception is a key underlying factor driving the dynamics of organisational teaming and change. Such theorizing has important implications for managerial practice; namely, that a company which perceives itself as a 'victim' may be highly inclined to view stakeholders as a source of negative influence, and would therefore be potentially unable to benefit from the positive influence of engagement. Such a selfperception can blind the firm from seeing stakeholders in a more positive, contributing light, which suggests that such firms may not be inclined to embrace external sources of innovation and teaming, as they are focussed on protecting the firm against disturbing environmental influences (through buffering), and remain more likely to perform better within an existing paradigm (single-loop teaming). By contrast, a company that perceives itself as a 'leader' may be highly inclined to view stakeholders as a source of positive influence. On the downside, such a firm might have difficulty distinguishing when stakeholder contributions are less pertinent as it is deliberately more open to elements in operating environment (including stakeholders) as potential sources of learning and change, as the firm is oriented towards creating space for fundamental change (through boundary redefinition), opening issues to entirely new ways of thinking and addressing issues from whole-system perspective. A significant implication of this study is that potentially only those companies who see themselves as a leader are ultimately able to tap the innovation potential of stakeholder dialogue.

Analyse par spectrométrie de masse d’hormones stéroïdiennes dans les eaux usées et abattement par oxydation chimique

Thèse réalisée en cotutelle avec Michèle Prévost (Ph.D), Professeure titulaire au département des génies civil, géologique et des mines de l'École Polytechnique de Montréal.