|
|
Abstraction in Automata Algorithms
Kocourek, Tomáš ; Lengál, Ondřej (referee) ; Holík, Lukáš (advisor)
Tato práce si klade za cíl implementaci a experimentální porovnání protiřetězcových algoritmů s abstrakcí a bez abstrakce, které testují prázdnost alternujících automatů. Autor také navrhuje vlastní algoritmy s abstrakcí a navrhuje několik optimalizací pro existující abstraktní algoritmy. Práce popisuje teoretické pozadí studovaných algoritmů a navrhuje efektivní způsob implementace datových struktur, které jsou těmito algoritmy používány. Experimentální vyhodnocení na náhodných automatech ukazuje, že algoritmy bez abstrakce vykazují obecně lepší výsledky, neboť nevyužívají náročné operace průniku a komplementace shora a zdola uzavřených množin. V případě automatů s vysokou hustotou přechodů však algoritmy bez abstrakce zpomalují a algoritmy s abstrakcí naopak zrychlují.
|
|
|
Specialized Computer System Automatic Control
Opálka, Jan ; Šátek, Václav (referee) ; Kunovský, Jiří (advisor)
This work deals with the automatic control of calculations of specialized system. The reader is acquainted with the numerical solution of differential equations by Taylor series method and numerical integrators. The practical aim of this work is to analyze parallel characteristics of Taylor series method, specification of parallel math operations and design of control of this operations.
|
|
|
Hardware Realization of Higher Order Numerical Integrator
Matečný, František ; Veigend, Petr (referee) ; Šátek, Václav (advisor)
This work describes numerical integration and solution for ordinary differential equations by the Taylor series by different types of integrators. The next part is a description of floating point and fixed point arithmetic. Subsequently, we are presenting designs and calculation methods for parallels multiplication and division integrators in floating point and fixed point arithmetic. The designs were realized in VHDL and implemented on FPGA. Finally we summarizes the proposed solution and compare time complexity with another numerical methods.
|
|
| |
|
|
Automatic Computation Control
Opálka, Jan ; Šátek, Václav (referee) ; Kunovský, Jiří (advisor)
This work deals with the automatic control of numerical calculations. The reader is acquainted with the numerical solution of differential equations and the parallel, serial, and series-parallel numerical integrator. The practical aim of this work is to design control circuits for the three mentioned variants of integrators. The design includes the development of a software simulator of the control circuit for the series-parallel integrator in a fixed point.
|
|
|
Fractional-Octave Analysis of Acoustic Signals
Ryšavý, Marek ; Frenštátský, Petr (referee) ; Schimmel, Jiří (advisor)
The diploma thesis is focused on design and optimalization of digital octave and fraction-octave band filters. This thesis describe the behavior of filters in systems with fixed point arithmetics and investigate the impact of quantization coefficients for frequency response of filter. Filters, whitch has been designed, are implemented into simple software in C. Designed filters are in accordance with standard IEC 61260.
|
|
|
Implementation of fixed-point arithmetic unit in FPGA
Kalocsányi, Vít ; Fujcik, Lukáš (referee) ; Dvořák, Vojtěch (advisor)
This thesis deals with a design of fixed-point arithmetic unit for FPGA circuits and its model in Matlab. The thesis explains a number representation in digital circuits and both basic and selected additional arithmetic operations with fixed-point numbers. The arithmetic unit’s model is designed in Matlab, the realization of the unit in VHDL is described and its implementation into FPGA is carried out. A specific example of use of designed arithmetic unit’s model for simulation of complex systems in Simulink environment is shown at the end of the thesis.
|
|
|
FPGA implementation of artificial neural network
Čermák, Justin ; Šteffan, Pavel (referee) ; Bohrn, Marek (advisor)
This master's thesis describes the design of effective working artificial neural network in FPGA Virtex-5 series with the maximum use of the possibility of parallelization. The theoretical part contains basic information on artificial neural networks, FPGA and VHDL. The practical part describes the used format of the variables, creating non-linear function, the principle of calculation the single layers, or the possibility of parameter settings generated artificial neural networks.
|
|
|
Loading and Printing ASCII Numbers in FPGA
Závodník, Tomáš ; Zachariášová, Marcela (referee) ; Bartoš, Václav (advisor)
The topic of this work is the issue of processing decimal numbers using binary hardware units. Making use of specialized hardware for this purpose is problematic due to both number systems being incompatible. The thesis is focused specifically on fixed point decimal numbers passed in the form of ASCII character strings and on the FPGA technology. The proposed solution lies in creating hardware units that allow sequential loading and printing of decimal numbers in the mentioned form digit by digit. In terms of the content of this work, it introduces suitable algorithms and describes the realization of the proposed units. It results in their efficient, configurable, portable and reusable implementation.
|
|
|
A Library for Convolutional Neural Network Design
Rek, Petr ; Mrázek, Vojtěch (referee) ; Sekanina, Lukáš (advisor)
In this diploma thesis, the reader is introduced to artificial neural networks and convolutional neural networks. Based on that, the design and implementation of a new library for convolutional neural networks is described. The library is then evaluated on widely used datasets and compared to other publicly available libraries. The added benefit of the library, that makes it unique, is its independence on data types. Each layer may contain up to three independent data types - for weights, for inference and for training. For the purpose of evaluating this feature, a data type with fixed point representation is also part of the library. The effects of this representation on trained net accuracy are put to a test.
|
guest ::
