Author: GARCÍA PASTOR, DAVID
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN SIGNAL THEORY AND COMMUNICATIONS
Department: Department of Signal Theory and Communications (TSC)
Mode: Confidentiality
Deposit date: 03/02/2021
Reading date: 09/03/2021
Reading time: 14:00
Reading place: Aula Teleensenyament - B3 - ETSETB - Campus Nord - videoconferència: meet.google.com/frc-hbet-ydy
Thesis director: MATEU MATEU, JORDI | COLLADO GOMEZ, JUAN CARLOS
Committee:
     PRESIDENT: AIGNER, ROBERT
     SECRETARI: UBEDA FARRE, EDUARDO
     VOCAL NO PRESENCIAL: MIREA, TEONA
Thesis abstract: Nowadays, mobile devices have become a key technology in our lives, making us become part of a connected world, in which millions of mobile handsets are sold every year. In order to satisfy the user demands, many mobile communication standards are working together in the same device (LTE-A, 5G-NR, IEEE Wireless LAN Standards, etc.). High data rates are demanded for different wireless services working on a very crowded frequency spectrum, where all the devices and systems need to operate at the same time without detrimental of mutual interference. Although this major objective has to be faced from different perspectives, such as new modulations and new networking protocols, the radio frequency (RF) and microwave components, forming the communication chain in a transceiver, play a significant role on the whole system performance. Passive devices, like filters, contribute to the signal degradation due to passive intermodulation (PIM). PIM can be generated by the lack of mismatch between metallic contacts, oxidation, or basically, by the inherent nonlinear material properties used on the fabrication of the devices. Although the impact of PIM is lower than the distortion introduced by active devices, for current wireless services and their demanding requirements, any type of nonlinear generation is certainly an issue.Acoustic wave (AW) devices have become a fundamental technology for portable handsets supporting current wireless communication networks. Bulk Acoustic Wave (BAW) has become the main technology nowadays for high frequency filters. This technology allows a high degree of miniaturization, easily making more than 40 filters per device, and also gives other advantages as low insertion losses and multiple frequency operation. Despite these major advantages, BAW resonators exhibit a nonlinear behavior, i.e., they may suffer of intermodulation and harmonic distortion, which might become a real bottle neck for the full expansion of the technology in the ever increasing stringent requirements.The aim of this thesis is focused on characterizing this nonlinear limitation, quite present in acoustic resonators, by defining the origin of different nonlinear manifestations in BAW resonators, and providing new nonlinear models to reduce the computational time required to simulate those undesired effects.The first part of this work explains the basic knowledge about acoustic propagation and piezoelectricity, presenting the linear and nonlinear constitutive equations, including the typical nonlinear manifestations appearing in those devices. Then, the typical circuit models for AW devices and, more specifically its applicability to BAW devices, is presented.The second part of this work is focused on the nonlinear analysis of BAW resonators, by measuring different nonlinear manifestation of 8 different BAW devices, identifying the origin of the nonlinear effects and demonstrating that this unique hypothesis is consistent with those manifestations for all the measured resonators.

Author: HÜMBELI, PATRICK
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN PHOTONICS
Department: Institute of Photonic Sciences (ICFO)
Mode: Normal
Deposit date: 11/02/2021
Reading date: pending
Reading time: pending
Reading place: pending
Thesis director: ACÍN DAL MASCHIO, ANTONIO
Committee:
     PRESIDENT: VAN NIEUWENBURG, EVERT
     SECRETARI: GRASS, TOBIAS DANIEL
     VOCAL: DUNJKO, VEDRAN
Thesis abstract: Research at the intersection of machine learning (ML) and quan- tum physics is a recent growing field due to the enormous expec- tations and the success of both fields. ML is arguably one of the most promising technologies that has and will continue to dis- rupt many aspects of our lives. The way we do research is almost certainly no exception and ML, with its unprecedented ability to find hidden patterns in data, will be assisting future scientific discoveries. Quantum physics on the other side, even though it is sometimes not entirely intuitive, is one of the most successful physical theories and we are on the verge of adopting some quan- tum technologies in our daily life. Quantum many-body physics is a subfield of quantum physics where we study the collective behavior of particles or atoms and the emergence of phenomena that are due to this collective behavior, such as phases of matter. The study of phase transitions of these systems often requires some intuition of how we can quantify the order parameter of a phase. ML algorithms can imitate something similar to intu- ition by inferring knowledge from example data. They can, there- fore, discover patterns that are invisible to the human eye which makes them excellent candidates to study phase transitions. At the same time, quantum devices are known to be able to perform some computational task exponentially faster than classical com- puters and they are able to produce data patterns that are hard to simulate on classical computers. Therefore, there is the hope that ML algorithms run on quantum devices show an advantage over their classical analog. This thesis is devoted to study two different paths along the front lines of ML and quantum physics. On one side we study the use of neural networks (NN) to classify phases of mater in many-body quantum systems. On the other side, we study ML algorithms that run on quantum computers. The connection be- tween ML for quantum physics and quantum physics for ML in this thesis is an emerging subfield in ML, the interpretability of learning algorithms. A crucial ingredient in the study of phase transitions with NNs is a better understanding of the predictions of the NN, to eventually infer a model of the quantum system and interpretability can assist us in this endeavor. The interpretabil- ity method that we study analyzes the influence of the training points on a test prediction and it depends on the curvature of the NN loss landscape. This further inspired an in-depth study of the loss of quantum machine learning (QML) applications which we as well will discuss. In this thesis we give answers to the questions of how we can leverage NNs to classify phases of matter and we use a method that allows to do domain adaptation to transfer the learned "in- tuition" from systems without noise onto systems with noise. To map the phase diagram of quantum many-body systems in a fully unsupervised manner, we study a method known from anomaly detection that allows us to reduce the human input to a mini- mum. We will as well use interpretability methods to study NNs that are trained to distinguish phases of matter to understand if the NNs are learning something similar to an order parame- ter and if their way of learning can be made more accessible to humans. And finally, inspired by the interpretability of classical NNs, we develop tools to study the loss landscapes of variational quantum circuits to identify possible differences between classi- cal and quantum ML algorithms that might be leveraged for a quantum advantage.

Author: SOSSA ARANCIBIA, VIVIANA ALEJANDRA
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN EARTHQUAKE ENGINEERING AND STRUCTURAL DYNAMICS
Department: (DECA)
Mode: Normal
Deposit date: 30/11/2020
Reading date: 11/03/2021
Reading time: 11:30
Reading place: PhD thesis by videoconference. Escola Camins. https://meet.google.com/ckz-kgee-amp
Thesis director: PEREZ GRACIA, MARIA DE LA VEGA | GONZALEZ DRIGO, JOSE RAMON
Committee:
     PRESIDENT: GARCÍA GARCÍA, FRANCISCO
     SECRETARI: DI CAPUA, DANIEL
     VOCAL: SOLLA CARRACELA, MERCEDES
Thesis abstract: This thesis aims to studythe capacities and limitations of Ground Penetrating Radar (GPR) to support inspection, assessment, maintenance and rehabilitation tasks, helping to earlydetection and evaluation ofpathologies in unreinforced masonryconstructions, as well as in reinforcing concrete structures.One ofthe most common pathologies in existing reinforced concrete structures is corrosion damage. In fact, this is a pathologythat, in more advanced stages, may comprom ise the integrity of sorne structural elements, or e\en of the whole structural system. Corros ion dam age reduces the effecti\e cross section of steel reinforcing bars , dim inishing the bond between reinforcement and concrete, and affecting, dueto physical and chemical processes, the concrete that inwl\es the corroded steel reinforcing bars. In reinforced concrete structures, the corrosion ofsteel reinforcing bars is, in general, hidden. Accordingly , remotely or at distance assessment techniques are desirable , which aims to assess the integrity of steel elements, generally inaccessible through direct contact.The implementation of non-destructi\e tests (NDT) is a com mon practice when analysing pathologies in historical buildings, where inter'A:lntions should be as minimal invasi\e as possible . Sorne ofthe mostcommon methods to assess corrosion, such as the potential of corrosion, the petrographic analysis and the resistivimeter, present sorne disadvantages in comparison to the GPR. In fact, the uses of such methods require a considerable time, they are slightty invasi\e, and its application is localized, which means thattheir application only co\ers a reduced area of analysi s.In orderto assess the GPR capacity for detecting corros ion dam age, different laboratory tests were respecti\ely de\eloped, being the obtained results extremely satisfactory. Accordingly, it is possible to identify the following achie\ements: (i) a fast data acquisition planning methodologywas well established; (ii) the key elements and parameters for corrosion detection, within a qualitati\e analysis, were respecti\elyidentified; (iii) the contoureffects, related to a change on geometry, were identified and duly controlled; and (iv) a numerical approximation method, based on energy attenuation of GPR, for assessing the presence and degree ofcorrosion in damaged samples, was proposed with success. Obtained results are themsel\es a strong contribution of de\eloped research studies for the state of art.Four in-field tests, implemented on existing structures were respecti\ely developed, aim ing to apply and validate the methodologies, results and conclusions, obtained from laboratory tests. Toree case studies are related to historical constructions, belonging to the cultural heritage of the city of Barcelona, in Spain. The obtained results from GPR, combined, in sorne cases, with boreholes and seism ic tomography, allowed to identify pathologies dueto hum idity, corrosion and differential setttements. A fourth case study, de\eloped on a contemporary structure, in reinforcing concrete, and only using GPR, allowed to identifyand localize pathologies dueto corrosion damage.The laboratorytests and the case studies, using GPR, combined or not with other techniques, showed a strong capacityto detect and assess early stage pathologies, especially dueto corrosion damage, through fast, non-invasi\e and non­ destructi\e prospections.This strategy has econom ical and social advan tages . lndeed, a fast detection of pathologies , allow the rehabilitation of existing structures with a minimum amount of budget, when comparing with the needed resources to face damage in a more se\ere stage. The assessment ofpathologies in earlystages with GPR, especiallywhen combined with other techniques, allow then to ensure more resilient structures, by increasing their reliability, safety, and availability, and by decreasing their operational costs.

Author: KERIDOU, INA
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN POLYMERS AND BIOPOLYMERS
Department: Department of Chemical Engineering (EQ)
Mode: Temporary seizure
Deposit date: 16/02/2021
Reading date: pending
Reading time: pending
Reading place: pending
Thesis director: PUIGGALI BELLALTA, JORGE | FRANCO GARCIA, MARIA LOURDES
Committee:
     PRESIDENT: MEAURIO ARRATE, EMILIANO
     SECRETARI: FERNANDEZ FRANCOS, XAVIER
     VOCAL: REINA LOZANO, JOSE ANTONIO
Thesis abstract: In the last decades, polymeric biomaterials, due to their advanced physical and mechanical properties, have been used in a vast variety of surgical and pharmaceutical applications. In this thesis, a study on crystallinity, properties, and degradability ofpoly-4-hydroxybutyrate (P4HB), commercially marketed as MonoMax® by B. Braun Surgical S.A.U. as an absorbable monofilament suture, and related polyesters is presented. ·Part of this work has been executed under the collaborative research project established between PSEP (Polímeros Sintéticos. Estructura y Propieda s) research group of Universitat Politécnica de Catalunya and B. Braun Surgical S.A.U. (Center of Excellence far Closur :Technologies) with the principal purpose of evolving ahd investigating polymeric systems with specific biomedical ápplications.This study covers different tapies related to P4HB such as (a) the physical, chemical, thermal and mechanical characterization;(b) the study of the crystallization kinetics; (c) the study of the degradation behavior of P4HB; and (d) the production of nanofibers by the electrospinning technique. Moreover, the work includes a study on blends of two different polyesters: PGA/PCL blends also produced employing the electrospinning technique and biphasic PLA/PA blends.Crystallization was studied under both isothermal and non-isothermal conditions using optical microscopy and differential scanning calorimetry. Furthermore, supplementary experiments performed in a synchrotron radiation facility provided us with further information about the lamellae morphology, crystal structure, and molecular orientation.Hydrolytic and enzymatic degradation of different matrices of P4HB was carried out in different buffered media. The hydrolytic degradation has been studied considerírig media of different pH values and temperatures. Enzymatic degradation has also been evatu ated at physiotogical conditions using two different lipases. The hydrolytic degradation mechanism differs from the enzymatic, where bulk degradation and a random éháin scission are characteristic of samples exposed to hydrolytic media, whereas surface erosion and successive removal of monomer units are charac!eristlc of samples exposed to enzymatic media. Far annealed fibers, small-angle X-ray diffraction studies revealed a supramolecular structure with two different types of !amellar stacks. The spherulitic morphology of the enzymatically degraded films was highlighted by the elimination of the amorphous regions.Nanofibers of PGA/PCL and P4HB were preparad using the electrospinning technique. The PGA/PCL nanofibers were also loaded with pharmacological drugs and Lised as a reinforcing agent of biodegradable polymer matrices. On the other hand, the biologicat performance of P4HB nanofibers was achieved with the incorporation of fibroblast growth factors. Far this purpose, wound healing assay for two different cell lines (e.g., epithelial and fibroblast) were studied.Finally, a study on the blends of PLA and Polyamide 6,1O was carried out to investigate the confinement effect of neighboring polymer domains on the phase separation and the structure and the influence of the disperse phase of polyamide in promotingPLA crystallization.

Author: AYMERICH GUBERN, JOAN
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN ELECTRONIC ENGINEERING
Department: Department of Electronic Engineering (EEL)
Mode: Normal
Deposit date: 01/12/2020
Reading date: 18/03/2021
Reading time: 10:00
Reading place: Aula Teleensenyament Edifici B3 - UPC Campus Nord - Videoconferència: https://meet.google.com/psk-qkma-ctk
Thesis director: SERRA GRAELLS, FRANCESC | DEI, MICHELE
Committee:
     PRESIDENT: BALDI COLL, ANTONI
     SECRETARI: MOLL ECHETO, FRANCESC DE BORJA
     VOCAL: PANTELIS, GEORGIOU
Thesis abstract: Electrochemical sensors are expanding rapidly over other chemical sensoring technologies because of their potential to generate precise, selective, miniaturized and cost-effective analytical devices. These features satisfy the emerging global demand for disposable testing systems at the point-of-need, where usability, portability, and price counts most, enabling to detect critical analytical evidence by anyone, any-where and at any time, without concerning about recalibration and limited shelf life. In particular a disposable electrochemical device must include a paired smart electronic interface to specifically bias the electrochemical cell, acquire signals, per-form data conversion and communicate measurements through a standard digital interface, all under severe restrictions of size and power consumption.This thesis describes the development of a novel, cost-effective, disposable, high-performance and user-friendly electrochemical sensing platform that combines the smartness of CMOS integrated circuits (ICs) with the flexibility of printed electronics.Two practical µW-range readout integrated circuit (ROIC) realizations in 65-nm and0.18-µmCMOS technologies are presented and specifically optimized for the potentiostatic biasing and amperometric read-out of electrochemical sensors. The proposed frontend architectures yield very elegant and compact CMOS implementations by reusing the dynamic properties of the sensor itself to implement continuous-time mixed electrochemical delta-sigma modulators (¿SM). The topologies include differential potentiostats to extend its range. Furthermore, low limit of detection (LOD) values can be achieved by implementing a novel cancellation mechanism of the flicker noise coming from the feedback DAC of the electrochemical ¿SM. A standard interface based onI2Cis included on-chip not only to control the extensive system configuration but also to limit the number of IC pads towards a low-cost flip-chip assembly on flexible substrates. Experimental results from both electrical and electrochemical tests are presented and compared to other state-of-the-art electrochemical sensor frontends.A cost-effective hybrid electronics interfacing approach is proposed, where the electrochemical sensor is directly printed on a flexible PEN substrate that also hosts the CMOS readout integrated circuit (ROIC) as a bare die without wire bonding. Low-cost inkjet printing technology is employed for the development of a three-electrode sensor and all the required connectivity. Anisotropic conductive adhesives are investigated as an emerging approach for mechanical and electrical contact between the IC die and printed inks in order to obtain a disposable flexible smart electrochemical sensory device.