Signal processing is one of the most important research and education areas of the Department of Measurement and Information Systems, Budapest University of Technology and Economics. Signal processing methods can be divided into two main groups, based on whether structural or parametric information about the system under test are used during the processing or none. This latter method is more general, they are not limited to problem-groups, however, they limitations come from their generality, as in most cases they do not use all available information about the system. Thus, methods using parametric or structural models promise significant new results. At the same time, due to the diversity of physical systems, a whole family of the specialized methods is required for adequate signal processing. Some important topics, which is examined by this proposal: Examination of system identification (system model creation) problems. Restoration of signals distorted by the measurement system, using model of the measurement system: inverse-filtering. Active noise cancellation/reduction by using models of the environment. Theory and application of new adaptive filter structures to model different phenomena and to create more robust algorithms. Examination of the quantization and round-off errors of the analog-digital conversion and signal processing algorithms to understand their behavior and to minimize their side-effects. Sound synthesis methods based on different (signal or physics-based) models of the musical instruments.