TY - CONF AB - A combination of GPS (global positioning system) and INS (inertial navigation system) is known to provide high precision and highly robust vehicle localization. Notably during times when the GPS signal has a poor quality, e.g. due to the lack of a sufficiently large number of visible satellites, the INS, which may consist of a gyroscope and an odometer, will lead to improved positioning accuracy. In this paper we show how velocity information obtained from GSM (global system for mobile communications) signalling, rather than from a tachometer, can be used together with a gyroscope sensor to support localization in the presence of temporarily unavailable GPS data. We propose a sensor fusion system architecture and present simulation results that show the effectiveness of this approach. AU - Peschke, Sven AU - Bevermeier, Maik AU - Haeb-Umbach, Reinhold ID - 11881 KW - cellular radio KW - distance measurement KW - global positioning system KW - Global Positioning System KW - global system for mobile communications KW - GPS positioning approach KW - GSM velocity KW - gyroscopes KW - gyroscope sensor KW - inertial navigation KW - inertial navigation system KW - odometer KW - sensor fusion system architecture KW - sensors T2 - 6th Workshop on Positioning Navigation and Communication (WPNC 2009) TI - A GPS positioning approach exploiting GSM velocity estimates ER - TY - CONF AU - Peschke, Sven AU - Bevermeier, Maik AU - Haeb-Umbach, Reinhold ID - 11882 T2 - DGON Navigationskonvent 2009 TI - Verbesserung von GPS-basierter Ortung durch GSM-Geschwindigkeitsschaetzungen ER - TY - JOUR AB - In automatic speech recognition, hidden Markov models (HMMs) are commonly used for speech decoding, while switching linear dynamic models (SLDMs) can be employed for a preceding model-based speech feature enhancement. In this paper, these model types are combined in order to obtain a novel iterative speech feature enhancement and recognition architecture. It is shown that speech feature enhancement with SLDMs can be improved by feeding back information from the HMM to the enhancement stage. Two different feedback structures are derived. In the first, the posteriors of the HMM states are used to control the model probabilities of the SLDMs, while in the second they are employed to directly influence the estimate of the speech feature distribution. Both approaches lead to improvements in recognition accuracy both on the AURORA2 and AURORA4 databases compared to non-iterative speech feature enhancement with SLDMs. It is also shown that a combination with uncertainty decoding further enhances performance. AU - Windmann, Stefan AU - Haeb-Umbach, Reinhold ID - 11937 IS - 5 JF - IEEE Transactions on Audio, Speech, and Language Processing KW - AURORA2 databases KW - AURORA4 databases KW - automatic speech recognition KW - feedback structures KW - hidden Markov models KW - HMM KW - iterative methods KW - iterative speech feature enhancement KW - model probabilities KW - speech decoding KW - speech enhancement KW - speech feature distribution KW - speech recognition KW - switching linear dynamic models TI - Approaches to Iterative Speech Feature Enhancement and Recognition VL - 17 ER - TY - JOUR AB - In this paper, parameter estimation of a state-space model of noise or noisy speech cepstra is investigated. A blockwise EM algorithm is derived for the estimation of the state and observation noise covariance from noise-only input data. It is supposed to be used during the offline training mode of a speech recognizer. Further a sequential online EM algorithm is developed to adapt the observation noise covariance on noisy speech cepstra at its input. The estimated parameters are then used in model-based speech feature enhancement for noise-robust automatic speech recognition. Experiments on the AURORA4 database lead to improved recognition results with a linear state model compared to the assumption of stationary noise. AU - Windmann, Stefan AU - Haeb-Umbach, Reinhold ID - 11938 IS - 8 JF - IEEE Transactions on Audio, Speech, and Language Processing KW - AURORA4 database KW - blockwise EM algorithm KW - covariance analysis KW - linear state model KW - noise covariance KW - noise-robust automatic speech recognition KW - noisy speech cepstra KW - offline training mode KW - parameter estimation KW - speech recognition KW - speech recognition equipment KW - speech recognizer KW - state-space methods KW - state-space model TI - Parameter Estimation of a State-Space Model of Noise for Robust Speech Recognition VL - 17 ER - TY - JOUR AU - Sommer, Christoph AU - Dietrich, Isabel AU - Dressler, Falko ID - 12060 JF - Mobile Networks and Applications SN - 1383-469X TI - Simulation of Ad Hoc Routing Protocols using OMNeT++ ER - TY - JOUR AU - Börstler, Jürgen AU - S. Hall, Mark AU - Nordström, Marie AU - H. Paterson, James AU - Sanders, Kate AU - Schulte, Carsten AU - Thomas, Lynda ID - 15680 IS - 4 JF - SIGCSE Bulletin TI - An evaluation of object oriented example programs in introductory programming textbooks VL - 41 ER - TY - CONF AU - Schulte, Carsten AU - Tolksdorf, Robert ID - 15681 T2 - DeLFI Workshops TI - Qualitätssicherung in einer interaktiven und lerneraktivierenden E-Learning-Umgebung ER - TY - CONF AU - Ehlert, Albrecht AU - Schulte, Carsten ID - 15682 T2 - ICER TI - Empirical comparison of objects-first and objects-later ER - TY - CONF AU - Brinda, Torsten AU - Puhlmann, Hermann AU - Schulte, Carsten ID - 15683 T2 - ITiCSE TI - Bridging ICT and CS: educational standards for computer science in lower secondary education ER - TY - CONF AU - Ehlert, Albrecht AU - Schulte, Carsten ID - 15684 T2 - INFOS TI - Unterschiede im Lernerfolg von Schülerinnen und Sch\üern in Abhängigkeit von der zeitlichen Reihenfolge der Themen (OOP-First bzw. OOP-Later) VL - P-156 ER - TY - CONF AU - Koubek, Jochen AU - Schulte, Carsten AU - Schulze, Peter AU - Witten, Helmut ID - 15685 T2 - INFOS TI - Informatik im Kontext (IniK) - Ein integratives Unterrichtskonzept für den Informatikunterricht VL - P-156 ER - TY - CONF AU - Schulte, Carsten ID - 15686 T2 - INFOS TI - Dualitätsrekonstruktion als Hilfsmittel zur Entwicklung und Planung von Informatikunterricht VL - P-156 ER - TY - CONF AU - Cheng, W. AU - Hüllermeier, Eyke ID - 15773 T2 - In Proceedings MLD-2009 1st International Workshop on Learning from Multi-Label Data, Bled, Slovenia TI - A simple instance-based approach to multilabel classification using the Mallows model ER - TY - CONF AU - Senge, Robin AU - Hüllermeier, Eyke ED - Hoffmann, F. ED - Hüllermeier, Eyke ID - 15774 T2 - in Proceedings 19th Workshop Computational Intelligence, Dortmund Germany TI - Learning pattern tree classifiers using a co-evolutionary algorithm ER - TY - CONF AU - Fober, T. AU - Mernberger, M. AU - Moritz, R. AU - Hüllermeier, Eyke ED - Grosse, I. ED - Neumann, S. ED - Posch, S. ED - Schreiber, F. ED - Stadler, P.. ID - 15775 T2 - In Proceedings GCB-2009 German Conference on Bioinformatics Halle (Saale), Germany TI - Graph-kernels for the comparative analysis of protein active sites ER - TY - CONF AU - Senge, Robin AU - Hüllermeier, Eyke ID - 15776 T2 - In Proceedings Workshop LWA-2009, Lernen-Wissensentdeckung-Adaptivität, Darmstadt, Germany TI - Learning pattern tree classifiers using a co-evolutionary algorithm ER - TY - CONF AU - Fober, T. AU - Mernberger, M. AU - Melnikov, Vitaly AU - Moritz, R. AU - Hüllermeier, Eyke ID - 15778 T2 - In Proceedings Workshop LWA-2009, Lernen-Wissensentdeckung-Adaptivität, Darmstadt, Germany TI - Extension and empirical comparison of graph-kernels for the analysis of protein active sites ER - TY - JOUR AU - Kwong, N. H. AU - Schumacher, Stefan AU - Binder, R. ID - 15847 JF - Physical Review Letters SN - 0031-9007 TI - Electron-Spin Beat Susceptibility of Excitons in Semiconductor Quantum Wells ER - TY - GEN AU - Dahle, Claudia AU - Bäumer, Michaela ID - 14916 TI - Cross-Border Group-Taxation and Loss-Offset in the EU - An Analysis for CCCTB (Common Consolidated Corporate Tax Base) and ETAS (European Tax Allocation System) VL - 66 ER - TY - JOUR AU - Halberstadt, Alexander AU - Sureth-Sloane, Caren AU - Voß, Armin ID - 14917 IS - 6 JF - Die Wirtschaftsprüfung TI - Der Einfluss der Abgeltungssteuer auf die Vorteilhaftigkeit von Anlagen in Genussscheine und Aktien VL - 62 ER -