UNITA' DI SIENA

Università degli Studi di SIENA

Dipartimento di Ingegneria dell’Informazione e Scienze Matematiche
 

Research topics 

1.              STUDY  AND DEVELOPMENT OF CHEMICAL SENSORS AND SENSOR-BASED ELECTRONIC SYSTEMS

F. Bertocci,  A. Fort,  M. Mugnaini,  S. Rocchi, , V. Vignoli

Collaborations:  University of Florence, University of Brescia, CNR-IFAC Firenze, US Army.

 

2.              ANALYSIS AND DESIGN OF ELECTRONIC CIRCUITS BASED ON NONLINEAR DYNAMICAL SYSTEMS

T. Addabbo, A. Fort, S. Rocchi, V. Vignoli

Collaborations:University of Naples “Federico II”, INLS - University of California San Diego, Macedonian Academy of Sciences and Arts in Skopje
 

3.              OPTIMIZED DESIGN OF DIGITAL ULTRA-LOW-POWER INTEGRATED CIRCUITS

M. Alioto 

Collaboraziotions: University of California Berkeley, University of Michigan Ann Arbor, EPFL, University of Catania, University of Calabria

 

STUDY  AND DEVELOPMENT OF CHEMICAL SENSORS AND SENSOR-BASED ELECTRONIC SYSTEMS

F. Bertocci,  A. Fort,  M. Mugnaini,  S. Rocchi, , V. Vignoli


The research activity concerns the study and development of different chemical sensors, and chemical sensor based measurement systems. In this context a large amount of research work is devoted to the characterization and modeling of Metal Oxide (MOX) sensors and of novel electrochemical nafion-based sensors in collaboration with the University of Perugia. Another field of investigation concerns the use of sensors based on Quartz Chemical Microbalances (QCM). In the past years, a non conventional measurement system was developed for chemical species classification, which exploits the dynamic response of QCMs. For such system new nanostructured materials like PbS and CdS were developed in collaboration with the University of Florence and tested as potential sensing layers to detect small concentration of NOx (ppm order). Polymeric (PEDOT) films were developed as well and grown on AT-cut crystal surfaces to be tested as sensors for NOx and CO. Recently also YCoO3 as well as and Potassium and Silver Hollandites were studied and tested for the development of CO sensors. The experience gained during the last years in this research field, brought the LEEME laboratory to develop sensor-based systems, not only limited to chemical sensors, with a high degree of flexibility, accuracy and measurement speed.  

 

ANALYSIS AND DESIGN OF ELECTRONIC CIRCUITS BASED ON NONLINEAR DYNAMICAL SYSTEMS

T. Addabbo, A. Fort, S. Rocchi, V. Vignoli

This research activity is focused on the design of nonlinear circuits and systems for the generation of stochastic signals with predetermined statistical characteristics, with a special reference to the design of True Random Number Generators (TRNGs) and Pseudo Random Number Generators (PRNGs), and to the development of testing techniques for AD converters. 

As far as the TRNGs are concerned, discrete-time chaotic circuits have been investigated.

In particular, a novel feedback control strategy that allows for reducing the TRNG statistical characteristics deterioration caused by the circuit parameter variability has been proposed. The approach is based on a feedback loop which exploits the estimation of the chaotic system parameters directly from the output bit stream. Concerning the PRNGs, the effcient implementation of discrete-chaotic maps have been proposed, with a particular reference to the digitized Rényi map and the digitized Tent map.

In this research activity the digitization strategy used for the definition of the  discrete-chaotic maps has been theoretically studied, whereas the efficiency of the digital designs is intended in terms of chip area occupation, power consumption and speed. The performances have been compared with those of traditional solutions, such as the Linear Feedback Shift Registers (LFSRs) and the Linear Congruential Generator (LCGs).

The comparison shows that the considered discrete-chaotic maps can exhibit  very good statistical properties, whereas involving a circuit complexity which is comparable to that one of LFSRs, which are well known to represent a lower bound in terms of gate count.

The statistical characteristics of sequences generated by chaotic analog circuits have been also investigated for the definition of techniques for AD converters testing. In detail, exploiting a chaos-based discrete-time noise generator, a method for generating test samples with a distribution arbitrarly close tothe uniform one has been proposed for the Code Density Test (also known as Histogram Test).




OPTIMIZED DESIGN OF DIGITAL ULTRA-LOW-POWER INTEGRATED CIRCUITS
 
M. Alioto 

This research activity is focused on the development of optimized design strategies at the transistor level of abstraction targeting circuits with a high speed, and low/ultra-low power consumption. In particular, the following topics were explored from a research point of view:

- Ultra-low power VLSI circuits: design strategies and circuit techniques for ultra-low power energy-scavenged active RFID tags (in collaboration with Univ. of California – Berkeley), models of standard CMOS logic gates operating in subthreshold, current-mode logic styles for extremely low power circuits.

- FinFET VLSI circuits: analysis of leakage-delay tradeoff, as well as layout efficiency versus process parameters and circuit approach.

- Flip-flops in nanometer CMOS technologies: extensive analysis and comparison including the effect arising at sub-100nm technology generations (leakage, process/voltage/temperature variations, impact of layout parasitics), design strategies and techniques, design strategies for energy-efficient clock networks.

- Process variability and impact on VLSI circuits: modelling and comparison of static and dynamic logic styles in CMOS technology, design metrics for variability-aware synthesis of VLSI circuits.

- Low standby current caches: innovative approaches to reduce leakage associated with the array of 6T caches.

 

Ultime notizie

Questo sito utilizza cookies. Continuando la navigazione si accetta di riceverli.