Data-driven Signal Processing per Jamming Detection in Sistemi Satellitari

In questo elaborato vengono analizzate differenti tecniche per la detection di jammer attivi e costanti in una comunicazione satellitare in uplink. Osservando un numero limitato di campioni ricevuti si vuole identificare la presenza di un jammer. A tal fine sono stati implementati i seguenti classificatori binari: support vector machine (SVM), multilayer perceptron (MLP), spectrum guarding e autoencoder. Questi algoritmi di apprendimento automatico dipendono dalle features che ricevono in ingresso, per questo motivo è stata posta particolare attenzione alla loro scelta. A tal fine, sono state confrontate le accuratezze ottenute dai detector addestrati utilizzando differenti tipologie di informazione come: i segnali grezzi nel tempo, le statistical features, le trasformate wavelet e lo spettro ciclico. I pattern prodotti dall’estrazione di queste features dai segnali satellitari possono avere dimensioni elevate, quindi, prima della detection, vengono utilizzati i seguenti algoritmi per la riduzione della dimensionalità: principal component analysis (PCA) e linear discriminant analysis (LDA). Lo scopo di tale processo non è quello di eliminare le features meno rilevanti, ma combinarle in modo da preservare al massimo l’informazione, evitando problemi di overfitting e underfitting. Le simulazioni numeriche effettuate hanno evidenziato come lo spettro ciclico sia in grado di fornire le features migliori per la detection producendo però pattern di dimensioni elevate, per questo motivo è stato necessario l’utilizzo di algoritmi di riduzione della dimensionalità. In particolare, l'algoritmo PCA è stato in grado di estrarre delle informazioni migliori rispetto a LDA, le cui accuratezze risentivano troppo del tipo di jammer utilizzato nella fase di addestramento. Infine, l’algoritmo che ha fornito le prestazioni migliori è stato il Multilayer Perceptron che ha richiesto tempi di addestramento contenuti e dei valori di accuratezza elevati.

Motor Unit Spike Trains Recontruction from sEMG Signals for Gesture Classification on a Low-Power Processing Platform

Hand gesture recognition based on surface electromyography (sEMG) signals is a promising approach for the development of intuitive human-machine interfaces (HMIs) in domains such as robotics and prosthetics. The sEMG signal arises from the muscles' electrical activity, and can thus be used to recognize hand gestures. The decoding from sEMG signals to actual control signals is non-trivial; typically, control systems map sEMG patterns into a set of gestures using machine learning, failing to incorporate any physiological insight. This master thesis aims at developing a bio-inspired hand gesture recognition system based on neuromuscular spike extraction rather than on simple pattern recognition. The system relies on a decomposition algorithm based on independent component analysis (ICA) that decomposes the sEMG signal into its constituent motor unit spike trains, which are then forwarded to a machine learning classifier. Since ICA does not guarantee a consistent motor unit ordering across different sessions, 3 approaches are proposed: 2 ordering criteria based on firing rate and negative entropy, and a re-calibration approach that allows the decomposition model to retain information about previous sessions. Using a multilayer perceptron (MLP), the latter approach results in an accuracy up to 99.4% in a 1-subject, 1-degree of freedom scenario. Afterwards, the decomposition and classification pipeline for inference is parallelized and profiled on the PULP platform, achieving a latency < 50 ms and an energy consumption < 1 mJ. Both the classification models tested (a support vector machine and a lightweight MLP) yielded an accuracy > 92% in a 1-subject, 5-classes (4 gestures and rest) scenario. These results prove that the proposed system is suitable for real-time execution on embedded platforms and also capable of matching the accuracy of state-of-the-art approaches, while also giving some physiological insight on the neuromuscular spikes underlying the sEMG.

Neural Radiance Fields for Relighting and View Synthesis

Neural scene representation and neural rendering are new computer vision techniques that enable the reconstruction and implicit representation of real 3D scenes from a set of 2D captured images, by fitting a deep neural network. The trained network can then be used to render novel views of the scene. A recent work in this field, Neural Radiance Fields (NeRF), presented a state-of-the-art approach, which uses a simple Multilayer Perceptron (MLP) to generate photo-realistic RGB images of a scene from arbitrary viewpoints. However, NeRF does not model any light interaction with the fitted scene; therefore, despite producing compelling results for the view synthesis task, it does not provide a solution for relighting. In this work, we propose a new architecture to enable relighting capabilities in NeRF-based representations and we introduce a new real-world dataset to train and evaluate such a model. Our method demonstrates the ability to perform realistic rendering of novel views under arbitrary lighting conditions.

Synthesis of thioenol ether functionalized cyclooctyne and its orthogonal cycloaddition sequence

In this work, the synthesis of a new bifunctionalized cyclooctyne for a possible layer by layer surface functionalization is presented. The main objective is to find a more stable molecule than the literature known methyl enol ether substituted cyclooctyne. Accordingly, the two target functionalities are an internal alkyne group and a vinyl methyl sulfide group. The synthesis was achieved in 9 steps and consists first of all in the preparation of an aldehyde starting from 1,5-cyclooctadiene with a cyclopropanation reaction followed by a reduction and the SWERN oxidation to an aldehyde. The new functionality was introduced by exploiting the WITTIG reaction. For the alkyne group a bromination followed by a double elimination gave good results. The reactivity of the new molecule was tested using a sequential application of SPAAC and iEDDA reactions, comparing it with the cyclooctyne functionalized with a methyl enol ether. Concerning the comparison of both compounds the sulfur ether is significantly slower and therefore more stable. It will be tested in the future for surface functionalization from the KOERT group.

Studio del comportamento meccanico e tribologico della lega AlSi10Mg prodotta mediante Laser-based Powder Bed Fusion sottoposta ad un ciclo tecnologico ottimizzato

AlSi10Mg alloy is one of the most widely used alloys for producing structural components by Laser-based Powder Fusion (L-PBF) technology due to the high mechanical and technological properties. The present work aims to characterize mechanically and tribologically the L-PBF AlSi10Mg alloy subjected to both heat treatment and surface modification cycles. Specifically, the effects of three heat treatments on the tribological and mechanical properties of the alloy were analyzed: T5 (artificial aging at 160 °C for 4 h), T6 rapid solution heat treatment (solution heat treatment at 510 °C for 1h and aging at 160 °C for 6 h), and T6 benchmark (solution heat treatment at 540 °C for 1h and aging at 160 °C for 4 h), the latter used as a benchmark. The study highlighted how the better balance between strength and ductility properties induced by the introduction of heat treatments leads to lower wear resistance and not significant variations in the friction coefficient of the alloy. The tribological and mechanical behavior of the alloy coated with two different coating structures, consisting of (i) chemical Ni (Ni-P) and (ii) Ni-P + DLC, was also evaluated. The goal was the identification of a deposition cycle such as to guarantee the optimization of the mechanical and tribological behavior of the alloy. The Ni-P coating provided good wear resistance but an increase in the coefficient of friction. In contrast, using the DLC top coating resulted in excellent tribological performance in wear resistance and friction coefficient. The samples characterized by the Ni-P + DLC multilayer coating were subsequently subjected to mechanical characterization. The results obtained highlighted problems of adhesion and incipient breaking of the material due to the different mechanical behavior of the coating, considerably reducing the mechanical performance of the alloy coated with Ni-P+DLC multilayer solution compared to the specimens in the un-coated condition.

Life cycle assessment (lca) of six separation methods of polyethylene and aluminum in tetra pak

In food and beverage industry, packaging plays a crucial role in protecting food and beverages and maintaining their organoleptic properties. Their disposal, unfortunately, is still difficult, mainly because there is a lack of economically viable systems for separating composite and multilayer materials. It is therefore necessary not only to increase research in this area, but also to set up pilot plants and implement these technologies on an industrial scale. LCA (Life Cycle Assessment) can fulfil these purposes. It allows an assessment of the potential environmental impacts associated with a product, service or process. The objective of this thesis work is to analyze the environmental performance of six separation methods, designed for separating the polymeric from the aluminum fraction in multilayered packaging. The first four methods utilize the chemical dissolution technique using Biodiesel, Cyclohexane, 2-Methyltetrahydrofuran (2-MeTHF) and Cyclopentyl-methyl-ether (CPME) as solvents. The last two applied the mechanical delamination technique with surfactant-activated water, using Ammonium laurate and Triethanolamine laurate as surfactants, respectively. For all six methods, the LCA methodology was applied and the corresponding models were built with the GaBi software version 10.6.2.9, specifically for LCA analyses. Unfortunately, due to a lack of data, it was not possible to obtain the results of the dissolution methods with the solvents 2-MeTHF and CPME; for the other methods, however, the individual environmental performances were calculated. Results revealed that the methods with the best environmental performance are method 2, for dissolution methods, and method 5, for delamination methods. This result is confirmed both by the analysis of normalized and weighted results and by the analysis of 'original' results. An hotspots analysis was also conducted.

Fabrication by magnetron sputtering and characterization of electrodes based on CuIn0.7Ga0.3Se2 (CIGS)

This thesis work aims to produce and test multilayer electrodes for their use as photocathode in a PEC device. The electrode developed is based on CIGS, a I-III-VI2 semiconductor material composed of copper (Cu), indium (In), Gallium (Ga) and selenium (Se). It has a bandgap in the range of 1.0-2.4 eV and an absorption coefficient of about 105cm−1, which makes it a promising photocathode for PEC water splitting. The idea of our multilayer electrode is to deposit a thin layer of CdS on top of CIGS to form a solid-state p–n junction and lead to more efficient charge separation. In addition another thin layer of AZO (Aluminum doped zinc oxide) is deposit on top of CdS since it would form a better alignment between the AZO/CdS/CIGS interfaces, which would help to drive the charge transport further and minimize charge recombination. Finally, a TiO2 layer on top of the electrodes is used as protective layer during the H2 evolution. FTO (Fluorine doped tin oxide) and Molybdenum are used as back-contact. We used the technique of RF magnetron sputtering to deposit the thin layers of material. The structural characterization performed by XDR measurement confirm a polycrystalline chalcopyrite structural with a preferential orientation along the (112) direction for the CIGS. From linear fit of the Tauc plot, we get an energy gap of about 1.16 eV. In addition, from a four points measurements, we get a resistivity of 0.26 Ωcm. We performed an electrochemical characterization in cell of our electrodes. The results show that our samples have a good stability but produce a photocurrent of the order of μA, three orders of magnitude smaller than our targets. The EIS analysis confirm a significant depletion of the species in front of the electrode causing a lower conversion of the species and less current flows.