Enabling Techniques and Algorithms for Integrated Communication and Navigation Satellite Systems

This thesis presents the outcomes of my Ph.D. course in telecommunications engineering. The focus of my research has been on Global Navigation Satellite Systems (GNSS) and in particular on the design of aiding schemes operating both at position and physical level and the evaluation of their feasibility and advantages. Assistance techniques at the position level are considered to enhance receiver availability in challenging scenarios where satellite visibility is limited. Novel positioning techniques relying on peer-to-peer interaction and exchange of information are thus introduced. More specifically two different techniques are proposed: the Pseudorange Sharing Algorithm (PSA), based on the exchange of GNSS data, that allows to obtain coarse positioning where the user has scarce satellite visibility, and the Hybrid approach, which also permits to improve the accuracy of the positioning solution. At the physical level, aiding schemes are investigated to improve the receiver’s ability to synchronize with satellite signals. An innovative code acquisition strategy for dual-band receivers, the Cross-Band Aiding (CBA) technique, is introduced to speed-up initial synchronization by exploiting the exchange of time references between the two bands. In addition vector configurations for code tracking are analyzed and their feedback generation process thoroughly investigated.

Hybrid Artificial Intelligence to Extract Patterns and Rules from Argumentative and Legal Texts

This Thesis is composed of a collection of works written in the period 2019-2022, whose aim is to find methodologies of Artificial Intelligence (AI) and Machine Learning to detect and classify patterns and rules in argumentative and legal texts. We define our approach “hybrid”, since we aimed at designing hybrid combinations of symbolic and sub-symbolic AI, involving both “top-down” structured knowledge and “bottom-up” data-driven knowledge. A first group of works is dedicated to the classification of argumentative patterns. Following the Waltonian model of argument and the related theory of Argumentation Schemes, these works focused on the detection of argumentative support and opposition, showing that argumentative evidences can be classified at fine-grained levels without resorting to highly engineered features. To show this, our methods involved not only traditional approaches such as TFIDF, but also some novel methods based on Tree Kernel algorithms. After the encouraging results of this first phase, we explored the use of a some emerging methodologies promoted by actors like Google, which have deeply changed NLP since 2018-19 — i.e., Transfer Learning and language models. These new methodologies markedly improved our previous results, providing us with best-performing NLP tools. Using Transfer Learning, we also performed a Sequence Labelling task to recognize the exact span of argumentative components (i.e., claims and premises), thus connecting portions of natural language to portions of arguments (i.e., to the logical-inferential dimension). The last part of our work was finally dedicated to the employment of Transfer Learning methods for the detection of rules and deontic modalities. In this case, we explored a hybrid approach which combines structured knowledge coming from two LegalXML formats (i.e., Akoma Ntoso and LegalRuleML) with sub-symbolic knowledge coming from pre-trained (and then fine-tuned) neural architectures.

Fabrication and characterization of hybrid ferromagnetic-organic heterostructures for spintronics application

Recent research in the field of organic spintronics highlighted the peculiar spin-dependent properties of the interface formed by an organic semiconductor (OSC) chemisorbed over a 3d ferromagnetic metal, also known as spinterface. The hybridization between the molecular and metallic orbitals, typically π orbitals of the molecule and the d orbitals of the ferromagnet, give rise to spin dependent properties that were not expected by considering the single components of interfaces, as for example the appearance of a magnetic moment on non-magnetic molecules or changes in the magnetic behavior of the ferromagnet. From a technological viewpoint these aspects provide novel engineering schemes for spin memory and for spintronics devices, featuring unexpected interfacial magnetoresistance, spin-filtering effects and even modulated magnetic anisotropy. Applications of these concepts to devices require nevertheless to transfer the spinterface effects from an ideal interface to room temperature operating thin films. In this view, my work presents for the first time how spinterface effects can be obtained even at room temperature on polycrystalline ferromagnetic Co thin films interfaced with organic molecules. The considered molecules were commercial and widely used in the field of organic electronics: Fullerene (C60), Gallium Quinoline (Gaq3) and Sexithiophene (T6). An increase of coercivity, up to 100% at room temperature, has been obtained on the Co ultra-thin films by the deposition of an organic molecule. This effect is accompanied by a change of in-plane anisotropy that is molecule-dependent. Moreover the Spinterface effect is not limited to the interfacial layer, but it extends throughout the whole thickness of the ferromagnetic layer, posing new questions on the nature of the 3d metal-molecule interaction.