Browsing by Author "Tokhi, M. O."
Now showing 1 - 4 of 4
Results Per Page
Sort Options
- Comparative study of different Doppler spectrum estimator implementationsPublication . Madeira, M. M.; Tokhi, M. O.; Ruano, M. GraçaThis paper presents a comparative performance study of some parallel processing architectures aiming at real-time implementation of a blood flow spectral estimator. Three homogeneous and an heterogeneous architecture incorporating transputers, digital signal processors (DSP's) and a vector processor were considered. The performance of these architectures was evaluated and compared in terms of execution times and gradient measurements. Communication times among processing elements were also considered. Analysis of the results revealed that both the DSP's homogeneous architecture and the heterogeneous architecture met real-time requirements; the latter presenting better execution times with less communications time load.
- High-performance computing for real-time spectral estimationPublication . Madeira, M. M.; Bellis, S. J.; Beltran, L. A. A.; Gonzalez, J. S.; Nocetti, D. F. G.; Marnane, W. P.; Tokhi, M. O.; Ruano, M. GraçaThis paper presents two separate investigations into the real-time implementation of the modified covariance spectral estimator: one comparing performances on digital signal processors, the TMS320C40, and the recently released ADSP2016x (SHARC); another using application-specific custom circuitry. The advantages and disadvantages of each of these different approaches are reviewed, resulting in the design of a field-programmable gate array/digital signal processor-based high-performance system that combines the hardware and software approaches. In conjunction with pulsed Doppler ultrasound blood-flow detectors, the spectral estimator offers increased sensitivity in the non-invasive detection of arterial disease. (C) 1999 Elsevier Science Ltd. All rights reserved.
- High-performance real-time implementation of a spectral estimatorPublication . Madeira, M. M.; Beltran, L. A. A.; Gonzalez, J. S.; Nocetti, F. G.; Tokhi, M. O.; Ruano, M. GraçaDoppler blood flow spectral estimation is a common technique of noninvasive cardiovascular disease detection. Blood flow velocity and disturbance may be evaluated by measuring spectral mean frequency and bandwidth respectively. Aiming at minor stenosis diagnosis, parametric spectral estimators may be employed. These models present better spectral resolution than the FFT based ones, at the expense of higher computational burden. Seeking for an efficient real-time implementation of a blood now spectral estimation system, high performance techniques are being investigated. This paper compares the implementation of the Modified Covariance (MC) spectral estimator on two different DSP architectures: the TMS320C40 and the ADSP2016x (SHARC). Implementations are described and their performance assessed. Considerations about portability of algorithms, compiler optimisation levels and system dependence features are addressed. Copyright (C) 1998 IFAC.
- Real-time implementation of a Doppler signal spectral estimator using sequential and parallel processing techniquesPublication . Madeira, M. M.; Tokhi, M. O.; Ruano, M. GraçaDoppler signal spectral estimation has been used to evaluate blood flow parameters in order to diagnose cardiovascular diseases. The modified covariance (MC) method has proved to provide accurate estimation of the two spectral parameters employed in clinical diagnosis, namely mean frequency and bandwidth. The aim of the work reported in this paper is to determine an efficient real-time implementation of the MC spectral estimator by investigating several architectures and implementation methods. A comparative performance analysis of the implementation of the MC algorithm on several homogeneous and heterogeneous architectures incorporating transputers, digital signal processing (DSP) devices and a vector processor is reported. The performances of these architectures are evaluated and compared in terms of computational time (execution and communication) and gradient measurements. Analysis of the results reveals that both the homogeneous and heterogeneous DSP-based parallel architectures meet the real-time requirements. (C) 2000 Elsevier Science B.V. All rights reserved.