You are invited to attend a lecture

Lightning-fast solution of scattering problems in nanophotonics: an effortless modal approach

:By

Parry Yu Chen

Unit of electro-optics Engineering, Ben-Gurion University

Abstract

Nanophotonic structures are capable of generating field hotspots, which can enhance quantum light-matter interactions by many orders of magnitude. However, numerical simulations for applications such as radiative heat transfer, electron energy loss spectroscopy, van der Waals forces, Purcell factor throughout a volume, and many others are challenging and often computationally prohibitive. Common to these simulations is that the Green’s function or local photonic density of states must be known at each point across a volume of space, necessitating the solution of Maxwell’s equations perhaps many thousands of times.

We propose a modal solution, which requires just a single simulation to find the modes of the nanophotonic system, from which we immediately obtain the Green’s function everywhere in space. This not only reduces simulation time by approximately 2 orders of magnitude, but also offers ready physical insight into the spatial variation of Green’s function. Modal methods have long been used for closed systems, where the formulation is exceedingly simple. We have generalized modal methods to open systems while maintaining this simplicity, catering to the explosion of research interest in nanophotonics. We furthermore present a highly-efficient exponentially-convergent method of generating the modes themselves

On Thursday, November 29, 2018, 15:00

Room 011, Kitot building

School of Mechanical Engineering Seminar
Wednesday, December 05, 2018 at 14.00
Wolfson Building of Mechanical Engineering, Room 206

The Parametric HFGMC Micromechanics for Nonlinear and Damage Modeling of Multiphase Materials

Aviad Levi Sasson

 Ph.D. student of Prof. Rami Haj-Ali and Prof. Jacob Aboudi

This study expands the parametric HFGMC capabilities to model and solve engineering problems. New condensed formulation is proposed and proved to yield good results for the effective thermo-mechanical properties and the spatial elastic fields in terms of computing time. For the first time, the parametric HFGMC is extended to predict the effective thermal properties of composite material: coefficients of thermal expansion, effective thermal conductivity and heat capacity at constant pressure and constant volume.

A continuum damage model has been integrated with the parametric HFGMC. This damage model was originated from the principal of energy dissipation proposed by Marigo (1981) model and was coupled with Ledeveze-Lemaitre (1984) model. The Marigo damage model was revised in the current work to eliminate nonphysical behavior under shear loading conditions. The corrected damage model is implemented at the subcell level so each subcell can detect failure due to its local stress state. Material softening behavior due to damage is implemented as well. The isotropic damage law and its evolution generates both an effective anisotropic damage behavior in the global (composite) level and global nonlinear stress-strain response. Several material systems were used to demonstrate the new damage and failure prediction capabilities of the parametric HFGMC. The obtained failure envelops from the parametric HFGMC were compared with experimental results and showed very good agreement. For the first time, micromechanical based failure envelops are matched with new analytical-mathematical expressions for future design and analysis and to reduced computational effort during analyses.

Multiscale analysis of laminated composite structures is proposed by integrating the parametric HFGMC with the classical FE. The multiscale analysis is compared with the experimental results for the cases of notched laminated composite coupons that are subjected to tension and compression loading. In additions, the parametric HFGMC is used to model the mechanical behavior of a Dyneema based soft composite. The new capacities of the parametric HFGMC are shown again to succeed with the prediction of stress-strain curves of soft composite. New experimental tests for the soft composite are designed and proposed in the framework of this study in order to compare the periodic HFGMC prediction with some measurements. The parametric HFGMC shows good agreement with the measured elastic properties.  This talk will conclude by discussing potential future extensions of the proposed micromechanical framework.

Speaker: Moran Rubin

M.Sc. student under the supervision of Prof. Natan T. Shaked

Wednesday, November 28^th 2018 at 15:30

Room 011, Kitot Bldg., Faculty of Engineering

TOP-GAN: Label-Free Cancer Cell Classification Using Deep Learning with a Small Training Set

Abstract

We propose a deep learning approach for medical imaging that copes with the problem of a small labeled training set, the main bottleneck of deep learning, and apply it for classification of healthy and cancer cells acquired by quantitative phase imaging. The proposed method is hybridization between transfer learning and generative adversarial networks (GANs). Healthy cells and cancer cells of different metastatic potential have been imaged by low-coherence off-axis holography. After the acquisition, the optical path delay maps of the cells have been extracted and directly used as an input to the deep networks. In order to cope with the small number of classified images, we have used the GAN setup to train a large number of unclassified images from another cell type (sperm cells). After this preliminary training, and after transforming the last layers of the network with new ones, we have designed an automatic classifier that copes with a small training set and classified the correct cell type (healthy/primary cancer/metastatic cancer) with 90-99% accuracy. We believe that our approach makes the combination of holographic microscopy and deep learning networks more accessible to the medical field by enabling a rapid, automatic and accurate classification in stain-free imaging flow cytometry.

Speaker:  Marina Haikin

M.Sc. student under the supervision of Prof. Ram Zamir and Dr. Matan Gavish

Wednesday, December 5^th, 2018 at 15:00

Room 011, Kitot Bldg., Faculty of Engineering

Analog Coding Frame-work
 

Abstract

            Analog coding is a low-complexity method to combat erasures, based on linear redundancy in the signal space domain. Previous work examined "band-limited discrete Fourier transform (DFT)" codes for Gaussian channels with erasures or impulses. We extend this concept to source coding with "erasure side-information" at the encoder and show that the performance of band-limited DFT can be significantly improved using irregular spectrum, and more generally, using equiangular tight frames (ETF).

            Frames are overcomplete bases and are widely used in mathematics, computer science, engineering, and statistics since they provide a stable and robust decomposition. Design of frames with favorable properties of random subframes is motivated in variety of applications, including code-division multiple access (CDMA), compressed sensing and analog coding.     

            We present a novel relation between deterministic frames and random matrix theory. We show empirically that the MANOVA ensemble offers a universal description of the spectra of randomly selected subframes with constant aspect ratios, taken from deterministic near-ETFs. Moreover, we derive an analytic framework and bring a formal validation for some of the empirical results, specifically that the asymptotic form for the moments of high orders of subsets of ETF agree with that of MANOVA.

            Finally, when exploring over-complete bases, the Welch bound is a lower bound on the root mean square cross correlation between vectors. We extend the Welch bound to an erasure setting, in which a reduced frame, composed of a random subset of Bernoulli selected vectors, is of interest. The lower bound involves moment of the reduced frame, and it is tight for ETFs and asymptotically coincides with the MANOVA moments. This result offers a novel perspective on the superiority of ETFs over other frames.