| |
|
Parallel signal processing with help of GPU
Václavík, Jiří ; Frýza, Tomáš (referee) ; Mego, Roman (advisor)
In the introduction, the bachelor thesis outlines the origins of modern graphic processors. The theoretical part of the text describes the minimum of required information from parallel programming model essential to program simple DSP algorithms. The next part elaborate on three common DSP algorithms, finite impulse response filter, naive implementation of discrete cosine transform type II, and fast Fourier transform. To demonstrate parallel capability of GPU, algorithm for JPEG compression was chosen as JPEG compression is favorable because it illustrates both advantages and disadvantages of parallel data processing on GPU, and compromises needed to be considered.
|
|
Active Loudspeaker Crossover with DSP
Václavík, Jiří ; Fedra, Zbyněk (referee) ; Povalač, Aleš (advisor)
Semestral thesis is devoted to mapping and verification technologies for the project active loudspeaker crossover using DSP. Three way systém will use open baffle. Design of analog crossover is complex due to the need of compensating for acoustic short circuit and properties of laudspeakers.
|
| |
|
Debugger for Windows
Václavík, Jiří ; Yaghob, Jakub (advisor) ; Bednárek, David (referee)
This work deals with analysis of available debugging means of the operating system Microsoft Windows XP on the IA-32 platform and ensuing implementation of an application debugger. The resulting implementation (the project RazeDebugger) supports multithreaded debugging, SIMD instructions, and viewing of accessible debugging information from source code. The work also includes comparison of the project RazeDebugger and other existing debuggers. The comparison aims at these aspects of debuggers: available functions, user friendliness, CPU usage, performance, stability, and also reaction to special behavior of the debuggee.
|
|
CSG modeling for polygonal objects
Václavík, Jiří ; Křivánek, Jaroslav (advisor) ; Beneš, Jan (referee)
This work deals with an efficient and robust technique of performing Boolean operations on polygonal models. Full robustness is achieved within an internal representation based on planes and BSP (binary space partitioning) trees, in which operations can be carried out exactly in mere fixed precision arithmetic. Necessary conversions from the usual representation to the inner one and back, including their consequences are analyzed in detail. The performance of the method is optimized by a localization scheme in the form of an adaptive octree. The resulting implementation RazeCSG is experimentally compared with implementations used in practice Carve and Maya, which are not fully robust. For large models, RazeCSG shows only twice lower performance in the worst case than Carve, and is at least 130 times faster than Maya.
|
| |
| |
|
Active Loudspeaker Crossover with DSP
Václavík, Jiří ; Fedra, Zbyněk (referee) ; Povalač, Aleš (advisor)
Semestral thesis is devoted to mapping and verification technologies for the project active loudspeaker crossover using DSP. Three way systém will use open baffle. Design of analog crossover is complex due to the need of compensating for acoustic short circuit and properties of laudspeakers.
|
|
CSG modeling for polygonal objects
Václavík, Jiří ; Křivánek, Jaroslav (advisor) ; Beneš, Jan (referee)
This work deals with an efficient and robust technique of performing Boolean operations on polygonal models. Full robustness is achieved within an internal representation based on planes and BSP (binary space partitioning) trees, in which operations can be carried out exactly in mere fixed precision arithmetic. Necessary conversions from the usual representation to the inner one and back, including their consequences are analyzed in detail. The performance of the method is optimized by a localization scheme in the form of an adaptive octree. The resulting implementation RazeCSG is experimentally compared with implementations used in practice Carve and Maya, which are not fully robust. For large models, RazeCSG shows only twice lower performance in the worst case than Carve, and is at least 130 times faster than Maya.
|