30/12/15 Ron Peleg

 

You are invited to attend a lecture

By

 

Ron Peleg

(Msc.  student under supervision of Prof. Raphael Kastner,

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

 

On the subject:

 

Side Lobe Level Reduction for a Thinned Array and Aperture Sharing

In many applications, real estate allocated for antennas is constrained. Several different systems may need to co-exist within the same area, implying overlapping apertures and interlacing arrays.

 

In this study, we aim to optimize the usage of area by thinning the elements in one array in order to free up area for a different array. Our optimization approach ensures control of side lobe level, and aspires to minimize gain loss.

An algorithm that implements the optimization is presented and includes the combination of several methods that were developed to handle the increased side lobe level of the massively thinned array. The method was tested for the 1-D and 2-D cases.

 

Using these techniques, several arrays can be placed in a given aperture and work simultaneously. Development of such multi-functional arrays is explored and implemented. The results confirm that this procedure can save substantial amount of elements from the aperture with well controlled side lobe level and sets the ground for new designs of antennas arrays.

 

 

Wednesday, 30 December 2015 at 12:00

Room 011, Kitot Bldg., Faculty of Engineering

30 בדצמבר 2015, 12:00 
Room 011, Kitot Bldg., Faculty of Engineering  
30/12/15 Ron Peleg

 

30/12/15

 

You are invited to attend a lecture

By

 

Tom Heller

 

 

M.Sc. student of Professor Eran Socher

Co-adviser: Professor Emanuel Cohen, Technion

Electrical Engineering, Physical Electronics Department

Tel Aviv University

 

 

An F-Band IQ Receiver Front-end in 28 nm CMOS

 

Abstract

 

This talk will discuss the design of a millimeter-wave F-band zero-IF receiver front-end for wireless chip-to-chip communication in 28-nm CMOS. The talk will focus on the design of the LNA chain and the quadrature splitter, but will also discuss the passive ring mixer and zero-IF buffers.

The LNA consists of a chain of four capacitively neutralized differential pairs. Dissipative losses, a major cause of noise figure degradation, are minimized by choosing appropriate LNA device dimensions and neutralization capacitors. The effect of capacitive neutralization on the minimum noise figure and noise sensitivity of a differential pair will also be discussed.

The quadrature splitter, which provides the I and Q carrier signals, can be tuned over a wide LO bandwidth. The splitter is implemented as a transformer-based directional coupler with an isolation port terminated by a tunable load impedance.

The receiver front-end consumes 18 mW from a 1 V supply and is driven by a -4 dBm F-band LO signal power. The zero-IF chain, used for testing purposes, consumes 33 mW from a 1.5 V supply.

 

 

Wednesday, 30 December 2015, at 9:30

Room 206, Wolfson Mechanical Engineering Building

 

30 בדצמבר 2015, 9:30 
Room 206, Wolfson Mechanical Engineering Building  
30/12/15

 

30/12/15 Yuval Dafna

 

You are invited to attend a lecture

By

 

Yuval Dafna

 

 

M.Sc. student of Professor Eran Socher

Co-adviser: Professor Emanuel Cohen, Technion

Electrical Engineering, Physical Electronics Department

Tel Aviv University

 

 

DIRECT DIGITAL TO D-BAND TRANSSMITTER FOR TENS OF Gb/s COMMUNICATION IN NANO-METER CMOS

 

Abstract

 

This talk will discuss the design of a millimeter-wave direct conversion transmitter for wireless chip-to-chip communication in 28-nm CMOS. The talk will focus on the design of a wide band 95-140GHz PA and the design of a full transmitter chain including: baseband buffer and active DAC-mixer design.

The PA consists of three differential pair’s stages. Low K transformers were used as matching networks to achieve high bandwidth while maintaining relatively low losses. Cross couple neutralization technique is used in order to maintain stability and increase gain. Stages layout was optimized in order to reduce parasitic losses and increase the PA gain.

The transmitter is a direct conversion transmitter with four baseband circuits driving DAC-Mixer and one PA stage. The transmitter is capable supporting up to 16-QAM modulation.

Both designs were taped out in 28nm CMOS technology and measured using millimeter wave probe station. The PA measured gain is above 10dB across 95-140GHz. The OP1dB is above 0dBm and the saturation power is 6dBm. The transmitter operates in the frequency regime of 100-130GHz, reaches up to -1dBm output power, consumes 84mW and achieves data rate of 27Gbps with QPSK modulation and without any digital calibrations (for example LO or IQ).  The Transmitter measured EVM with digital DPD and IQ imbalance is -23dB.

 

 

Wednesday, 30 December 2015, at 11:00

Room 206, Wolfson Mechanical Engineering Building

 

30 בדצמבר 2015, 11:00 
Room 206, Wolfson Mechanical Engineering Building  
30/12/15 Yuval Dafna

 

28/12/15

 

You are invited to attend a lecture

By

 

Samuel Jameson

 

Ph.D. student of Prof. Eran Socher

Electrical Engineering, Physical Electronics Department

Tel Aviv University

 

Millimeter-Wave and THz CMOS Radiating Transmitters

 

Abstract

CMOS was originally not considered suitable for RF-circuits however, the CMOS technology kept improving its operation frequency, enabling nowadays to reach THz frequencies with sufficient power that will allow us to create low-cost THz scanners. If efficient transmitters and receivers can be made efficient enough with high gain and compact antennas, THz scanner of the size of a hand or even less could be realized one day at low cost. Such devices could be then plugged to any smart-phone allowing extremely quick and regular scan of our body detecting and identifying diseases at the earliest stage for quicker prevention and treatment. On the road to the THz and its applications, systems based on millimeter-wave frequencies (24-300 GHz) are developed but mainly in targeting automotive systems (76.5 GHz) and in image screening for weapon detection (94 GHz). CMOS will be the key for mass-production for its low-cost and large scale integration properties that would allow integrating a whole system on-chip. In this research, we will focus on the development and improvement of CMOS MMW and THz transmitters and on their connection from the micro-scale environment to the human scale environment while conserving their state-of-the-art characteristics.

Thursday, December 28, 2015, at 16:00

Room 001, Kitot building

28 בדצמבר 2015, 16:00 
Room 001, Kitot building  
28/12/15

 

Elinor Knaani-Kashani 27/12/15

 

You are invited to attend a lecture

By

 

Elinor Knaani-Kashani

 

 

Ms.c. student of Doctor Eran Socher

Electrical Engineering, Physical Electronics Department

Tel Aviv University

 

 

Digital Closed Loop Design for Wideband Envelope Tracking Systems

 

Abstract

 

Power supply control schemes such as envelope tracking (ET) offer the greatest potential of high average efficiency operation for high peak-to-average power ratio (PAPR) signals. Envelope tracking systems are becoming a great opportunity since high PAPR systems are becoming very common (e.g. wifi, wimax, LTE, etc.), and power saving is needed. Naturally, in time, the need for throughput is increased, which implicates on the signal bandwidth, and traditional tracking systems become less efficient.

In this talk, we describe two topologies of digital closed loop envelope tracking systems for high bandwidth communication systems: A fast switching digital DC2DC based ET and a Multilevel ET based on multiple high BW digital low dropout (DLDO) voltage regulators. In both solutions, we introduce fast digital controllers, as well as digital mechanisms to overcome noise and voltage level transitions, which were not seen before in analog ET systems.

Both topologies were tested on a high-level simulation platform, examining the challenges each one presents, and showing the mechanisms to overcome them. We will show that although the existing highly non-linear PA in our system, both architectures provide the solution in terms of mask limitation, and error vector magnitude (EVM).

 

Sunday, 27 December 2015, at 12:30

Room 101, Software Building

 

27 בדצמבר 2015, 12:30 
Room 101, Software Building  
Elinor Knaani-Kashani 27/12/15

 

27/12/15 Assaf Azoulay

 

You are invited to attend a lecture

By

 

Assaf Azoulay

 

 

M.Sc. student of Dr. Eran Socher

Electrical Engineering, Physical Electronics Department

Tel Aviv University

 

 

Devices for Multiband Communication in MM-Waves

 

 

Abstract

 

Since the introduction of the iPhone, the demand for high data rate has increased drastically. More and more applications require wide bandwidth to allow for many bytes of digital transmission in a short period. Moreover, the demand for multifunction devices is inevitable whereas in today's phone, one can find many - LTE, GPS, Bluetooth and many more. In parallel to these increasing demands, mmW applications have been developed - Satcom, Wigig, automotive Rardar and others. Much of the RF research is focused on improving the performance of  existing bands on one hand and exploring new bands, such as 100-300GHz on the other hand.

 

This thesis presents the design of key components that enable the use of multiple applications and high data rates on the same chip. We start by introducing a tri-band VCO which oscillates at 3 different bands 30, 60 and 90 GHz. This device can also be used as an FSK by modulating the control voltage. The following device is the wideband antenna. The antenna design should enable the wideband demand in terms of good Return Loss and radiation pattern stability. The difficulties encountered with integrating the antenna with the RF chip by using interconnecting bonding wires is addressed and a matching circuit solution is shown. The last device this thesis explores is a combiner that will combine all bands into a single radiating element while maintaining low loss and compact size.

 

The VCO was designed on a 65nm CMOS technology from TSMC. The simulation results showed reasonable power across the band, ranging from -4dBm at 30GHz to -13dBm at 60GHz and -30dBm at 90GHz. The power was simulated into a single ended, 50 ohm load. The designed antenna was realized on a Rogers 5880 Duroid laminate, and we measured a peak gain of +6dBi at 120GHz at end-fire direction. The results were in good agreement with the EM simulations performed on CST.

The combiner was designed as a Triplexer on CMOS chip with the same technology as the VCO and was simulated to have average loss of 3dB across each sub-band.

 

 

Sunday, 27 November 2015, at 14:00

Room 101, Wolfson Building of Computer and Software Engineering

 

27 בדצמבר 2015, 14:00 
Room 101, Wolfson Building of Computer and Software Engineering  
27/12/15 Assaf Azoulay

 

 

24/12/15 Yehuda Meir

 

Yehuda Meir

 

 

Ph.D. student of Prof. Eli Jerby

Electrical Engineering, Physical Electronics Department

Tel Aviv University

 

 

Localized Microwave Heating (LMH) and its Applications in Solids, Powders, and Plasmas

 

Abstract

Various LMH effects are presented in a paradigmatic approach. The LMH is intentionally induced, as in the microwave drill, by a near-field applicator to form a small hotspot. The thermal-runaway instability rapidly leads to a phase transition, to liquid, gas or plasma. Here we show that LMH can also be induced by low microwave power (~10 W) generated by a solid-state device such as LDMOS-FET. The LMH applicability is demonstrated for the ignition of thermite powders (mixtures of iron oxides and aluminum powders) with high ignition temperature (~1,500°C). Using the bubble-marble (BM) effect discovered in this study, thermite combustion also becomes feasible underwater. The hydrophobic powder is inserted as a confined BM into water, by an external magnetic field, and its combustion generates significant gas pressure underwater. Constructive LMH applications such as additive manufacturing of solid object from metal powders are presented. The inverse process of nano-powder production from solids is demonstrated as well by the LMH dusty plasma. The potentials and limitations of various LMH applications are discussed.

 

Thursday, December 24, 2015, at 15:00

Room 011, Kitot building

 

24 בדצמבר 2015, 15:00 
 
24/12/15 Yehuda Meir

 

 

 

24/12/16 Doron Bar Lev

 

Doron Bar-Lev

(PhD student under the supervision of Prof. Jacob Scheuer)

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

Plasmonic Metasurfaces Applications for Manipulation of Optical Radiation

 

Surface plasmons modes are characterized by a propagating or a localized behavior, leading to a variety of unique phenomena, particularly enhanced near field intensities at sub-wavelength dimensions. The intense and localized field can be harnessed to increase the sensitivity and performance of numerous phenomena and applications. Moreover, introduction of carefully designed sub-wavelength patterns to the surface, form a plasmonic metasurface that can shape the amplitude, phase and polarization of the electromagnetic (EM) field in the vicinity of the structure, thus offering a new paradigm for manipulation and control of optical radiation.

In this talk the unique nature of plasmonic waves and metausrfaces will be discussed leading to two new applications: 1) extremely efficient laser-driven particle accelerators and 2) generation and dynamic control of arbitrary plasmonic wave-fronts using simple and static plasmonic surfaces. The first application demonstrates the opportunity of using a plasmonic metasurface to tailor the EM field in its vicinity, particularly for particle beam manipulation. I will present general design rules for plasmonic metasurface laser-driven accelerators (MLAs) and a specific structure for efficient acceleration of relativistic particles. The presented MLA attains a huge acceleration gradient of 11.6GV/m at 16fsec operation. This value is almost four times larger than that of contemporary short pulse laser accelerators and two orders of magnitude larger than that of contemporary radio-frequency based accelerators. The second application will refer to the reciprocal process in which manipulations on the surrounding EM field affects the generated plasmonic waves. Here, I will show that by tailoring the exciting beam incident on a simple and static one-dimensional metallic grating, efficient generation and dynamic control of practically every possible arbitrary plasmonic wave-front is achievable. A complete and rigorous theory will be presented and demonstrated, leading to a general design formalism for generating arbitrary plasmonic wave fronts. Finally, using this formalism I will demonstrate excitation of optically controlled surface plasmon hotspots that can lead to exciting new applications as dark field plasmonic microscopy and surface optical tweezers.

Thursday, December 24, 2015, at 10:00

Room 011, Kitot building

 

24 בדצמבר 2015, 10:00 
011 Kitot  
24/12/16 Doron Bar Lev

 

סמינר מחלקתי - גייל גלבוע-פרידמן

Characterization of Privacy Loss

19 בינואר 2016, 14:00 
וולפסון 206  

Abstract:

How much privacy is lost during a process? This question is not a new question. However, there is no unequivocal answer in the scientific literature. Our research aims to formalize the conceptual notion of privacy.

In this talk, I’ll present two results:

  1. Construction of a preference relation model over privacy jeopardizing processes. We list 4 natural properties (axioms) to describe such relation. We prove a Characterization Theorem stating that any preference relation model which satisfies these axioms is represented by f-divergence. We conclude that  f-divergence can serve as a natural measure of privacy-loss.
  2. Characterization of Differential Privacy, which is an ad-hoc standard in computer-science literature. We choose 5 axioms for the purpose of justifying an existing measure. Some of the axioms are not natural in our eyes. Thus, Differential Privacy remains unjustified, at least until another construction - which is based on natural axioms - is demonstrated.

 

Our study leads to a recommendation on measuring privacy loss by f-divergence functions, such as KL-divergence or Hellinger-distance. The applicability of our research is for having a rigorous methodology of prioritizing procedures by their level of privacy-loss.

 

*Joint work with Prof. Rann Smorodinsky and Prof. Kobbi Nissim

 

Bio :

Gail Gilbao-Freedman received her BA degree in Mathematics and Computer Science (cum laude) from the Technion in 2001 and her M.Sc. in Applied Mathematics from the Technion in 2005. She received her Ph.D. from the department of Statistics and Operations Research in the school of Mathematical Sciences in Tel Aviv University in 2011.

Over the years she has worked for Elbit Systems, IBM Research Lab, and the Innovation Center of Citi Group. In 2014-15, she was a Postdoctoral Research Fellow at the faculty of Industrial Engineering and Management in the Technion. She is currently a Postdoctoral Research Fellow at the school of Computer Science in Tel Aviv University and the department of Electrical Engineering in Columbia University.

סמינר מחלקתי - עופר משי

Scalable Machine Learning for structured high-dimensional outputs

07 בינואר 2016, 12:00 
אודיטוריום 020  

Abstract:

In recent years, machine learning has emerged as an important and influential discipline in computer science 

and engineering. Modern applications of machine learning involve reasoning about complex objects like 

images, videos, and large documents. Treatment of such high-dimensional data requires the development of 

new tools, since traditional methods in machine learning no longer apply. In this talk I will present two recent 

works in this direction. The first work introduces a family of novel and efficient methods for inference and 

learning in structured output spaces. This framework is based on applying principles from convex optimization 

while exploiting the special structure of these problems to obtain efficient algorithms. The second work 

studies the success of a certain type of approximate inference methods based on linear programming 

relaxations. In particular, it has been observed that such relaxations are often tight in real applications, and I 

will present a theoretical explanation for this interesting phenomenon.

Bio :

Ofer Meshi is a Research Assistant Professor at the Toyota Technological Institute at Chicago. Prior to that he 

obtained his Ph.D. and M.Sc. in Computer Science from the Hebrew University of Jerusalem. His B.Sc. in 

Computer Science and Philosophy is from Tel Aviv University. Ofer’s research focuses on machine learning, 

with an emphasis on efficient optimization methods for inference and learning with high-dimensional 

structured outputs. During his doctoral studies Ofer was a recipient of Google's European Fellowship in 

Machine Learning.

עמודים

אוניברסיטת תל אביב עושה כל מאמץ לכבד זכויות יוצרים. אם בבעלותך זכויות יוצרים בתכנים שנמצאים פה ו/או השימוש שנעשה בתכנים אלה לדעתך מפר זכויות
שנעשה בתכנים אלה לדעתך מפר זכויות נא לפנות בהקדם לכתובת שכאן >>