TCAD approaches to multidimensional simulation of advanced semiconductor devices

Technology scaling increasingly emphasizes complexity and non-ideality of the electrical behavior of semiconductor devices and boosts interest on alternatives to the conventional planar MOSFET architecture. TCAD simulation tools are fundamental to the analysis and development of new technology generations. However, the increasing device complexity is reflected in an augmented dimensionality of the problems to be solved. The trade-off between accuracy and computational cost of the simulation is especially influenced by domain discretization: mesh generation is therefore one of the most critical steps and automatic approaches are sought. Moreover, the problem size is further increased by process variations, calling for a statistical representation of the single device through an ensemble of microscopically different instances. The aim of this thesis is to present multi-disciplinary approaches to handle this increasing problem dimensionality in a numerical simulation perspective. The topic of mesh generation is tackled by presenting a new Wavelet-based Adaptive Method (WAM) for the automatic refinement of 2D and 3D domain discretizations. Multiresolution techniques and efficient signal processing algorithms are exploited to increase grid resolution in the domain regions where relevant physical phenomena take place. Moreover, the grid is dynamically adapted to follow solution changes produced by bias variations and quality criteria are imposed on the produced meshes. The further dimensionality increase due to variability in extremely scaled devices is considered with reference to two increasingly critical phenomena, namely line-edge roughness (LER) and random dopant fluctuations (RD). The impact of such phenomena on FinFET devices, which represent a promising alternative to planar CMOS technology, is estimated through 2D and 3D TCAD simulations and statistical tools, taking into account matching performance of single devices as well as basic circuit blocks such as SRAMs. Several process options are compared, including resist- and spacer-defined fin patterning as well as different doping profile definitions. Combining statistical simulations with experimental data, potentialities and shortcomings of the FinFET architecture are analyzed and useful design guidelines are provided, which boost feasibility of this technology for mainstream applications in sub-45 nm generation integrated circuits.

Porous Polymeric Bioresorbable Scaffolds for Tissue Engineering

Tissue engineering is a discipline that aims at regenerating damaged biological tissues by using a cell-construct engineered in vitro made of cells grown into a porous 3D scaffold. The role of the scaffold is to guide cell growth and differentiation by acting as a bioresorbable temporary substrate that will be eventually replaced by new tissue produced by cells. As a matter or fact, the obtainment of a successful engineered tissue requires a multidisciplinary approach that must integrate the basic principles of biology, engineering and material science. The present Ph.D. thesis aimed at developing and characterizing innovative polymeric bioresorbable scaffolds made of hydrolysable polyesters. The potentialities of both commercial polyesters (i.e. poly-e-caprolactone, polylactide and some lactide copolymers) and of non-commercial polyesters (i.e. poly-w-pentadecalactone and some of its copolymers) were explored and discussed. Two techniques were employed to fabricate scaffolds: supercritical carbon dioxide (scCO2) foaming and electrospinning (ES). The former is a powerful technology that enables to produce 3D microporous foams by avoiding the use of solvents that can be toxic to mammalian cells. The scCO2 process, which is commonly applied to amorphous polymers, was successfully modified to foam a highly crystalline poly(w-pentadecalactone-co-e-caprolactone) copolymer and the effect of process parameters on scaffold morphology and thermo-mechanical properties was investigated. In the course of the present research activity, sub-micrometric fibrous non-woven meshes were produced using ES technology. Electrospun materials are considered highly promising scaffolds because they resemble the 3D organization of native extra cellular matrix. A careful control of process parameters allowed to fabricate defect-free fibres with diameters ranging from hundreds of nanometers to several microns, having either smooth or porous surface. Moreover, versatility of ES technology enabled to produce electrospun scaffolds from different polyesters as well as “composite” non-woven meshes by concomitantly electrospinning different fibres in terms of both fibre morphology and polymer material. The 3D-architecture of the electrospun scaffolds fabricated in this research was controlled in terms of mutual fibre orientation by properly modifying the instrumental apparatus. This aspect is particularly interesting since the micro/nano-architecture of the scaffold is known to affect cell behaviour. Since last generation scaffolds are expected to induce specific cell response, the present research activity also explored the possibility to produce electrospun scaffolds bioactive towards cells. Bio-functionalized substrates were obtained by loading polymer fibres with growth factors (i.e. biomolecules that elicit specific cell behaviour) and it was demonstrated that, despite the high voltages applied during electrospinning, the growth factor retains its biological activity once released from the fibres upon contact with cell culture medium. A second fuctionalization approach aiming, at a final stage, at controlling cell adhesion on electrospun scaffolds, consisted in covering fibre surface with highly hydrophilic polymer brushes of glycerol monomethacrylate synthesized by Atom Transfer Radical Polymerization. Future investigations are going to exploit the hydroxyl groups of the polymer brushes for functionalizing the fibre surface with desired biomolecules. Electrospun scaffolds were employed in cell culture experiments performed in collaboration with biochemical laboratories aimed at evaluating the biocompatibility of new electrospun polymers and at investigating the effect of fibre orientation on cell behaviour. Moreover, at a preliminary stage, electrospun scaffolds were also cultured with tumour mammalian cells for developing in vitro tumour models aimed at better understanding the role of natural ECM on tumour malignity in vivo.

Heterogeneous multicore systems for signal processing

This thesis explores the capabilities of heterogeneous multi-core systems, based on multiple Graphics Processing Units (GPUs) in a standard desktop framework. Multi-GPU accelerated desk side computers are an appealing alternative to other high performance computing (HPC) systems: being composed of commodity hardware components fabricated in large quantities, their price-performance ratio is unparalleled in the world of high performance computing. Essentially bringing “supercomputing to the masses”, this opens up new possibilities for application fields where investing in HPC resources had been considered unfeasible before. One of these is the field of bioelectrical imaging, a class of medical imaging technologies that occupy a low-cost niche next to million-dollar systems like functional Magnetic Resonance Imaging (fMRI). In the scope of this work, several computational challenges encountered in bioelectrical imaging are tackled with this new kind of computing resource, striving to help these methods approach their true potential. Specifically, the following main contributions were made: Firstly, a novel dual-GPU implementation of parallel triangular matrix inversion (TMI) is presented, addressing an crucial kernel in computation of multi-mesh head models of encephalographic (EEG) source localization. This includes not only a highly efficient implementation of the routine itself achieving excellent speedups versus an optimized CPU implementation, but also a novel GPU-friendly compressed storage scheme for triangular matrices. Secondly, a scalable multi-GPU solver for non-hermitian linear systems was implemented. It is integrated into a simulation environment for electrical impedance tomography (EIT) that requires frequent solution of complex systems with millions of unknowns, a task that this solution can perform within seconds. In terms of computational throughput, it outperforms not only an highly optimized multi-CPU reference, but related GPU-based work as well. Finally, a GPU-accelerated graphical EEG real-time source localization software was implemented. Thanks to acceleration, it can meet real-time requirements in unpreceeded anatomical detail running more complex localization algorithms. Additionally, a novel implementation to extract anatomical priors from static Magnetic Resonance (MR) scansions has been included.

La Previsione dei Consumi Elettrici nel Mercato Energetico Italiano : un nuovo modello empirico con struttura triangolare

L'obiettivo principale della tesi è lo sviluppo di un modello empirico previsivo di breve periodo che sia in grado di offrire previsioni precise ed affidabili dei consumi di energia elettrica su base oraria del mercato italiano. Questo modello riassume le conoscenze acquisite e l'esperienza fatta durante la mia attuale attività lavorativa presso il Romagna Energia S.C.p.A., uno dei maggiori player italiani del mercato energetico. Durante l'ultimo ventennio vi sono stati drastici cambiamenti alla struttura del mercato elettrico in tutto il mondo. Nella maggior parte dei paesi industrializzati il settore dell'energia elettrica ha modificato la sua originale conformazione di monopolio in mercato competitivo liberalizzato, dove i consumatori hanno la libertà di scegliere il proprio fornitore. La modellazione e la previsione della serie storica dei consumi di energia elettrica hanno quindi assunto un ruolo molto importante nel mercato, sia per i policy makers che per gli operatori. Basandosi sulla letteratura già esistente, sfruttando le conoscenze acquisite 'sul campo' ed alcune intuizioni, si è analizzata e sviluppata una struttura modellistica di tipo triangolare, del tutto innovativa in questo ambito di ricerca, suggerita proprio dal meccanismo fisico attraverso il quale l'energia elettrica viene prodotta e consumata nell'arco delle 24 ore. Questo schema triangolare può essere visto come un particolare modello VARMA e possiede una duplice utilità, dal punto di vista interpretativo del fenomeno da una parte, e previsivo dall'altra. Vengono inoltre introdotti nuovi leading indicators legati a fattori meteorologici, con l'intento di migliorare le performance previsive dello stesso. Utilizzando quindi la serie storica dei consumi di energia elettrica italiana, dall'1 Marzo 2010 al 30 Marzo 2012, sono stati stimati i parametri del modello dello schema previsivo proposto e valutati i risultati previsivi per il periodo dall'1 Aprile 2012 al 30 Aprile 2012, confrontandoli con quelli forniti da fonti ufficiali.

Efficacia delle griglie in titanio con osso particolato nella ricostruzione dei difetti alveolari tridimensionali dei mascellari

Questa tesi valuta l’efficacia della tecnica delle griglie in titanio con osso particolato nella ricostruzione dei difetti alveolari tridimensionali ai fini della riabilitazione dentale implanto-protesica. Il primo studio ha considerato la metodica in termini di complicanze post-operatorie e di risultati implanto-protesici. Sono stati considerati 24 pazienti con difetti tridimensionali trattati con l’applicazione di 34 griglie di titanio e osso particolato e riabilitati protesicamente dopo circa 8-9 mesi. 4 su 34 griglie sono state rimosse prima dell’inserimento implantare (11.76% di fallimento totale); 20 su 34 griglie si sono esposte per deiscenza dei tessuti molli (58.82% di complicanze): 4 (11.77%) prima e 16 (47.05%) dopo le prime 4-6 settimane dall’intervento; in nessun caso il piano di trattamento implanto-protesico ha subito variazioni. Dopo un follow-up medio di 20 (3-48) mesi dal carico protesico, nessuno degli 88 impianti ha perso la propria osteo-integrazione (100% di sopravvivenza implantare), con un valore complessivo di successo implantare di 82.9%. Il secondo studio ha calcolato in termini volumetrici la ricostruzione ossea ottenuta con griglie e la sua corre-lazione con l’estensione dell’esposizione e la tempistica del suo verificarsi. Sono stati valutati 12 pazienti con 15 difetti alveolari. Per ciascun sito sono state studiate le immagini TC con un software dedicato per misurare i volumi in tre dimensioni: il volume di osso non formatosi rispetto a quanto pianificato, lacking bone volume (LBV), è stato calcolato sottraendo il volume di osso ricostruito, reconstructed bone volume (RBV) in fase di ri-entro chirurgico dal volume di osso pianificato pre-operativamente, planned bone volume (PBV). LBV è risultato direttamente proporzionale all’area di esposizione della griglia, con un valore del 16.3% di LBV per ogni cm2 di griglia esposta. Si sono evidenziate, inoltre, correlazioni positive tra LBV , la tempistica precoce di esposizione e il valore di PBV.

Scene Reconstruction And Understanding By RGB-D Sensors

This thesis investigates interactive scene reconstruction and understanding using RGB-D data only. Indeed, we believe that depth cameras will still be in the near future a cheap and low-power 3D sensing alternative suitable for mobile devices too. Therefore, our contributions build on top of state-of-the-art approaches to achieve advances in three main challenging scenarios, namely mobile mapping, large scale surface reconstruction and semantic modeling. First, we will describe an effective approach dealing with Simultaneous Localization And Mapping (SLAM) on platforms with limited resources, such as a tablet device. Unlike previous methods, dense reconstruction is achieved by reprojection of RGB-D frames, while local consistency is maintained by deploying relative bundle adjustment principles. We will show quantitative results comparing our technique to the state-of-the-art as well as detailed reconstruction of various environments ranging from rooms to small apartments. Then, we will address large scale surface modeling from depth maps exploiting parallel GPU computing. We will develop a real-time camera tracking method based on the popular KinectFusion system and an online surface alignment technique capable of counteracting drift errors and closing small loops. We will show very high quality meshes outperforming existing methods on publicly available datasets as well as on data recorded with our RGB-D camera even in complete darkness. Finally, we will move to our Semantic Bundle Adjustment framework to effectively combine object detection and SLAM in a unified system. Though the mathematical framework we will describe does not restrict to a particular sensing technology, in the experimental section we will refer, again, only to RGB-D sensing. We will discuss successful implementations of our algorithm showing the benefit of a joint object detection, camera tracking and environment mapping.