Politecnio die Bari - Catalogo di prodotti della Ricerca
Not a member yet
36616 research outputs found
Sort by
An Improved Charge Recycling Bipolar Pulser Architecture for Ultrasound Imaging
No full textHigh-voltage pulsers are crucial in medical ultrasound imaging systems, significantly influencing both transmitter's signal quality and power efficiency. In this context, we introduce a new structure for a 5-level bipolar pulser based on a voltage-doubling architecture, that is extended with four additional switches to enable charge recycling and dual-supply operation. Circuit-level simulations demonstrate substantial improvements, including reduced power consumption, enhanced harmonic rejection and higher fundamental output voltage
Toward Holistic Evaluation of Recommender Systems Powered by Generative Models
No full textRecommender systems powered by generative models (Gen-RecSys) extend beyond classical item-ranking by producing open-ended content, which simultaneously unlocks richer user experiences and introduces new risks. On one hand, these systems can enhance personalization and appeal through dynamic explanations and multi-turn dialogues. On the other hand, they might venture into unknown territory-hallucinating nonexistent items, amplifying bias, or leaking private information. Traditional accuracy metrics cannot fully capture these challenges, as they fail to measure factual correctness, content safety, or alignment with user intent. This paper makes two main contributions. First, we categorize the evaluation challenges of Gen-RecSys into two groups: (i) existing concerns that are exacerbated by generative outputs (e.g., bias, privacy) and (ii) entirely new risks (e.g., item hallucinations, contradictory explanations). Second, we propose a holistic evaluation approach that includes scenario-based assessments and multi-metric checks-incorporating relevance, factual grounding, bias detection, and policy compliance. Our goal is to provide a guiding framework so that researchers and practitioners can thoroughly assess Gen-RecSys, ensuring both effective personalization and responsible deployment
Reconstruction of the benzene absorption band at 14.8 μm in amplitude modulation exploiting a quartz tuning fork as optical detector
No full textThis work presents a single-pass optical sensor designed for the detection of a broadband absorption feature of benzene. It is based on light-induced thermoelastic absorption spectroscopy (LITES) employing a novel InAs/ AlSb-based quantum cascade laser source emitting at 14.85 mu m as light source, and a custom-designed quartz tuning fork (QTF) as infrared photodetector. The sensor system was operated in both amplitude (AM) and wavelength modulation (WM) mode to characterize the absorption band parameters and assess the sensor ultimate detection limit. The limitations of WM and double-frequency detection (2f-WM) approach for broadband absorption features are thoroughly analyzed and discussed. An absorption cross-section of (1.1 +/- 0.1) center dot 10 cm2 was measured in AM mode with a minimum detection limit of 1.6 part-per-million, while a minimum detection limit of 0.6 part-per-million was achieved using the 2f-WM mode, at 0.1 s of integration time, employing a 12 cm long single-pass cell
IoD-Sim: an open-source simulator for 6G integrated terrestrial/non-terrestrial networks
No full textIntegrated Terrestrial/Non-Terrestrial Networks (T/NTNs) are paving the way to ubiquitous connectivity in future Sixth Generation (6G) communication systems, as they enable novel heterogeneous environments with drones, High-Altitude Platforms (HAPs), and satellites. For the challenges brought by this diverse spectrum of mobile network entities, it is imperative to have tools capable of assessing such a complex network infrastructure before moving to a real-world deployment and testing phase. To this end, this work emphasizes the need for a comprehensive system-level communication simulation platform to study and develop these emerging architectures. It surveys existing simulators and introduces the Internet of Drones Simulator (IoD-Sim), an open-source platform designed for T/NTN simulations, including the envisioned 6G technology of Intelligent Reflective Surfaces (IRSs)
Adhesive contact of a viscoelastic fibrillar surface — A homogenized model
No full textA fibrillar interface is modeled as a regular array of cylindrical micropillars bonded to a substrate, with a focus on the viscoelastic properties of the fibrils. A one-dimensional linear constitutive model describes the coupled deformation of an individual fibril and the substrate. For a compliant viscoelastic substrate, the interaction backing layer effect between fibrils through the substrate deformations is also accounted for. In the case of a viscoelastic Winkler-type foundation whose adhesive mechanism is described by the Shrimali–Lopez-Pamies criterion of maximum rate-independent elongation of the foundation elements, an exact analytical solution is derived for the displacement-controlled loading protocol. A leading-order discrete asymptotic model is developed for the Schapery-type rate-independent adhesive contact between the viscoelastic fibrillar substrate and a rigid punch. By neglecting the influence of the backing layer, a homogenized model is derived in detail. The debonding incubation time is introduced, and an analytical approximation for the pull-off force is obtained under conditions of strong adhesion and fast unloading after a long dwell time
Dual-Wavelength Add-Drop filters with aperiodic gratings on a Hybrid SiN-TFLN Platform
No full textWe demonstrate a compact dual-wavelength optical filter based on a contra-directional coupler with superimposed Bragg gratings on a hybrid SiN-TFLN (Silicon Nitride on thin-film lithium niobate) platform. The device supports add-drop functionality at 1533 nm and 1563 nm through an aperiodic grating formed by the superposition of two distinct Bragg grating periods. This structure eliminates the need for cascaded designs while achieving low insertion losses of -0.70 dB and -1.39 dB. Three-dimensional Finite Difference Time Domain (3D-FDTD) simulations demonstrate effective spectral separation with a 30 nm channel spacing, making the design suitable for sparse WDM applications
Insights and perspectives on entrainment and detrainment in natural stratified flows
No full textThe entrainment hypothesis, first introduced by Taylor and later refined by Turner, has been instrumental in elucidating the mechanisms by which turbulence engulfs ambient fluid. Although entrainment dynamics has been extensively analyzed in geophysical applications, detrainment, the expulsion of fluid from turbulent regions, has received comparatively less scrutiny. This review synthesizes the current understanding of entrainment and detrainment, emphasizing their role in natural and engineered flow systems. Particular attention is paid to recent investigations of detrainment in obstructed plane jets, where the presence of obstacles significantly modifies shear layer development, mixing efficiency, and scalar transport properties. A critical evaluation of the prevailing models and their inherent limitations is presented, along with prospective research directions aimed at enhancing predictive frameworks for applications such as sediment transport, pollutant dispersion, and atmospheric boundary layer modeling
Formulation and test of a predictive dispatch problem for the optimal management of energy resources and operating reserve in isolated distribution networks
No full textEnergy transition and decarbonization pose significant challenges to isolated distribution networks, historically dependent on conventional fuel-fired generation. Lacking external control resources, these systems must face several operational issues due to the increasing integration of intermittent energy resources. This paper focuses on the development of a predictive dispatch problem for optimal energy resource management and operating reserve allocation in isolated networks. A key aspect of this study is the integration of reserve constraints within the designed predictive control. In this regard, the paper evaluates two different probability-based methods for reserve assessment, examining their performance under different levels of RES integration and conservativeness. Furthermore, the paper addresses different ways of allocating storage resources in the mathematical formulation of the proposed problem. To validate the impact of the discussed reserve assessment and allocation methods on the predictive control's effectiveness, a comprehensive testing algorithm is presented. This algorithm allows a thorough evaluation of the performance of various predictive control designs, including both static day-ahead and recursive dispatch approaches. The algorithm also integrates real-time system behavior, capturing the interplay between predictive scheduling and actual system evolution, and providing a realistic assessment of overall system performance. The performed analysis includes network simulations based on a real-world case study of a small Italian island's distribution network, characterized by significant seasonal load variations and increasing RES penetration scenarios. The analysis of test results seeks to provide valuable insights to practitioners, enabling them to design and implement analogous optimal control strategies for the efficient operation of isolated power grids
Adaptive User Modeling in Visual Merchandising: Balancing Brand Identity with Operational Efficiency
No full textMaintaining a consistent brand identity across a global network of retail stores while adhering to local constraints has long challenged Visual Merchandisers. Legacy processes, often reliant on subjective "by-eye" adjustments, can drive up operational costs and lead to inconsistent in-store execution. We formalize a user modeling framework implementing a multi-criteria utility function that balances brand identity and operational overhead. We integrated our framework in a 3D virtual tour design platform, deploying it in the ecosystem of OVS, a global fashion firm. Through a preliminary user study, we showcase that our solution enables lower iteration cycles and decreases store-to-store discrepancies