Performance Evaluation of Doppler Frequency Estimation with Variable Range Sampling

碩士 === 國立臺灣大學 === 電信工程學研究所 === 87 === A new technique for determining the velocity in PW (Pulse Wave) Doppler system is presented. Since Doppler frequencies exceeding half PRF (Pulse Repetition Frequency) will exhibit aliasing, they contribute with a wrong sign (and magnitude) to the Doppler frequen...

Full description

Bibliographic Details
Main Authors: Ta-Jung Su, 蘇大榮
Other Authors: JenHo Taso
Format: Others
Language:en_US
Published: 1999
Online Access:http://ndltd.ncl.edu.tw/handle/51432345343651350935
id ndltd-TW-087NTU00435048
record_format oai_dc
spelling ndltd-TW-087NTU004350482016-02-01T04:12:41Z http://ndltd.ncl.edu.tw/handle/51432345343651350935 Performance Evaluation of Doppler Frequency Estimation with Variable Range Sampling 變距取樣於都卜勒頻率估測之效能評估 Ta-Jung Su 蘇大榮 碩士 國立臺灣大學 電信工程學研究所 87 A new technique for determining the velocity in PW (Pulse Wave) Doppler system is presented. Since Doppler frequencies exceeding half PRF (Pulse Repetition Frequency) will exhibit aliasing, they contribute with a wrong sign (and magnitude) to the Doppler frequency causing a significant error in the estimate. The PRF is limited by the depth of penetration: emission of a next burst is only allowed if the current amplitude of returned signal of previous burst has been sufficiently subsided, otherwise aliasing in depth will occur. The maximum velocity which can be detected without range/velocity ambiguity is limited by Nyquist criterion indirectly. Therefore, a compromise between PRF and detectable range is currently existed. In spite of the fact that frequency can be down shifted by enlarging the angle or decreasing the center frequency, the error increases as the observation angle increases or the center frequency decreases. Accurate extraction of Doppler shifts using current signal processing approaches is extremely difficult because Doppler ultrasound signals are highly non-stationary. The finite size of sample volume will cause spectral broadening, that is, a single velocity induces a band of frequencies. The width depends on the angle of observation and the spatial sensitivity of the Doppler system. A much simpler treatment for evaluating the velocity precisely by tracking the minimum phase variation as well as selecting alternative sampling line in time domain with employing autocorrelation method is described in this paper. These algorithms give rigorously accurate values in terms of selecting an optimal tracking line for velocity estimation and outperform the conventional time domain or frequency domain methods. In addition, the maximum detectable depth and velocity are not constrained mutually as the sign is taken into consideration. Experimental results are shown which confirm the superiority of these techniques. JenHo Taso 曹建和 1999 學位論文 ; thesis 92 en_US
collection NDLTD
language en_US
format Others
sources NDLTD
description 碩士 === 國立臺灣大學 === 電信工程學研究所 === 87 === A new technique for determining the velocity in PW (Pulse Wave) Doppler system is presented. Since Doppler frequencies exceeding half PRF (Pulse Repetition Frequency) will exhibit aliasing, they contribute with a wrong sign (and magnitude) to the Doppler frequency causing a significant error in the estimate. The PRF is limited by the depth of penetration: emission of a next burst is only allowed if the current amplitude of returned signal of previous burst has been sufficiently subsided, otherwise aliasing in depth will occur. The maximum velocity which can be detected without range/velocity ambiguity is limited by Nyquist criterion indirectly. Therefore, a compromise between PRF and detectable range is currently existed. In spite of the fact that frequency can be down shifted by enlarging the angle or decreasing the center frequency, the error increases as the observation angle increases or the center frequency decreases. Accurate extraction of Doppler shifts using current signal processing approaches is extremely difficult because Doppler ultrasound signals are highly non-stationary. The finite size of sample volume will cause spectral broadening, that is, a single velocity induces a band of frequencies. The width depends on the angle of observation and the spatial sensitivity of the Doppler system. A much simpler treatment for evaluating the velocity precisely by tracking the minimum phase variation as well as selecting alternative sampling line in time domain with employing autocorrelation method is described in this paper. These algorithms give rigorously accurate values in terms of selecting an optimal tracking line for velocity estimation and outperform the conventional time domain or frequency domain methods. In addition, the maximum detectable depth and velocity are not constrained mutually as the sign is taken into consideration. Experimental results are shown which confirm the superiority of these techniques.
author2 JenHo Taso
author_facet JenHo Taso
Ta-Jung Su
蘇大榮
author Ta-Jung Su
蘇大榮
spellingShingle Ta-Jung Su
蘇大榮
Performance Evaluation of Doppler Frequency Estimation with Variable Range Sampling
author_sort Ta-Jung Su
title Performance Evaluation of Doppler Frequency Estimation with Variable Range Sampling
title_short Performance Evaluation of Doppler Frequency Estimation with Variable Range Sampling
title_full Performance Evaluation of Doppler Frequency Estimation with Variable Range Sampling
title_fullStr Performance Evaluation of Doppler Frequency Estimation with Variable Range Sampling
title_full_unstemmed Performance Evaluation of Doppler Frequency Estimation with Variable Range Sampling
title_sort performance evaluation of doppler frequency estimation with variable range sampling
publishDate 1999
url http://ndltd.ncl.edu.tw/handle/51432345343651350935
work_keys_str_mv AT tajungsu performanceevaluationofdopplerfrequencyestimationwithvariablerangesampling
AT sūdàróng performanceevaluationofdopplerfrequencyestimationwithvariablerangesampling
AT tajungsu biànjùqǔyàngyúdōubolēipínlǜgūcèzhīxiàonéngpínggū
AT sūdàróng biànjùqǔyàngyúdōubolēipínlǜgūcèzhīxiàonéngpínggū
_version_ 1718174465198653440