fondazione bruno kessler

About: fondazione bruno kessler is a facility organization based out in Trento, Italy. It is known for research contribution in the topics: Silicon photomultiplier & Detector. The organization has 1145 authors who have published 4730 publications receiving 94404 citations. The organization is also known as: Trentino Institute of Culture.

Semi-Blind Noise Extraction Using Partially Known Position of the Target Source

Abstract: An extracted noise signal provides important information for subsequent enhancement of a target signal. When the target's position is fixed, the noise extractor could be a target-cancellation filter derived in a noise-free situation. In this paper we consider a situation when such cancellation filters are prepared for a set of several possible positions of the target in advance. The set of filters is interpreted as prior information available for the noise extraction when the target's exact position is unknown. Our novel method looks for a linear combination of the prepared filters via Independent Component Analysis. The method yields a filter that has a better cancellation performance than the individual filters or filters based on a minimum variance principle. The method is tested in a highly noisy and reverberant real-world environment with moving target source and interferers. A post-processing by Wiener filter using the noise signal extracted by the method is able to improve signal-to-noise ratio of the target by up to 8 dB.

Multi-modal vibration based MEMS energy harvesters for ultra-low power wireless functional nodes

Abstract: In this work we discuss a novel design concept of energy harvester (EH), based on Microsystem (MEMS) technology, meant to convert mechanical energy, available in the form of vibrations scattered in the surrounding environment, into electrical energy by means of the piezoelectric conversion principle. The resonant structure, named four-leaf clover (FLC), is circular and based on four petal-like double mass-spring systems, kept suspended through four straight beams anchored to the surrounding Silicon frame. Differently from standard cantilever-type EHs that typically convert energy uniquely in correspondence with the fundamental vibration frequency, this particular shape is aimed to exploit multiple resonant modes and, thereby, to increase the performance and the operation bandwidth of the MEMS device. A preliminary non-optimized design of the FLC is discussed and physical samples of the sole mechanical resonator, fabricated at the DIMES Technology Center (Delft University of Technology, the Netherlands), are experimentally characterized. Their behaviour is compared against simulations performed in ANSYS Workbench™, confirming good accuracy of the predictive method. Furthermore, the electromechanical multiphysical behaviour of the FLC EH is also analysed in Workbench, by adding a layer with piezoelectric conversion properties in the simulation. The measured and simulated data reported in this paper confirm that the MEMS converter exhibits multiple resonant modes in the frequency range below 1 kHz, where most of the environmental vibration energy is scattered, and extracted power levels of 0.2 μW can be achieved as well, in closed-loop conditions. Further developments of this work are expected to fully prove the high-performance of the FLC concept, and are going to be addressed by the authors of this work in the on-going activities.

Convolutional Long Short-Term Memory Networks for Recognizing First Person Interactions

Abstract: In this paper, we present a novel deep learning based approach for addressing the problem of interaction recognition from a first person perspective. The proposed approach uses a pair of convolutional neural networks, whose parameters are shared, for extracting frame level features from successive frames of the video. The frame level features are then aggregated using a convolutional long shortterm memory. The hidden state of the convolutional long short-term memory, after all the input video frames are processed, is used for classification in to the respective categories. The two branches of the convolutional neural network perform feature encoding on a short time interval whereas the convolutional long short term memory encodes the changes on a longer temporal duration. In our network the spatio-temporal structure of the input is preserved till the very final processing stage. Experimental results show that our method outperforms the state of the art on most recent first person interactions datasets that involve complex ego-motion. In particular, on UTKinect-FirstPerson it competes with methods that use depth image and skeletal joints information along with RGB images, while it surpasses all previous methods that use only RGB images by more than 20% in recognition accuracy.

Flexible MISFET Devices From Transfer Printed Si Microwires and Spray Coating

Abstract: This paper presents two types of metal insulator field effect transistors (MISFETs) devices fabricated from Si microwires through a new manufacturing route involving a combination of printing and microfabrication technologies. Si microwires, developed through standard photolithography and etching steps, are transferred from a silicon on insulator wafer onto polyimide using stamp-assisted transfer printing. The MISFETs are then obtained by spray coating the dielectric layer and metal contact layer. Spray coating has been introduced here for the first time for deposition of organic dielectric on transfer printed Si microwires. Two groups of the devices are fabricated, one based on a single Si microwire and the other based on the array of 15 microwires of similar dimensions. The variations in the output response of the two groups of devices has been investigated. The devices based on array of microwires are observed to have less variation in the output response, with lesser standard deviations as compared to MISFETs made from single Si microwires.