הפקולטה להנדסה ע"ש איבי ואלדר פליישמן

16 מרץ 2015

ידיעות נוספות בנושא

EE Seminar: Practical-Optimal design of grid connected PV system

~~Speaker: Avi Aronescu
M.Sc. student under the supervision of Prof. Appelbaum Joseph

Wednesday, March 25th, 2015 at 15:00
Room 011, Kitot Bldg., Faculty of Engineering

Practical-Optimal design of grid connected PV system
Abstract
As the world's energy demand increases, more and more are turning to renewable energy sources. On this work we will focus on a practical approach of designing a grid connected PV system. Any grid connected PV system is based upon 2 fundamental building blocks. The first is of course the PV panel (or module). The Second is the Inverter, which converts DC energy to AC energy and injects it to the power grid. For practical design one must take into account the PV panel and inverter specs such as voltages, power, etc. The design must also take into account the different losses such as mutual shading, cable power losses, dust, etc.
Due to the mutual shading between collector rows in contrast to the goal of producing a required amount of energy from a given field size or least cost per unit energy we may define optimization problems. For example, what is the maximum output energy of a PV system from a given area or what is the minimum cost of the PV system? These problems depend on the PV systems practical design parameters. The final purpose is to present a practical optimal design of a grid connected PV system.

25 במרץ 2015, 15:00

חדר 011, בניין כיתות-חשמל

EE Seminar: A VC-dimension-based Outer Bound for the Zero-Error Capacity of the Binary Adder Channel

~~(The talk will be given in English)

Speaker: Or Ordentlich
Institute for Problems in Mechanics, Moscow
Monday, March 23rd, 2015
15:00 - 16:00
Room 011, Kitot Bldg., Faculty of Engineering
A VC-dimension-based Outer Bound for the Zero-Error Capacity of the Binary Adder Channel
Abstract
The Binary Adder Channel (BAC) is a two-user multiple access channel whose inputs are binary and whose output is the real sum of the inputs. While the Shannon capacity region of this channel is well known, little is known regarding its zero-error capacity region, and a large gap remains between the best inner and outer bounds. In this work, we provide a new outer bound for this problem, improving the best known result by Urbanke and Li.
Our result is obtained via a confluence of combinatorial and information-theoretical arguments: Given any zero-error coding scheme for the BAC, we first construct another lower-dimensional coding scheme for the BAC with correlated messages. We then obtain a single letter outer bound for the Shannon capacity region of this more general setting, which translates back to an outer bound on the zero-error capacity of the BAC. Our construction is facilitated by introducing the notion of a "soft" VC-dimension of a codebook, which we lower bound by establishing a suitable variation of the Sauer-Perles-Shelah Lemma.

Joint work with Ofer Shayevitz.

23 במרץ 2015, 15:00

חדר 011, בניין כיתות-חשמל

EE Seminar: The Kalman Filter and Quantization

~~(The talk will be given in English)

Speaker: Prof. Alexander Ovseevich
Institute for Problems in Mechanics, Moscow
Wednesday, March 18th, 2015
16:00 - 17:00
Room 011, Kitot Bldg., Faculty of Engineering
The Kalman Filter and Quantization
Abstract
We show that the filtering of diffusion processes is intimately related with classical and quantum mechanics. We associate a classical Hamiltonian system with a linear or nonlinear filtering problem, so that the problem is similar to that of quantization of
the Hamiltonian system. In particular, the classical linear Kalman--Bucy filter turns out to be equivalent to the Heisenberg equations for a quantized linear Hamiltonian system. As a byproduct we obtain formulas for regularized determinants of Sturm-Liouville operators, as well as the Mehler formula for the fundamental solution of the Schrodinger operator for a harmonic oscillator.

18 במרץ 2015, 16:00

חדר 011, בניין כיתות-חשמל

סמינר המחלקה להנדסה ביו רפואית

ליאור טריימן

תלמידת המחלקה להנדסה ביו רפואית לתואר שני תרצה בנושא:

A computational study about the effect of oscillations in translation factors on translation rate

Gene expression is a central cellular process by which the protein coding potential inscribed in the DNA is converted into proteins. Translation, is an important step of gene expression. It is the process by which messenger RNA (mRNA) is decoded by a ribosome to produce a specific amino acid chain – protein. This fundamental process is related to all biomedical disciplines. Specifically, it may be engineered for various biotechnological applications. Oscillations and clocks are important components of many hardware systems, suggesting that they are required also in synthetic biology. As timing is a key factor in most engineered systems and also driving biological processes, this work focuses on the effect of oscillations on gene translation.

Effect of oscillations in tRNA pool levels, as well as oscillations in codon translation rates in different locations in the gene strand, were studied using the Totally Asymmetric Simple Exclusion Process (TASEP) as implemented in Matlab. This computational model takes into consideration the interactions between ribosomes and their size, concentration of tRNA molecules, efficiency of their interaction with codons, and the stochastic nature of translation.

This is the first work done exploring oscillation in translation factors using simulations in a computational model that takes into consideration the bio-physical nature of translation process, it is a genome scale analysis using real data parameters including ribosome size, typical decoding times and real genes from the yeast genome.

Results suggest oscillating translation factors results in a periodic translation rate. A strong negative Spearman correlation (P<10^-87) exists between gene length and gene's translation rate amplitude. Moreover, the change of -10% in gene length has a stronger effect on translation rate amplitude (+34%) than in long genes (+9%). Furthermore, there is a strong positive Spearman correlation (P<10^-87) between the oscillated codon efficiency and genomic mean amplitude of translation rate, such that a change in +20% of oscillated codon velocity results in +40% increase and translation rate amplitude. In total about 200% change in translation rate amplitude is controlled by the oscillated codon velocity. Another connection was found between the relative locations of oscillation in the gene. When oscillating codons in the middle of the genes ORF it has a ~15% higher genomic mean amplitude comparing with oscillating codons on the edge (upstream) of the gene.

העבודה נעשתה בהנחיית דר' תמיר טולר, המחלקה להנדסה ביו-רפואית,

אוניברסיטת תל-אביב

ההרצאה תתקיים ביום ראשון 15.03.15, בשעה 14:15,

בחדר 315, הבניין הרב תחומי, אוניברסיטת תל אביב

15 במרץ 2015, 14:15

סמינר

סמינר מחלקתי Miri Siton

15 ביוני 2015, 15:00

וולפסון 206

School of Mechanical Engineering Seminar
Monday, June 15, 2015 at 15:00
Wolfson Building of Mechanical Engineering, Room 206

גרים בהנדסת מערכואתת חוצת ארגונים

Miri Siton

עידן המידע מייצר הזדמנויות ואיומים חדשים לארגונים עתירי טכנולוגיה. התמודדות עם סוגיות הנוגעות בחוסר מידע לתמיכה בקבלת החלטות חוצות ארגון ובכשלים מערכתיים עשויה לקבל מענה באמצעות תהליכי הנדסת מערכות ברמת הארגון.

הנדסת מערכת מהוה נדבך משמעותי בתהליך הפיתוח של מערכות החל משלב הגדרת הדרישות, דרך אפיון פונקציונאלי, אפיון טכני, פיתוח ובדיקות. בארגונים רבים חל תהליך של מיכון תהליכים ידניים באמצעות מערכות ממוכנות. בעשור האחרון, עם הכניסה לעידן המידע, הפך המידע לנכס בעל ערך רב לארגון והחלו להתפתח רשתות ליצירת קישוריות בין מערכות המשרתות עולמות תוכן משותפים. תהליכים אלה של קישוריות היוו צעד ראשון בהבנה של ארגונים כי שיתוף מידע יכול לייצר ערך מוסף שלא היה קיים תחת פעילות עצמאית של כל מערכת ונבע מהקישוריות. כיום, יותר ויותר ארגונים, מבינים כי נדרש לייצר הסתכלות רוחבית כוללת על כלל המערכות והמידע בארגון, כך שניתן יהיה להגדיל את הערך המוסף לארגון. המחקר המוצע מתבסס על ידע שנצבר בשני ארגונים גדולים ליצירת מתודולוגיה להנדסת מערכת רוחבית בארגון.

מטרת המחקר: הגדרת מתודולוגיה להנדסת מערכות חוצת ארגון שתסייע בשיפור תהליכים חוצי ארגון.

המחקר כלל הגדרת מסגרת הנדסית מוכוונת לשיפור תהליכים חוצים תוך בחינת ישימותה בשני ארגונים גדולים עתירי טכנולוגיה בתחום הביטחון והתחבורה. המסגרת ההנדסית כוללת הגדרת תהליכי הנדסת מערכת למיפוי התהליכים האופרטיביים בארגון, הגדרת ארכיטקטורה אופרטיבית, הגדרת ארכיטקטורת מערכות ובניית מפת דרכים למימושה. במסגרת המחקר נאספו נתונים השוואתיים טרום הטמעת המתודולוגיה ולאחר הטמעתה לבחינת המדדים הבאים: שינוי בבניית תמונת מצב, שינוי במודעות המצב, שינוי בתהליכי קבלת החלטות ושינוי בביצועים (Performance) של תהליכים חוצי ארגון. הנתונים שנאספו נותחו והתוצאות מצביעות על שיפור משמעותי בכל המדדים.

EE Seminar: Optimal Adaptive Transmit Beamforming for Target Tracking in Cognitive MIMO Radar/Sonar

~~Speaker: Nathan Sharaga,
M.Sc. student under the supervision of Prof. Joseph Tabrikian and Hagit Messer

Wednesday, March 18th, 2015 at 15:00
Room 011, Kitot Bldg., Faculty of Engineering

Optimal Adaptive Transmit Beamforming for Target Tracking in Cognitive MIMO Radar/Sonar

Abstract
In this work, a new adaptive beamforming method for target tracking by cognitive multiple-input multiple-output (MIMO) radar or sonar is proposed. In this method, at each step, the transmit beampattern is sequentially determined based on history observations. The conditional Bayesian Cramér-Rao bound (BCRB) for one-step prediction of the state-vector in target tracking problem was used as the optimization criterion for beampattern design. The conditional BCRB for one-step prediction of the state vector is recursively computed at each step. The conditional BCRB is computationally expensive, therefore a low-complexity method for computing an approximation of the conditional BCRB is proposed. The proposed method is applied to the problem of target tracking in a shallow underwater environment in the presence of environmental uncertainties. It is shown that the method is able to automatically focus the transmit beampattern toward the target direction within a few steps at very low signal-to-noise ratios (SNR's). The method exhibits much better performance in terms of localization error compared to other methods, such as orthogonal (omni-directional) transmission.

18 במרץ 2015, 15:00

חדר 011, בניין כיתות-חשמל

EE Seminar: A Spectral Perspective on Shape Analysis

~~(The talk will be given in English)

Speaker: Prof. Ron Kimmel
Computer Science Department, Technion
Monday, March 16th, 2015
15:00 - 16:00
Room 011, Kitot Bldg., Faculty of Engineering
A Spectral Perspective on Shape Analysis
Abstract
We explore the power of the Laplace Beltrami Operator (LBO) in processing and analyzing visual and geometric information. The decomposition of the LBO at one end, and the heat operator at the other end provide us with efficient tools for dealing with images and shapes. Denoising, segmenting, filtering, exaggerating are just few of the problems for which the LBO provides a solution. We will review the optimality of a truncated basis provided by the LBO, and a selection of relevant metrics by which such optimal bases are constructed. Specific example is the scale invariant metric for surfaces that we argue to be a natural selection for the study of articulated objects.

16 במרץ 2015, 15:00

חדר 011, בניין כיתות-חשמל

Lior Yaron 12.3.15

12 במרץ 2015, 12:00

Kitot 011

You are invited to attend a lecture

Lior Yaron

(PhD. student under the supervision of Prof. Moshe Tur)

School of Electrical Engineering, Tel-Aviv University, Tel-Aviv 69978, Israel

Photonic True Time Delay

Over the last fifty years, the engineering community recognized the need for True Time Delay (TTD) in phased array antennas. However, the subject has become increasingly important in the last decade, as radar systems were required to achieve higher resolution, farther distances and wider scan angles. One of the preferred RF solutions is the Rotman lens, which uses RF guided waves in a specially designed geometrical structure to produce these delays for a number of discrete beams. However, the need for smaller volume and lower weight, as well as for still wider bandwidth, has made this option less attractive. Alternatively, TTD may be implemented digitally, using fast (many Giga-Samples/sec) Analog to Digital converters on every element. This power-hungry technology, though, has not yet reached maturity but may become a viable solution in the future. This leaves photonic technology as a mid-term feasible option.

In this seminar we will discuss the need for TTD in modern radar systems and the possibility to implement it using a photonic sub-system. We will present a photonic delays network for antenna array receiver employing multiple scanning beams and discuss the main design considerations. We will also present measurements results of a single TTD module with 4 independent beams and 8 different delays showing extremely smooth RF response in the X-band, fast switching capabilities and negligible crosstalk.

In the second part of this seminar we will discuss Brillouin Dynamic Gratings (BDGs), which have been recently suggested to realize tunable photonic delay elements with large delay-bandwidth product. A fundamental feature of Brillouin scattering that might present a limitation is the inevitable presence of spontaneous Brillouin scattering, generated by the scattering of the optical signal from thermally excited acoustic waves. In this talk we will show theoretically and experimentally how the undesired scattering of the signal can be considerably enhanced in BDG through interaction with the scattering of the orthogonally polarized pump which writes the grating. For a strong enough pump and even moderate signal power levels, the observed scattering is dominated by the pump, exhibiting an average power more than 30 dB stronger than that of the signal alone.

Thursday, March 12, 2015, at 12:00

Room 011, EE- KITOT building

12.3.15

12 במרץ 2015, 16:00

Kitot 011

You are invited to attend a lecture

Moty Roudstein

(M.Sc. student under the supervision of Prof. Amir Boag)

School of Electrical Engineering, Tel-Aviv University, Tel-Aviv 69978, Israel

Multilevel Physical Optics Algorithm for

Near-Field Double-Bounce Scattering

Many radar and sonar applications require evaluation of the field scattered from complex objects. Usually, the scattered field is of interest for various angles of interest at several frequencies within a certain band. The range of distances to the target is often such that, for the desired frequencies, it is within the target's near field. At high frequencies, where the physical Optics (PO) approximation is valid, the scattered field can be divided into a series of separate contributions, depending on the number of bounces. For concave geometries, accurate computation of the scattered field requires, taking into account at least the second order term in the series, namely the "double-bounce" (DB) contribution, i.e., in addition to the "single-bounce" (SB) contribution. In terms of the computational complexity (CC), this DB contribution is, often, significantly more difficult to compute than the easier SB contribution. This high CC impedes the calculation of DB contributions, and calls for a faster method.

In this talk, we present a fast algorithm for the evaluation of the DB contributions to the PO scattering integrals, over a range of aspect angles and frequencies. The work extends the preceding far-field algorithm, to encompass three-dimensional and near-field scenarios. The algorithm relies on the observation that partial contributions to the backscattered patterns, due to pairs of subsurfaces of a finite size, are essentially band-limited functions of the aspect angles, distance, and frequency. Multiplying these contributions by appropriate common phase factors allows for their sampling over a coarse grid of source/observation points at rates which are dictated by the subsurfaces' linear dimensions. Savings are achieved by directly evaluating the partial contributions due to very small subdomains on coarse grids at a very low cost, and then gradually interpolating and aggregating the contributions to obtain the backscattered field due to large subdomains, in a hierarchical fashion. The presented DB Multi-Level Physical Optics (MLPO) method achieves a significant reduction in the CC.

Thursday, March 12, 2015, at 16:00

Room 011, Kitot building

ברכות לדר' יצחק תמו

ידיעות נוספות בנושא

עמודים