Two single-pass methods for fitting multiexponentials to experimental data are described. These methods rely on the construction of a matrix whose characteristic polynomial is used to determine the rates of decay. In the first method, which we call the multiple-delay method, the matrix is constructed using time delays of the experimental data. This method is fast and highly accurate even if the experimental signal contains exponential components with similar rates of decay. In the second method, which we call the successive-integral method, the matrix is constructed using integrals of the experimental data. This procedure yields good results for noisy signals and is a generalization of the method of Martin et al. ((1993) J. Neurosci. Methods, 51: 135-146). In addition, a particular instability of the multiexponential curve fitting problem is identified and a method for overcoming this instability is given.

In this thesis we focus mainly on special finite differences and finite volume methods and apply them to the pricing of barrier options.The structure of this work is the following: in Chapter 1 we introduce the definitions of options and illustrate some properties of vanilla European options and exotic options.Chapter 2 describes a classical model used in the financial world, the  Black-Scholes model. We derive theBlack-Scholes formula and show how stochastic differential equations model financial instruments prices.The aim of this chapter is also to present the initial boundary value problem and the maximum principle.We discuss boundary conditions such as: the first boundary value problem, also called  Dirichlet problem that occur in pricing ofbarrier options and European options. Some kinds of put options lead to the study of a second boundary value problem (Neumann, Robin problem),while the Cauchy problem is associated with one-factor European and American options.Chapter 3 is about finite differences methods such as theta, explicit, implicit and Crank-Nicolson method, which are used forsolving partial differential equations.The exponentially fitted scheme is presented in Chapter 4. It is one of the new classesof a robust difference scheme that is stable, has good convergence and does not produce spurious oscillations.The stability is also advantage of the box method that is presented in Chapter 5.In the beginning of the Chapter 6 we illustrate barrier options and then we consider a novel finite volume discretization for apricing the above options.Chapter 7 describes discretization of the Black-Scholes equation by the fitted finite volume scheme. In  Chapter 8 we present and describe numerical results obtained by using  the finite difference methods illustrated in the previous chapters.

En acoustique sous marine, les ´etudes sur les zones en eau peu profondes sontredevenues strat´egiques. Cette th`ese porte sur l’ ´etude de la s´eparation et la d´etectionde trajet dans le cadre des eaux peu profondes tomographie acoustique oc´eanique. Dansune premi´ere ´etape de notre travail, nous avons donn´e un bref aperc¸u sur les techniquesexistantes de traitement acoustique sous-marine afin de trouver la difficult´e toujoursconfront´es `a ce type de m´ethodes. Par cons´equent, nous avons fait une conclusion qu’ilest encore n´e cessaire d’am´eliorer la r´esolution de s´eparation afin de fournir des informationsplus utiles pour l’ ´etape inverse de la tomographie acoustique oc´eanique.Ainsi, une enquˆete sur les mthodes haute r´esolution est effecut´ee. Enfin, nous avonspropos´e une m´ethode `a haute r´esolution appel´ee lissage MUSICAL (MUSIC Active largeband), qui combine le lissage de fr´equence spatiale avec l’algorithme MUSICAL, pourune s´eparation efficace de trajet coh´erentes ou totalement corr´el´es. Cependant, cettem´ethode est bas´ee sur la connaissance a priori du nombre de trajet. Ainsi, nous introduisonsun test (exponential fitting test) (EFT) `a l’aide de courte longueur des ´echantillonspour d´eterminer le nombre de trajets. Ces deux m´ethodes sont appliqu´ees `a la fois desdonn´ees synth´etiques et les donn´ees r´eelles acquises dans un r´eservoir `a petite ´echelle.Leurs performances sont compar´ees avec les m´ethodes conventionnelles pertinentes. / As the studies on shallow-water acoustics became an active field again, this dissertationfocuses on studying the separation and detection of raypaths in the context of shallowwaterocean acoustic tomography. As a first step of our work, we have given a briefreview on the existing array processing techniques in underwater acoustics so as to findthe difficulties still faced by these methods. Consequently, we made a conclusion thatit is still necessary to improve the separation resolution in order to provide more usefulinformation for the inverse step of ocean acoustic tomography. Thus, a survey on highresolutionmethod is provided to discover the technique which can be extended to separatethe raypaths in our application background. Finally, we proposed a high-resolutionmethod called smoothing-MUSICAL (MUSIC Actif Large band), which combines thespatial-frequency smoothing with MUSICAL algorithm, for efficient separation of coherentor fully correlated raypaths. However, this method is based on the prior knowledgeof the number of raypaths. Thus, we introduce an exponential fitting test (EFT)using short-length samples to determine the number of raypaths. These two methodsare both applied to synthetic data and real data acquired in a tank at small scale. Theirperformances are compared with the relevant conventional methods respectively.

En acoustique sous marine, les 'etudes sur les zones en eau peu profondes sontredevenues strat'egiques. Cette th'ese porte sur l' 'etude de la s'eparation et la d'etectionde trajet dans le cadre des eaux peu profondes tomographie acoustique oc'eanique. Dansune premi'ere 'etape de notre travail, nous avons donn'e un bref aperc¸u sur les techniquesexistantes de traitement acoustique sous-marine afin de trouver la difficult'e toujoursconfront'es 'a ce type de m'ethodes. Par cons'equent, nous avons fait une conclusion qu'ilest encore n'e cessaire d'am'eliorer la r'esolution de s'eparation afin de fournir des informationsplus utiles pour l' 'etape inverse de la tomographie acoustique oc'eanique.Ainsi, une enquˆete sur les mthodes haute r'esolution est effecut'ee. Enfin, nous avonspropos'e une m'ethode 'a haute r'esolution appel'ee lissage MUSICAL (MUSIC Active largeband), qui combine le lissage de fr'equence spatiale avec l'algorithme MUSICAL, pourune s'eparation efficace de trajet coh'erentes ou totalement corr'el'es. Cependant, cettem'ethode est bas'ee sur la connaissance a priori du nombre de trajet. Ainsi, nous introduisonsun test (exponential fitting test) (EFT) 'a l'aide de courte longueur des 'echantillonspour d'eterminer le nombre de trajets. Ces deux m'ethodes sont appliqu'ees 'a la fois desdonn'ees synth'etiques et les donn'ees r'eelles acquises dans un r'eservoir 'a petite 'echelle.Leurs performances sont compar'ees avec les m'ethodes conventionnelles pertinentes.

For noiseless sampled data, we describe the close connections between Prony--like methods, namely the classical Prony method, the matrix pencil method and the ESPRIT method. Further we present a new efficient algorithm of matrix pencil factorization based on QR decomposition of a rectangular Hankel matrix. The algorithms of parameter estimation are also applied to sparse Fourier approximation and nonlinear approximation.

For noiseless sampled data, we describe the close connections between Prony--like methods, namely the classical Prony method, the matrix pencil method and the ESPRIT method. Further we present a new efficient algorithm of matrix pencil factorization based on QR decomposition of a rectangular Hankel matrix. The algorithms of parameter estimation are also applied to sparse Fourier approximation and nonlinear approximation.