Giovanni Frattini

Hardware-Algorithms Co-Design and Implementation of an Analog-to-Information Converter for Biosignals Based on Compressed Sensing

We report the design and implementation of an Analog-to-Information Converter (AIC) based on Compressed Sensing (CS). The system is realized in a CMOS 180 nm technology and targets the acquisition of bio-signals with Nyquist frequency up to 100 kHz. To maximize performance and reduce hardware complexity, we co-design hardware together with acquisition and reconstruction algorithms. The resulting AIC outperforms previously proposed solutions mainly thanks to two key features. First, we adopt a novel method to deal with saturations in the computation of CS measurements. This allows no loss in performance even when 60% of measurements saturate. Second, the system is able to adapt itself to the energy distribution of the input by exploiting the so-called rakeness to maximize the amount of information contained in the measurements.

Practical Optimization of EMI Reduction in Spread Spectrum Clock Generators With Application to Switching DC/DC Converters

We here consider the most common technique used in spread spectrum clock generators that is the frequency modulation of a timing signal by means of a triangularly shaped waveform. As a first step, we develop a reliable mathematical model of a spectrum analyzer, which allows us to compute the power spectrum as measured by this instrument for any signal put at its input. This is particularly important when considering spread spectrum clocking methods for electromagnetic interference reduction, since international regulations impose constraints on the peak of the spectrum of interfering signals as measured by this instrument. Thanks to the developed mathematical tool, we are able to theoretically prove that the maximum peak reduction of the measured spectrum is achieved for a well-defined frequency of the triangular driving signal. This is in contrast with what one can obtain by optimizing the theoretical power density spectrum, where the minimum interference is ideally obtained when the triangular signal has a vanishing frequency. The results are confirmed by measurements on two commercial dc/dc switching converters.

Short-term Optimized Spread Spectrum Clock Generator for EMI Reduction in Switching DC/DC Converters

We present here a prototype of a Spread Spectrum Clock Generator designed for switching DC/DC converters, which has been optimized in accordance to EMC international regulations to improve EMI reduction with respect to standard approaches. Such an optimization has been obtained trough the theoretical computation of the power spectrum as measured by a spectrum analyzer when a chaotic or random-like PAM modulating signal is employed. To experimentally validate the achieved theoretical results, the circuit has been fabricated in CMOS 0.18 μm technology, and embeds two simple and effective modulators, the first one relying on a full analog hardware, the second one on a simple digital core. Measurements on a DC/DC converter in accordance with EMC regulations confirm that the prototype can achieve a 7 dB EMI reduction improvement with respect to classical solutions based on out of audible triangular waveforms.

Coping with saturating projection stages in RMPI-based Compressive Sensing

Though compressive sensing hinges on extracting linear measurements from the signals to acquire, actual implementations introduce nonlinearities whose effect can be far from negligible. We here address the problem of saturation in the circuit blocks needed by a Random Modulation Pre-Integration architecture. To allow a fair a comparison with previous analysis, we rely on a model capturing the essentials of saturations in actual implementations while being able to reproduce more abstract settings considered in the literature. Based on this, we analyze some methods already proposed to cope with simplified saturation mechanisms, briefly discussing their underlying principles. Finally, we introduce a novel approach that takes into account the more realistic model and, at the cost of an almost negligible hardware overhead, is extremely effective in countering saturation effects.

An Analytical Approach for the Design of Class-E Resonant DC–DC Converters

We present a new approach to design resonant dc-dc converters, that allows us to achieve both a more accurate implementation and a simpler architecture, by reducing the number of required passive components. The approach is applied to a class-E topology, and it is based on the analytic solution of the system of differential equations regulating the converter evolution. Our technique is also capable of taking into account the most important circuit nonidealities. This represents an important breakthrough with respect to the state of the art, where class-E circuit analysis is based on strong simplifying assumptions, and the final circuit design is achieved by means of numerical simulations after many time-consuming parametric sweeps. The developed methodology is dimensionless, and the achieved design curves can be denormalized to easily get the desired circuit design. Measurements on two different prototypes confirm an extremely high adherence to the developed mathematical approach.

A new semi-analytic approach for class-E resonant DC-DC converter design

This paper presents a new approach for the design of a class-E resonant dc-dc converter. The small number of passive components featured by the considered topology allows to exactly solve the differential equations regulating the circuit evolution, and to develop a semi-analytic design procedure based on the differential equations solution. This represents an important breakthrough with respect to the state-of-the-art, where class-E circuit analysis is always based on strong simplifying assumptions, and the exact circuit design is achieved by means of numerical simulations after many time-consuming parametric sweeps.

Realtime ECG baseline removal: An isoelectric point estimation approach

This paper presents a novel method for ECG baseline drift removal while preserving the integrity of the ST segment. Baseline estimation is achieved by tracking 3 isoelectric points within the ECG waveform as fiducial markers used in an interpolation filter. These points are determined relative to the QRS complex, which is extracted using a known method (Pan-Tompkins algorithm). The proposed algorithm has been tested extensively using synthetic signals and also validated with real data. The synthetic signals assume a 2mV p-p ECG signal and 300 μV p-p baseline drift in the presence of noise artefacts including EMG pickup (20 dB - max. 200 μV), and residual power-line interference (50 μV p-p). The results show a maximum (worst-case ST-segment distortion) error of 34.7 μV (mean), 27.8 μV (median) and 21.2 μV (std. dev.) across 50 randomly generated synthetic ECG signals each containing 100 heartbeats. Validation of the algorithm applied to signals from the MIT-BIH arrhythmia databases reveals maximum error per P-T interval with mean, median and std. dev. of 34.4 μV, 35.2 μV and 9.6 μV respectively with suppressed motion artefacts.

The ZetaBoost: A step-up DC/DC topology derived from the Zeta converter

This work investigates a step-up topology derived from the Zeta converter: the ZetaBoost. A ZetaBoost converter design for consumer applications is presented and its steady-state performances are compared to a classical Boost converter with the support of computer simulations and measurements. The results presented include the efficiency and output voltage ripple.

A first implementation of a semi-analytically designed class-E resonant DC-DC converter

Resonant power converters represent a step further in the effort of increasing the operating frequency, and consequently the power density, with respect to conventional switching converter architectures. Nevertheless, resonant converters are used only in very specific applications. The main issue is their design that, being not based on a solid mathematical background, results in a non-trivial task. In this paper we present a prototype of a class-E resonant converter with a simplified architecture, allowing both a small size (and so a higher density) and a simple mathematical analysis. Conversely with respect to the state-of-the-art approach, the circuit design is obtained by means of a semi-analytic mathematical approach without any support from circuital simulation. Measurements confirm the performance expected according to the mathematical model, and prove that the design of circuits with the proposed architecture can be effectively achieved with the developed mathematical model.

Modeling Hysteresis Losses in Magnetic Core Inductors for DC–DC Conversion

This article proposes the implementation of a method for modeling hysteresis loops and related losses in magnetic core inductors used for power converters applications. The method has been applied to some commercially available ferrite materials in order to verify the accordance of B–H loops and power losses density estimation with acceptable results. Once correctly calibrated, the model could help to characterize magnetic materials hysteresis losses and to optimize the inductor core material choice and dimensioning in high-power density converters design.