A partir do espectro de freqüência do sinal eletromiográfico pode-se calcular a freqüência mediana, a freqüência média e a moda da freqüência, podendo se avaliar a fadiga muscular de um indivíduo. Atualmente as análises miográficas, do Depto de Fisioterapia da UNESP - Presidente Prudente, a coleta das medidas e a análise dos resultados são feitas em sistemas separados, isto é, há um sistema para aquisição dos sinais miográficos e outro para análise dos dados, em computadores diferentes. Isto causa problemas pois o indivíduo é submetido ao exercício e somente depois do exercício concluído o sinal é analisado. Com a intenção de melhorar a análise dos sinais miográficos foi desenvolvido um programa computacional no qual os cálculos de FFT e o tratamento do sinal são feitos simultaneamente a aquisição, isto é, em tempo real. A grande vantagem é a possibilidade de se corrigir possíveis erros durante a aquisição devidos a eletrodos mal posicionados, ruídos, evitar o cansaço excessivo do paciente ou até detectar um possível processo de lesão musculara antes que este ocorra. O aplicativo foi implementado em linguagem Visual Basic e rotinas do Matlab, encontra-se em funcionamento e os resultados práticos são discutidos neste trabalho. / From the frequency spectrum of a electromiographic signal the frequency mediana, the mean frequency and the frequency modal can be calculated. Using such parameters the muscular fading of a patient can be evaluated, for instance, using the FFT procedure. In the Phisiotherapy Department of Unesp at Presidente Prudente, experimental data are collected and analyzed using two separated systems. The data corresponding to the electro miographic signal are measured in the first step and then transferred and analyzed in another computer. This procedure is not straightforward since the patient is submitted to a given exercise and only after finishing it the data could be analyzed. Aiming to improve the analysis of the electromiographic signals a software was developed allowing the FFT calculations and the signal analysis to be performed during the data acquisition. The advantage is the possibility to adjust and correct the experimental procedure during the acquisition. Effects of electrodes malpositioning on the patient, electrical noise and excessive fatigue of the patient or even it is possible to detect a muscular lesion during the exercise. The software was developed in the Visual Basic language and using Matlab routines, it is now running and practical results are discussed in this work.

A partir do espectro de freqüência do sinal eletromiográfico pode-se calcular a freqüência mediana, a freqüência média e a moda da freqüência, podendo se avaliar a fadiga muscular de um indivíduo. Atualmente as análises miográficas, do Depto de Fisioterapia da UNESP - Presidente Prudente, a coleta das medidas e a análise dos resultados são feitas em sistemas separados, isto é, há um sistema para aquisição dos sinais miográficos e outro para análise dos dados, em computadores diferentes. Isto causa problemas pois o indivíduo é submetido ao exercício e somente depois do exercício concluído o sinal é analisado. Com a intenção de melhorar a análise dos sinais miográficos foi desenvolvido um programa computacional no qual os cálculos de FFT e o tratamento do sinal são feitos simultaneamente a aquisição, isto é, em tempo real. A grande vantagem é a possibilidade de se corrigir possíveis erros durante a aquisição devidos a eletrodos mal posicionados, ruídos, evitar o cansaço excessivo do paciente ou até detectar um possível processo de lesão musculara antes que este ocorra. O aplicativo foi implementado em linguagem Visual Basic e rotinas do Matlab, encontra-se em funcionamento e os resultados práticos são discutidos neste trabalho. / From the frequency spectrum of a electromiographic signal the frequency mediana, the mean frequency and the frequency modal can be calculated. Using such parameters the muscular fading of a patient can be evaluated, for instance, using the FFT procedure. In the Phisiotherapy Department of Unesp at Presidente Prudente, experimental data are collected and analyzed using two separated systems. The data corresponding to the electro miographic signal are measured in the first step and then transferred and analyzed in another computer. This procedure is not straightforward since the patient is submitted to a given exercise and only after finishing it the data could be analyzed. Aiming to improve the analysis of the electromiographic signals a software was developed allowing the FFT calculations and the signal analysis to be performed during the data acquisition. The advantage is the possibility to adjust and correct the experimental procedure during the acquisition. Effects of electrodes malpositioning on the patient, electrical noise and excessive fatigue of the patient or even it is possible to detect a muscular lesion during the exercise. The software was developed in the Visual Basic language and using Matlab routines, it is now running and practical results are discussed in this work.

In recent years, the Ultra-wideband (UWB) has become one of the most important transmission technologies in the wireless personal area networks (WPAN). In this thesis, we will focus on the multi-band orthogonal frequency division multiplexing (MB-OFDM) based UWB and investigate the hardware implementation of the Fast Fourier Transform (FFT) processor required by the baseband receiver. Due to the requirement of high data rate transmission, a pipeline structure is adopted for parallel processing at a lower clock rate. The hardware implementation is first verified and simulated by using the Verilog hardware description language (HDL). Then, the proposed design is realized by the Taiwan Semiconductor Manufacturing Company (TSMC) 0.13 £gm single-poly eight-metal CMOS process.

During the typhoon season, the coastal Taiwan often receives swells follow by wind waves. The routes of typhoon and the geographic effects complicate the wave conditions. This study investigates the growth and decay of waves at the southwest coast of Taiwan due to two different paths of typhoon. Two different methods of wave analysis are applied: The Fast Fourier Transform (FFT), and the Hilbert Huang Transform (HHT). The results show that 1) the FFT analysis is simple and easy to apply. The HHT, however, provides detail information on wave height and period during typhoon passages. The typhoon went through northern Taiwan induced small waves and period. The reason is that the wind is blocked by the Central Ridge of Taiwan. Waves are generated after typhoon center pass the Taiwan Strait. 2) On the other hand, typhoon went though Luzon Strait induced large swells and wind waves. The geographic effect is the major factor of wave height and period generated. The distance to the typhoon center is another factor producing large waves which is related to the degree of gust. 3) The comparison of wind wave and swell show that the sea is dominated by wind waves during the approaching of typhoon, while swells dominate over when typhoon is leaving.

The goal of this thesis has been to reduce the hardware cost of SDF FFTs. Inorder to achieve this, two methods for simplifying rotations in FFTs are presented:Decimation and Reduction. When applied, these methods reduce the total amountof angles that the rotators need to rotate, as well as the total angle count of theFFT. This is useful for constant shift and add based rotators, as their hardwarecost are typically dependent on the amount of angles it needs to calculate.Decimation works by splitting a large twiddle factor into a smaller one plusan additional small rotator in series. This allows for the possibility to implementlarge FFTs without needing any large twiddle factors. Reduction is a method thattakes a twiddle factor and simplifies it by removing one angle from the rotator.This can be done without adding any hardware cost if applied correctly.In addition to the methods, the thesis also includes proposed designs for 64- upto 1024-point FFTs, as well as post-implementation results for a 32- and 64-pointFFT.

In the region of the harmonic parameter analysis, we often use the Fast Fourier Transform (FFT) to get the character of the signal, but due to the limit conditions of the FFT, the analysis results will appear some error effects, these error effects are: Aliasing effect¡BLeakage effect and Picket-fence effect. When the error effects appear, we can not get the harmonic parameter accurate. For the purpose of getting accuracy harmonic parameters, In this paper proposes, we will introduce the analysis method of FFT and it's limit conditions, then will based on FFT to develop an accuracy formula to find the harmonic parameter fast and accurately. This method will be divided into three parts. First, we will discuss the effect of harmonic to spectrum. Second, we will discuss the parameter calculation with no interfere. Third, we will rid interfere of the harmonic for getting accuracy harmonic parameter. Finally, we will analysis the actual signal to prove the ability of this method, and raise the future research direction for the successor.

Den här rapporten presenterar ett examensarbete där ett mätdatainsamlingssystem för dynamiska signaler har skapats. Arbetet har utförts på Siemens Industrial Turbomachinery AB. Problemet är av sådan natur att tre olika typer av signaler analyseras. Dessa signaler är dels dynamiska signaler med ett intressant frekvensinnehåll mellan 10 och 10 000 Hz, dels likspänningssignaler och dels varvtalssignaler. Dessa signaler datasamplas och analyseras med avseende på både tids- och frekvensinnehåll. Resultatet skickas vidare till en server med en uppdateringsfrekvens kring 1 Hz. Arbetet har inneburit att en kravanalys av systemet har tagits fram utifrån ett redan existerande system och från intervjuer med tänkta användare av systemet. Utifrån den skapade kravprofilen valdes lämplig hårdvara och mjukvara implementerades med hjälp av LabVIEW. Resultatet blev ett system som lever väl upp till förväntningarna.

To efficiently capture signal events when performing analog measurements, a competent toolbox is required. In this master thesis, a system for frequency domain triggering is designed and implemented. The implemented system provides advanced frequency domain trigger conditions, in order to ease the capture of a desired signal event. A real-time 1024-point pipelined feedforward FFT-core is implemented to transform the signal from the time domain to the frequency domain. The system is designed and synthesized for a Virtex-6 FPGA (XC6VLX240T) and is integrated into SP Devices’ digitizer ADQ1600. The implemented system is able to handle a continuous stream of 1.6GS/s at 16-bit. A small software API is developed that provides runtime configuration of the Triggering conditions.

Este trabalho é uma síntese da transformada de Fourier na forma discreta e uma de suas principais aplicações à computação gráfica, a restauração de imagens corrompidas por movimento, seja do aparelho óptico ou da própria imagem.

Este trabalho é uma síntese da transformada de Fourier na forma discreta e uma de suas principais aplicações à computação gráfica, a restauração de imagens corrompidas por movimento, seja do aparelho óptico ou da própria imagem.