Superconducting single flux quantum (SFQ) logic is an attractive novel approach to perform logic in the post-CMOS era. Its energy-delay is significantly better than CMOS and other logic technologies, but it requires a fundamental paradigm shift in materials, interconnect, and design.
This project will deal with studying the properties and opportunities of SFQ logic. Then, new designs and methodologies will be developed.
Ferroelectric Field Effect Transistors are emerging devices that can be used as non-volatile memories. This project, as part of a collaboration with Prof. Tso-Ping Ma from Yale University, is to study the fundamentals of FeFET and design circuits to support memory operations in FeFET arrays.
Deep learning algorithms have become the most successful machine learning methods.
In this project we will study the application of deep learning for communications systems, combining communications theory with machine learning.
This project is a joint collaboration with Prof. Andrea Goldsmith from Stanford.
Augmented and Virtual Reality are going to become a part of our lives in the coming years, with the first generation of products already on the market. These new technologies enable visualization of virtual objects that look realistic and vivid. This has opened up a need for solutions that enable physical interaction with virtual objects, to allow us to ``touch’’ and ``feel’’ with our hands the virtual environment.
In this project we will be analyzing images and videos in order to build tactile representations for them that are good for haptic sensing.
Physics and math have many fundamental constants that appear naturally in completely different processes, yet are sometimes related in surprising ways. Examples appear in self-similarity of fractals in chaos theory, phase transitions in percolation processes, and other critical parameters in statistical physics, as well as in combinatorics and number theory.
The project's goal is to develop methodical methods to find new mathematical connections between seemingly unrelated constants by applying advanced statistical and machine-learning algorithms.
This may lead to new equations in physics and mathematics that can help us "reverse engineer" new laws of nature.
Super-resolution fluorescence microscopy is considered the next step in imaging of the sub-cellular dynamics of living cells. The 2014 Nobel Prize was awarded to the PALM and STED methods for achieving 10X sub-diffraction resolution.
In this project we will develop new super-resolution techniques, applied to state-of-the-art microscopy imaging modalities in order to achieve high spatio-temporal resolution of living sub-cellular processes.
The project is a joint collaboration with Prof. Michal Irani from the Weizmann institute and leading researchers in microscopy.
Recent years have shown an increased interest in the use of traditional signal processing methods and operations on signals defined over the nodes of a graph.
In this project we explore the meaning of deep networks over graphs with the aim to extend traits of deep networks to graph signals.
Deep neural networks have shown remarkable performance in many tasks over the past few years.
However, many of the theoretical issues regarding training, choice of nonlinearity, exploiting structure and more are not yet well understood.
In this project, we will investigate some of these issues and explore their applications in various signal processing tasks.
Memristors are novel circuit elements that can be used both as memory devices and as building blocks in electronic circuits. This project aims to design logic gates and digital circuits, where memristors are combined with CMOS transistors. These circuits will have higher area efficiency and therefore they have the potential to go beyond Moore's law.
Cellular neural networks (CNN) are a parallel computing paradigm similar to neural network with the difference that communication is allowed between neighboring units only. CNN are attractive for applications such as image processing, biological vision, solving partial differential equations and more.
In this project, memristors, novel circuit elements, will be used in CNN to exploit their capability to combine learning and memory together. Memristive CNN can outperform conventional CNN with significant better area efficiency.
Memristors are novel circuit elements typically used as memory devices. Memristors can also be used in artificial neural networks mostly to store the weight of a synapse. The target of this project is to use memristors as binary weighted elements and investigate the appropriate design of them that can be used to efficiently execute various machine learning algorithms.
Speech signal processing technologies, which have made significant strides in the last few decades, now play various important roles in our daily lives. For example, speech communication technologies such as (mobile) telephones, video-conference systems, and hearing aids are widely available as tools that assist communication between humans. Speech recognition technology now enables a wide spectrum of innovative and exciting voice-driven applications. However, most of these applications consider a microphone located near the talker as a prerequisite for reliable performance, which prevents further proliferation. In this project, we explore the challenging problem of reverberation. Can one use distant microphones to capture speech and still obtain good automatic speech recognition (ASR) performance?
Given an image, parts of it attract our immediate attention - these are the important, or the "salient", parts. A fundamental challenge in computer vision is the detection of the salient pixels automatically. In the attached image, these are the white areas. New 3D cameras pose a new challenge - given a set of of points, can we detect the salient points automatically? In this project we wish to study the relation between the two problems: 2D saliency detection that is based on colors and 3D saliency detection that is based on geometry.
We work on the reconstruction of the internal structure of thick objects that cannot be seen directly, using new time of flight imaging technologies, combined with new optimization algorithms.
We will work on the development of cool cutting edge display technologies capable of displaying 3D data without glasses, reproducing highlights and reacting to illumination. We will use modern tools in computer vision, computer graphics and image processing.
Neuromorphic devices emulate some aspects of neuron activity, for future hardware based neural networks. Examples are memristors and FLASH devices.
Within the project electron trapping in gallium nitride transistors will be evaluated as a new technology for neuromorphic devices. Based on the experimental results, improved devices will be designed, fabricated, and tested.
The purpose of the research project is to investigate
the HDR protocol and to evaluate its performance. Our goal here is to extend and adapt the
algorithm to larger topologies and configurations and to investigate its
performance via a comprehensive simulation program. The research project will be performed in
collaboration with Prof. Gil Zussman from Columbia University.
-- Find more info in the PDF file --
We developed a novel device that allows quantifying and characterizing the eyelid motion using magnetic field. Thereby, facilitates diagnosing neurological diseases and medication effects.
The goal of this project is to identify blinks and extract parameters (such as energy, frequency and velocity) in order to characterize the eyelid motion. In the framework of the project we will analyze clinical data collected by doctors from Emek Medical Center in Afula.
נתגלה לאחרונה שלאור ופוטונים בסיבים ומהודים אופטיים יש תופעות תרמיות ועיבוי מיוחדות (קונדנסצית בוז-איינשטיין). אנו עוסקים בחקירת התופעות הללו .
מטרת הפרויקט הזה הוא בניית פלטפורמה של סיבים אופטיים וחקירת התופעות שתאפשרנה את העיבוי של הפוטונים.
סריגים בסיבים אופטיים הם רכיבים מרכזיים בתחום של התקשורת האופטית עבור צרכים של סינון וניתוב של אור בסיבים, וכן עבור יצירת לייזרי סיבים ועוד.
אנו לומדים ויוצרים סיבים כאלו במעבדה ומשתמשים בהם לצרכים שונים.
מטרת הפרויקט הוא ליצר ולפתח סריגים מורכבים מיוחדים לצרכי לייזרים חזקים בסיבים אופטיים.
הפרויקט הוא במסגרת המחקר שלנו אבל יכול גם להתכוון ליעדי מחקר משותף עם תעשיות בקונסורציום שמפתח לייזרי סיב מאוד חזקים.
is an umbrella term used to describe progressive lung diseases including emphysema, chronic bronchitis, refractory (non-reversible) asthma, and some forms of bronchiectasis. This disease is characterized by increasing breathlessness.
In this project we’ll develop a machine learning based method, using recording of breath sound (audio) from a mobile device in order to get a fast alarm of a coming seizure.
Registration plays an important role in 3D data processing. In the 3D domain, data is usually represented as a cloud of 3D points. The data of each such point cloud is given in its own coordinate system. The goal of registration is to find a transformation that optimally positions that data with respect to another point cloud. This is an important stage in solving many 3D problems such as 3D acquisition, where multiple views of an object need to be brought into a common coordinate system.
Sinkholes are very common in the Dead Sea area and are a cause of serious hazards as they appear very suddenly and can be very large. They mainly endanger those who travel along the road parallel to the Dead Sea (Road #90), the residents of the Ein Gedi Kibbutz and tourists in the southern part of the Dead Sea. It is therefore very important to accurately know the location of all sinkholes in the region.
In this project we will use a novel technique to locate the sinkholes. The technique is based on the measurement of the subsurface resistivity of the region being mapped. The resistivity depends on various geological parameters such as the mineral and fluid content, porosity and degree of water saturation in the rock. A complete resistivity map will allow accurate location of all sinkholes.
-- Experimental aspect is also included.--
The large size and very high cost of modern radiotherapy medical devices used for cancer treatment are mainly due to the dimensions of the electron accelerator. Acceleration techniques in use today require the electron accelerator to be about one meter in length in order to produce radiation with sufficient energy for radiotherapy.
The goal of the project is to develop a small, portable device that can be placed within a tumor site using standard endoscopic methods; allowing it to deliver the same radiation dose provided by current external beam technologies, without damaging surrounding tissue.
Using laser to accelerate electrons to very high energies, we will explore various acceleration structures and the dynamics of the electrons. Also, we assess the possibility of generation of stimulated X-ray by Compton scattering.