Publications Server of the Weierstrass Institute for Applied Analysis and Stochastics
Not a member yet
7600 research outputs found
Sort by
Analysis of a drift-diffusion model for perovskite solar cells
No full textThis paper deals with the analysis of an instationary drift-diffusion model for perovskite solar cells including Fermi--Dirac statistics for electrons and holes and Blakemore statistics for the mobile ionic vacancies in the perovskite layer. The free energy functional is related to this choice of the statistical relations. Exemplary simulations varying the mobility of the ionic vacancy demonstrate the necessity to include the migration of ionic vacancies in the model frame. To prove the existence of weak solutions, first a problem with regularized state equations and reaction terms on any arbitrarily chosen finite time interval is considered. Its solvability follows from a time discretization argument and passage to the time-continuous limit. Applying Moser iteration techniques, a priori estimates for densities, chemical potentials and the electrostatic potential of its solutions are derived that are independent of the regularization level, which in turn ensure the existence of solutions to the original problem
Linearization of finite-strain poro-visco-elasticity with degenerate mobility
No full textA quasistatic nonlinear model for finite-strain poro-visco-elasticity is considered in the Lagrangian frame using Kelvin--Voigt rheology. The model consists of a mechanical equation which is coupled to a diffusion equation with a degenerate mobility. Having shown existence of weak solutions in a previous work, the focus is first on showing boundedness of the concentration using Moser iteration. Afterwards, it is assumed that the external loading is small, and it is rigorously shown that solutions of the nonlinear, finite-strain system converge to solutions of the linear, small-strain system
Existence of similarity profiles for diffusion equations and systems
No full textWe study the existence of self-similar profiles for diffusion equations and reaction-diffusion systems on the real line, where different nontrivial limits are imposed at both sides of infinity. The theses profiles solve a coupled system of nonlinear ODEs that can be treated by monotone operator theory
Advances in Continuum Physics: In Memoriam Wolfgang Dreyer
No full textThis book is in honor of the late Professor Wolfgang Dreyer from the Weierstrass Institute for Applied Analysis and Stochastics (WIAS) in Berlin
Dimensionality reduction in multicepstral features for voice spoofing detection: Case studies with singular value decomposition, genetic algorithm, and auto-encoder: International Conference on Artificial Intelligence and Soft Computing
No full textRecognizing people by their voice is not only interesting and challenging on its own, but also a very necessary ability in a variety of contexts, e.g. for unlocking electronic devices, for authentication in banking transactions and much more. However, current automatic voice verification technologies are vulnerable to presentation attacks due to the increasing realm of audio falsifications ? also known as spoofing ? that rely on artificial intelligence technologies, for example. Due to the direct connection of speech recognition with security systems and privacy concerns, the development of measures against spoofing is not only crucial but also urgently needed. In this work, we propose an experimental approach that focuses on dimensionality reduction techniques together with a classification model to detect spoofing in voice biometric systems. Our method uses a multicepstral feature extraction framework to distinguish between real and synthetic speech signals. To validate the proposed method, tests were performed using the ASVSpoof 2017 v2.0 database. Dimensionality reduction techniques such as singular value decomposition, genetic algorithms and auto-encoder were applied before implementing the support vector machine model. The model is then evaluated using the Equal Error Rate metric. For comparison, we investigated the voice liveness detection problem using machine learning models with and without dimensionality reduction strategies, observing an improvement of up to 7.11% in the model error metric
Transient Simulation of k⋅p-Schrödinger Systems Using Discrete Transparent Boundary Conditions
No full textThis chapter deals with the derivation and analysis of discrete transparent boundary conditions (TBCs) for transient systems of Schrödinger-type equations in one space dimension. These systems occur i.e. in the physics of layered semiconductor devices as the so called k⋅p-Schrödinger equations, which are a well established tool for band structure calculations.The new TBCs are constructed directly for the chosen finite difference scheme, in order to ensure the stability of the underlying scheme and to completely avoid any numerical reflections. The discrete TBCs are constructed using the solution of the exterior problem with Laplace and Z-transformation, respectively.These discrete TBCs can easily be obtained by an inverse Z-transformation based on FFT, but these exact discrete TBCs are non-local in time and thus very costly. Hence, as a remedy, we present approximate discrete TBCs, that allow a fast calculation of the boundary terms using a sum-of-exponentials approach. This is a prev
Exploring multicepstral features in a new classical machine learning-based framework for replay attack detection
No full textThe integration of Internet of Things (IoT) technologies has accelerated the adoption of recognition and authentication systems, offering seamless access across devices from smart homes to workplace systems. Among biometric traits, voice stands out due to its simplicity, cleanliness, low capture cost, uniqueness, and the extensive computational resources supporting it in the scientific literature. Recently, however, spoofing risks have emerged as a serious challenge to the security of voice-based systems. To counteract these threats without additional hardware, techniques analyzing inherent voice signal features have been developed. This paper introduces a new soft computing framework based on classical machine learning classifiers such as Support Vector Machine (SVM), Random Forest (RF), and Logistic Regression (LR), comprising Gaussian-noise-based data augmentation, extraction and fusion of multiple cepstral and non-cepstral features, and dimensionality reduction through Singular Value Decomposition (SVD). In particular, we explore eight distinct cepstral extraction techniques, exemplified by popular approaches such as MFCC and CQCC, and sixteen additional non-cepstral metrics such as Zero Crossing Rate (ZCR) and Harmonic-to-Noise Ratio (HNR). Additionally, we generalize cepstral pattern representation by proposing “cepstral multiprojection”, a novel strategy designed to systematically reduce the dimensionality and redundancy of multicepstral matrices, thereby enhancing discriminative power and computational efficiency. Evaluated with the ASVSpoof 2017 v2.0 competition benchmark, our approach achieved competitive results, reaching 5.14% equal error rate (EER) on the “Dev” set and 10.58% on the “Eval” set, presenting an effective, interpretable, and computationally efficient alternative to state-of-the-art methods for replay attack detection in voice authentication systems. These findings provide a reproducible, reconfigurable, and modular soft computing framework that is interpretable, hardware-independent, and suitable for real-world deployment in voice spoofing detection systems
Improving voice spoofing detection through extensive analysis of multicepstral feature reduction
No full textVoice biometric systems play a critical role in numerous security applications, including electronic device authentication, banking transaction verification, and confidential communications. Despite their widespread utility, these systems are increasingly targeted by sophisticated spoofing attacks that leverage advanced artificial intelligence techniques to generate realistic synthetic speech. Addressing the vulnerabilities inherent to voice-based authentication systems has thus become both urgent and essential. This study proposes a novel experimental analysis that extensively explores various dimensionality reduction strategies in conjunction with supervised machine learning models to effectively identify spoofed voice signals. Our framework involves extracting multicepstral features followed by the application of diverse dimensionality reduction methods, such as Principal Component Analysis (PCA), Truncated Singular Value Decomposition (SVD), statistical feature selection (ANOVA F-value, Mutual Information), Recursive Feature Elimination (RFE), regularization-based LASSO selection, Random Forest feature importance, and Permutation Importance techniques. Empirical evaluation using the ASVSpoof 2017 v2.0 dataset measures the classification performance with the Equal Error Rate (EER) metric, achieving values of approximately 10%. Our comparative analysis demonstrates significant performance gains when dimensionality reduction methods are applied, underscoring their value in enhancing the security and effectiveness of voice biometric verification systems against emerging spoofing threats
Bifurcations and intermittency in coupled dissipative kicked rotors
Full text linkWe investigate the emergence of complex dynamics in a system of coupled dissipative kicked rotors and show that critical transitions can be understood via bifurcations of simple states. We study multistability and bifurcations in the single rotor model, demonstrating how these give rise to a variety of coexisting spatial patterns in a coupled system. A combined order parameter is introduced to characterize different spatial patterns and to reveal the coexistence of chaotic and regular attractors. Finally, we illustrate an intermittent phenomenon near the onset of chaos