J.B. Simoes

A high performance reconfigurable hardware platform for digital pulse processing

This paper presents a new reconfigurable hardware platform which uses both digital signal processors (DSP) and field programmable gate arrays (FPGA) to attain high resolution and real-time processing in nuclear spectrometry experiments. The module was designed in order to provide a high digital pulse processing (DPP) yield with the capacity of being reconfigurable according to the experimental conditions and the desired data output. This allows unprecedented real time processing capabilities implemented in the FPGA such as pulse pileup correction through adaptive filtering and effective signal-to-noise ratio (SNR) control and optimization. The module uses a Virtex-II Pro FPGA (Xilinx) and a DSP from the TMS320C64xx family (Texas Instruments) and is implemented on a PCI board to be used in a host workstation. Special emphasis is given to the scalability of the module along with the potential capacity of being used as a portable stand-alone instrument. This reconfigurable hardware platform allows us to simultaneously benefit from the advantages of the hardware based digital spectrometers, namely, their high throughput, and of the flexibility of the software based configurations. Besides, this same platform can be used as a general purpose high-speed data acquisition and processing unit.

The optical coupling of analog signals

An opto-coupling circuit based on the Siemens IL300 linear optocoupler and the methods to assess its static and dynamic performances are presented. It is shown that IL300, due to its built-in linearizing feedback photodiode, makes it possible to build coupling schemes that associate good linearity with the inherent properties of optical devices, as true galvanic isolation, high isolation voltage and transmission down to DC. The integral linearity, 0.029%, obtained on the coupling of typical nuclear spectroscopy pulses makes us believe that traditional capacitive coupling used in nuclear spectroscopy circuits can be replaced by optical coupling in the near future.

Nuclear spectroscopy pulse height analysis based on digital signal processing techniques

A digital approach to pulse height analysis is presented. It consists of entire pulse digitization, using a flash analog-to-digital converter (ADC), being its height estimated by a floating point digital signal processor (DSP) as one parameter of a model best fitting to the pulse samples. The differential nonlinearity is reduced by simultaneously adding to the pulse, prior to its digitization, two analog signals provided by a digital-to-analog converter (DAC), One of them is a small amplitude dither signal used to eliminate a bias introduced by the fitting algorithm. The other, with large amplitude, corrects the ADC nonlinearities by a method similar to the well known Gattis sliding scale. The simulations carried out showed that, using a 12-bit flash ADC, a 14-bit DAC and a dedicated floating point DSP performing a polynomial fitting to the samples around the pulse peak, it is actually possible to process about 10000 events per second, with a constant height pulse dispersion of only 4 on 8192 channels and a very good differential linearity. A prototype system based on the Texas Instruments floating point DSP TMS320C31 and built following the presented methodology has already been tested and performed as expected. >

A pulse processing station

This is the first of two papers concerning the architecture, circuitry design and performance of a pulse processing system based on a digital signal processor. This multifunction system, implemented as a single PC module, incorporates a high performance 16-bit Pulse Height Analyser (PHA) a Multichannel Scaler (MCS), a Digital Oscilloscope (DSO) and also a Digital Pulse Processor (DPP). This paper presents the PHA architecture with emphasis on the baseline restorer and peak stretcher circuits. Differential nonlinearities (DNL) are corrected by a new implementation of the sliding scale technique and performance ranges from better than 2% (at 16-bit resolution) up to less than 0.2% for 12-bit operation. The DNL correction technique is assessed for different sliding-scale ranges.

Programmable Test Bench for Hemodynamic Studies

The non-invasive assessment of hemodynamic parameters has been a permanent challenge posed to the scientific community. The literature shows many contributions to this quest expressed as algorithms dedicated to revealing some of its characteristics and as new probes or electronics, featuring some enhanced instrumental capability that can improve their insight.

Baseline restoration using current conveyors [nuclear spectrometry systems]

This paper presents simple circuits for baseline restoration based on a commercial current conveyor (CCUII01). Tests were performed, on two circuits, with periodic trapezoidal shaped pulses in order to measure the baseline restoration for several pulse rates and restorer duty cycles. For the current conveyor based Robinson restorer, the peak shift was less than 10 mV, for duty cycles up to 60%, at high pulse rates. Duty cycles up to 80% were also tested, being the maximum peak shift 21 mV. The peak shift for the current conveyor based Grubic restorer was also measured. The maximum value found was 30 mV at 82% duty cycle. Keeping the duty cycle below 60% improves greatly the restorer performance. The ability of both baseline restorer architectures to reject low frequency modulation is also measured, with good results on both circuits.

Optimization of digital spectrometers using a pulse streaming generator

This paper presents a digital pulse streaming generator developed to test and optimize the design of real-time digital spectrometers. The purpose of this generator is to produce a stream of digital data capable of reproducing the typical pulse response of a pre-amplifier used with solid state or other radiation detectors. This process not only avoids the use of X-ray sources during the test and calibration procedures as it also allows the experimentalist to take in account as many stream pulse scenarios as he desires. Using this tool it is possible to simulate a number of signal parameters that are characteristics of each front-end electronic apparatus. Such parameters include pulse amplitude distribution, rise-time components, decay-time of RC feedback pre-amps, frequency of occurrence, ballistic deficit, etc., as well as noise parameters (amplitude, power spectra density, type, etc.). In order to test a specific digital spectrometer implementation, this generator includes step, delta and optionally 1/|f| noise sources. Pulse pile-up effects arise due to the fact that we use a Poisson distribution to reproduce a real stream of events. The digital data stream is directly fed into the FPGA reproducing the behavior of the fast acquisition channel (FADC) of the spectrometer. This is an important auxiliary tool to the design and debug of the FPGA, namely when the implementation of near-optimum signal-to-noise ratio filters is desired.

Baseline restoration using current conveyors

This paper presents simple circuits for baseline restoration based on a commercial current conveyor (CCII01). Tests were performed, on two circuits, with periodic trapezoidal shaped pulses in order to measure the baseline restoration for several pulse rates and restorer duty cycles. For the current conveyor based Robinson restorer, the peak shift was less than 10 mV, for duty cycles up to 60% at high pulse rates. Duty cycles up to 80% were also tested, being the maximum peak shift 21 mV. The peak shift for the current conveyor based Grubic restorer was also measured. The maximum value found was 30 mV at 82% duty cycle. Keeping the duty cycle below 60% improves greatly the restorer performance. The ability of both baseline restorer architectures to reject low frequency modulation is also measured, with good results on both circuits.

Optimized linear pulse amplifier circuit based on a composite op amp configuration

We present a high count rate linear pulse amplifier with bipolar shaping to be used in Nuclear Spectrometry applications. The amplifier is based on a traditional single chain architecture containing an initial passive differentiation with a pole zero cancellation circuit, multi-stage controllable gain, active integration with 4/sup th/ order Butterworth filter and final active differentiation with output buffering. The gain stages use fast cascaded amplifiers with both voltage and current feedback amplifiers to combine the excellent DC performance of the former with the very good dynamic behavior of the last ones. The performance and the design cautions associated to this architecture are discussed.

Characterization of an Acoustic Based Device for Local Arterial Stiffness Assessment

Arterial stiffness, recognized as an independent predictor of cardiovascular events, can be assessed non-invasively by regional and local methods. The present work proposes and describes a novel and low-cost device, based on a double-headed acoustic probe (AP), to assess local arterial stiffness, by means of pulse wave velocity (PWV) measurements. Local PWV is measured over the carotid artery and relies on the determination of the time delay between the signals acquired simultaneously by both acoustic sensors, placed at a fixed distance. The AP was characterized with dedicated test setups, in order to evaluate its performance concerning waveform analysis, repeatability, crosstalk effect and time resolution. Results show that AP signals are repeatable and crosstalk effect do not interfere with its time resolution, when the cross-correlation algorithm for time delay estimation is used. The AP’s effectiveness in measuring higher PWV (14 m/s), with a relative error less than 5 %, when using two uncoupled APs, was also demonstrated. Finally, its clinical feasibility was investigated, in a set of 17 healthy subjects, in which local PWV and other hemodynamic parameters were measured. Carotid PWV yielded a mean value of 2.96 ± 1.08 m/s that is in agreement with the values obtained in other reference studies.