PhD thesis defence Yasser Sherazi
Date: 2014-01-17, at 10:15
Place: E: 1406, E building, Lund Institute of Technology
Opponent: Associate Professor Dejan Markovic. University of California, Los Angeles (UCLS), Electrical Engineering
Design Space Exploration of Digital Circuits for Ultra-low Energy Dissipation
ISSN 1654-790X, No 54, Printed in Sweden by Tryckeriet i E-huset, Lund University, Lund., 2013.
Abstract: The ever expanding market of ultra portable electronic products is compelling the designer to invest major efforts in the development
of small and low energy electronic devices. The driving force and benefactors of such devices are (but not limited to) e-health
system, sensor network applications, security systems, environmental applications, and home automation systems. These markets
have launched a massive trend towards ultra low-energy and ultra low-voltage devices. As the technology scales, the dimensions
of a transistors have become extremely small, leading to reliability and process variation issues. Above all, with the ability of
placing millions of gates in a small area, high current consumption have become one of the key factors in modern high-performance
technologies. In portable electronics, the battery life time is a major issue, as most of the time the device is accompanied with an
enclosed battery that has to last for long periods without compromise on performance. Furthermore, there are many applications
where the battery lifetime sets the lifetime of the device. Therefore, research is needed to identify the techniques and the impact of
them on the design operated for ultra low-energy. The low energy dissipation requirements on a design are achievable by employing
various optimization techniques. Voltage scaling is the most effective knob to reduce energy dissipation. For this reason ultra-low
energy design usually translates into ultra-low voltage or subthreshold (sub-VT) domain operation. This work presents an analysis
on design space for ultra-low energy dissipation of digital circuits. The circuits are operated in the sub-VT region with moderate
throughput constraints. The drawback of operating circuits in sub-VT is slow speed performances and reduced reliability. To
combat speed degradation due to scaling of the supply voltage, the architectural design space, needs exploration. Techniques such as
device sizing, body biasing, stacking transistors, dual threshold gates, multi threshold synthesis, pipelining, and loop unfolding, are
explored and applied to the designs. The designs are synthesized in a 65 nm CMOS technology with low-power and three threshold
options, both as single-VT and as multi-VT designs. A sub-VT energy model is applied to characterize the designs in the sub-
VT domain. Reliability in the sub-VT domain is analyzed by Monte-Carlo simulations. The minimum reliable operation voltage
(ROV) for gates in low power 65 nm CMOS technology is found to be around 250 mV. The applied energy model for designs to
be characterized for sub-VT domain operation is presented. The energy model encompasses single VT implementations and multi-
VT implementations. The energy modeling is based on the 65 nm CMOS standard cells provided by the technology vendor. The
energy model has been used to evaluate various techniques and constraints for a circuits operated in the sub-VT domain. The work
describes how the energy dissipation of architectures vary w.r.t. switching activity, e. The effects of pipelining together with supply
voltage scaling is analyzed, which shows that they have high benefits with respect to energy dissipation. Various half-band digital
(HBD) filter structures are evaluated for minimum energy dissipation in the sub-VT domain for a throughput constrained system. All
architectures, i.e., unfolded and the basic HBD filter, are implemented and simulated using 65 nm Low-Power High-Threshold (HVT)
standard cells. The application of a sub-VT energy model reveals that it is beneficial to use an unfolded implementation to achieve
low energy dissipation per sample at EMV, when compared to the energy dissipated by a basic simplified HBD filter implementation.
Various available threshold options are analyzed with the help of filter structures by using 65 nm Low-Leakage High-Threshold
(HVT), Standard-Threshold (SVT) and Low-Threshold (LVT) standard cells. Secondly, the design space is increased by utilization
of a combination of HVT + SVT and also HVT + LVT cells. The analysis with sub-VT energy model leads to the conclusion
that a suitable design is a synergy between parallelism, and utilization of various threshold options. In this analysis the multi-VT,
implementations did not show a major advantage over single VT implementations. A decimation filter chain consisting of 4 HBD
filters is fabricated and the silicon measurements demonstrate that SVT and different architectural flavors are suitable for a ultra
low energy (ULE) implementation. Silicon measurements prove functionality down to a supply at 350 mV, with a maximum clock
frequency of 500 kHz, having an energy dissipation of 102 fJ/cycle. Additionally, an alternative to SRAM macro is presented for
sub-VT operations. The memory is based on standard-cells and is referred to as SCMs. The energy per memory access as well as the
maximum achievable throughput in the sub-VT domain of various SCM architectures are evaluated by means of a gate-level sub-VT
energy characterization model.
2013-12-20 Ex-jobbspresentation: Olof Malmodin
Olof Malmodin will present his Master thesis work on Dec. 20th (Friday) at 13:15 in E:2311.
His thesis title is: "Design of a 60GHz Low-Noise Amplifier in SiGe Technology".
This thesis was done at Ericsson in Göteborg.
A low noise amplifier is designed for future applications in the 60GHz band, using an existing SiGe technology, BiCMOS8HP from IBM. Different topologies are analyzed and compared. Two different schematics of single ended three stage designs are compared. A differential four stage CE topology is designed and simulated with parasitic extraction. The final design shows a noise figure of less than 6 dB, gain of more than 29dB and input/output return loss of less than -13dB for the entire band. Results show a low-noise amplifier in BiCMOS technology is feasible, but issues such as unconditional stability, better linearity and more efficient biasing would have to be solved.
2013-12-18 Ex-jobbspresentation: Alexander Aulin Söderqvist och Niklas Claesson
Alexander Aulin Söderqvist and Niklas Claesson will present their Master thesis on Wednesday Dec 18th at 10:15 in E:3139
Title: A Timing System Application using White Rabbit
In their thesis they are evaluating a new system that is a candidate for ESS, a short demo will be presented.
The thesis was done at Cosylab in Slovenia
Two synchronization layers for timing systems in large experimental physics control systems were studied. White Rabbit (WR), which is an emerging standard, is compared against the well-established event-based approach. Several typical timing system services has been implemented on an FPGA using WR to explore its concepts and architecture, which is fundamentally different from an eventbased one. The requirements for the implemented prototype were decided based on typical requirements of current accelerator projects and with regard to other parameters, such as scalability and commercial availability. The proposed design methodology and prototype demonstrate one way of deploying WR in future accelerator projects.
2013-12-19 Ex-jobbspresentation: Stefan Granlund
Stefan Granlund will present his Master thesis work on Dec. 19th (Thursday) at 11:15 in E:2311.
His thesis title is: "Implementation of a Highly-Parallel Soft-Output MIMO Detector with Fast Node Enumeration".
This report presents a low latency, high throughput soft-output signal detector for a 4×4 64-QAM spatial-multiplexing MIMO system. To achieve high data-level parallelism and accurate soft information, the detector adopts a node perturbation technique to generate a list of candidate vectors around an initial estimation acquired by using the Zero Forcing detection. The initial estimation result is extended to the closest neighboring nodes to form a list of candidate vectors. A fast and hardware friendly enumeration scheme is developed to significantly reduce processing delay of the node extension. The Fast Node Enumeration exploits the symmetric geometric properties of the QAM constellation. The Euclidean distance of the candidate vectors is calculated and used to produce the soft output. The detector achieves a BER of 10^−4 at the SNR point of 13.5 dB. Compared to the K-best detector the number of visited nodes is reduced by 34 times. The detector was implemented in VHDL and synthesized using Synopsys Design Compiler with a 65nm CMOS standard cell library. The detector occupies a 0.58mm^2 core area with 290K gates. The peak throughput is 3Gb/s at 500 MHz clock frequency with a latency of 20ns. Energy consumption per detected bit is 33pJ.
2013-12-17 Ex-jobbspresentation: Hayder Al-Zubaidi och Baydai Abdulameer
On Tue, 17 Dec 2013, Hayder Al-Zubaidi and Baydai Abdulameer will present their Master’s thesis entitled “Statistical Characterization of User Effects on MIMO Terminal Antennas”.
Location: E:2311 (E-building, LTH)
The thesis is supervised by Ivaylo Vasilev and I am the examiner.
You are very welcome to attend!
Boun Kiong Lau
2013-11-28 Seminar: Magnus Herberthson
On Thursday November 28 at 10.15 in E:2311, Dr Magnus Herberthson from FOI and Linköping University, will give a seminar.
"A rigorous high frequency limit investigation of the MIE scattering from a PEC sphere"
We look at the MIE scattering from a PEC (perfect electric conductor) sphere, i.e., we consider the standard situation in frequency domain where the sphere is illuminated by a plane wave. Due to the spherical symmetry, the scattered and hence the total field can be written down explicitly through a series of spherical vector waves. It is often assumed that in the high frequency region, the scattered field at the surface of the sphere is well approximated by the so called physical optics approximation. This assumptions can, for the sphere, be formulated as an explicit limit statement. We will give this formulation, and discuss its validity. The talk consists of two parts. The first part describes the formulation mentioned above, some difficulties one encounters in the proof, and some useful mathematical facts/tools. The second part discusses (parts of) the proof itself.
2013-11-26 Seminar: Torleif Martin
On Tuesday 26/11 at 13.15 in E:3139, Dr Torleif Martin from Saab Aeronautics AB, Linköping, will give a seminar with the title
"Electromagnetic theory – an industrial perspective"
Dr Torleif Martin has collaborated with the group of Electromagnetic Theory at EIT on many aspects of scattering and computational problems the last several years. He took his PhD in Theoretical Physics from Linköping Institute of Technology, and has mainly been with FOI and Saab, where he has worked on electromagnetic problems in areas such as antennas, radar cross section, and electromagnetic compatibility. The group of Electromagnetic Theory is currently recruiting him as Adjunct Professor, and in his seminar he will give bit of background of himself, his previous activities, and his future involvement with the EIT department.
2013-11-27 Seminar: John B Anderson
The Communication group invites you to our “Wednesday seminar”, given on November 27th. The talk will be given by John B. Anderson.
Title: "Finding Best Convolutional Codes for Faster-Than-Nyquist Signaling" Speaker: John B. Anderson
Date: November 27, 2013
Combining convolutional codes with an intersymbol interference channel - “turbo equalization” - is a powerful method of coding when 3-5 bits/Hz-s are to be transmitted. However, which code works best with an iterative decoder depends more on the EXIT tunnel and a correlation property of the code than it does on the minimum distance. Faster-than-Nyquist signaling and a new type of EXIT chart are the keys to defining the problem and identifying best codes.
2013-11-06 Seminar: Norbert Goertz
Title: Dense-Signal Recovery from Incomplete Observations
Speaker: Norbert Goertz, TU Wien
Date: November 6, 2013
John B Anderson
We report on experiments with what we define as "dense signals'' -- a class of signal vectors with both discrete-amplitude as well as continuous-amplitude components -- that, similar to sparse signals, can be recovered exactly from incomplete linear measurements with high very probability, when a suitably defined convex optimization problem is solved. We present a recovery algorithm based on the minimization of the maximum-norm as well as a fast scheme based on a variation of iterative hard thresholding that can be used for dense-signal recovery in practice.
2013-11-07 Licentiatavhandling: Linnéa Larsson
Welcome to attend Linnéa Larssons licentiate seminar on thursday Nov 7 at 10.15 in E:2311. Her thesis is entitled "Event Detection in Eye-Tracking Data for ?Use in Applications with Dynamic Stimuli?". Opponent is Christer Ahlström who works with signal processing and human-vehicle-transport system interaction at the Swedish National Road and Transport Research Institute (VTI).
This licentiate thesis has signal processing of eye-tracking data as its main theme. The thesis comprises two papers describing two parts of a novel algorithm for detection of four types of eye movements in data recorded using dynamic stimuli. The four eye movements are: saccades, fixations, smooth pursuit movements, and postsaccadic oscillations (PSO). The first paper presents a method for detection of saccades and PSO. The saccades are detected using the acceleration signal and three specialized criteria based on directional information. In order to detect the PSO, the interval after each saccade is modeled and the parameters of the model are used to determine if PSO are present or not. The algorithm was evaluated by comparing the detection results to manual annotations and to the detection results using the most recent PSO detection algorithm. The results show that the algorithm is in good agreement with the annotations, and has better performance than the most recent algorithm. In the second paper, a method for separation of fixations and smooth pursuit movements is proposed. In the intervals between the detected saccades/PSO, the algorithm uses different spatial scales of the position signal in order to separate the two types of eye movements. The algorithm is evaluated by computing five different performance measures, showing both general and detailed aspects of the detection performance. The detections of the algorithm are compared to the results of a velocity and dispersion based algorithm (I-VDT) proposed for smooth pursuit detection, and to manual annotations of two experts. The results show that the proposed algorithm performs considerably better than the compared algorithm.
2013-11-06 Seminar: Maria Tengner
Title: "A story of quantum information – photons, entanglement and keeping secret"
Speaker: Maria Tengner, SAAB Dynamics, Linköping
Date: November 6.
Any information processing requires the information in question to be carried by some physical system. The aspiration of quantum information sciences is to utilize quantum mechanical systems rather than classical systems as such information carriers, i.e. we want to encode information on quantum states, then store, transmit, process, and detect this information by manipulating these quantum states. Such control over individual quantum states is not easily achieved, but if it can be done it opens up new possibilities for what is attainable within information sciences. The reason for this is that by using quantum systems instead of classical systems, the properties of quantum physics, such as superpositions and entanglement, become accessible, making it possible to perform tasks with a quantum information system that are simply impossible with a classical information system. The exploration of these possibilities constitutes the field of quantum information sciences. In this talk I will give a brief introduction to the field of quantum information sciences by looking at the development of the field and some of its key concepts. In particular, I will focus on optical quantum information, i.e. when photons are used as information carriers. I will discuss entanglement, and why it is so important in quantum information sciences, and how entangled photons can be used for secure communication.
2013-10-29 Ex-jobbspresentation: Apostolos Tsiaras
On Tue, 29 Oct 2013, Apostolos Tsiaras will present his Master’s thesis entitled “SAR Evaluation in Multi-antenna Mobile Handsets”.
The work is performed in collaboration with ART-FI, a startup company in France.
Location: E:3139 (E-building, LTH)
The thesis is supervised by Hui Li and I am the examiner.
You are very welcome to attend!
Buon Kiong Lau
2013-10-30 Ex-jobbspresentaion: Tobias Lindqvist
On Wednesday October 30 14.15 Tobias Lindqvist will present his master's thesis "Field parametrization by trigonometric expansion combined with particle tracking in superconducting cavities" in E3139.
The European Spallation Source research facility will be built in Lund. The driver of the spallation process will be a linear proton accelerator that will deliver 2.5 GeV protons to a tungsten target. At a beam power of 5 MW, ESS has set the ambitious goal of maintaining a beam loss of 0.1 MW/m . To achieve such a goal the properties of the accelerator need to be explored in great detail. This master thesis examines and implements a model for tracking particles inside the superconducting elliptical cavities of a linear accelerator. The model uses a parametrization of the eld to vastly reduce the amount of data required to describe the eld. The model can be used to compare and examine different designs of the accelerator, and predict accelerator behavior once the accelerator is running.
2013-10-22 ELLIIT Annual Workshop
Welcome to the ELLIIT Annual Workshop
October 22-23, 2013,
Faculty of Engineering, Lund University Lund, Sweden.
The workshop covers reports and discussions on the research activities carried out within the ELLIIT center of excellence over the past year.
The workshop also features keynote speakers Karl Iagnemma from MIT and Björn Ekelund from Ericsson as well as two half-day workshops, one on autonomous and unmanned vehicles as well as one on mobile cloud technologies.
More information about the workshop can be found here
The workshop is free of charge.
Contact: Pia Bruhn
2013-10-24: Prof. Yann Deval
Thursday 24 October at 14.15-15.00 in room E:2311 Prof. Yann Deval from University of Bordeaux will give a talk titled
“An overview of Full Software Radio Transceivers”
He has been invited here as a faculty opponent to Carl Bryant. We are very happy that he will visit us, and after the PhD defense we now also have this opportunity to learn more about his research. The abstract of the talk is included below. I hope that many of you have possibility to attend!
This talk addresses the current status and trends in research on Full Software Radio (FSR) systems. Concerning the receiver path after defining Software Defined Radio (SDR) versus FSR a new version of the Sampled Analog Signal Processor (SASP) is presented and experimental results are discussed. The new version doubles the signal bandwidth of the initial prototype while the power consumption is reduced to less than 100 mW. For the transmitter path, two new FSR architectures are presented based on Riemann’s algorithm and a charge pump (namely the Riemann’s Pump), first, and a Fourier recombination transmitter (FoRTX) in nanoscale CMOS, second. A GaN demonstrator of the Riemann’s Pump is presented, which generates any signal including concurrent emissions in the 0-to-1 GHz band, and a 28nm SOI CMOS FoRTX is presented as well addressing roughly the same frequency band.
PhD thesis defence Carl Bryant
Date: 2013-10-24, at 10:15
Place: Room E:1406, E-building, Ole Römers väg 3, Lund University, Faculty of Engineering
Opponent: Professor Yann Deval. University of Bordeaux, France
Receiver Front-Ends in CMOS with Ultra-Low Power Consumption
ISBN 978-91-7473-708-0, Dep. Electrical & Information Technology, 2013.
Abstract: Historically, research on radio communication has focused on improving range and data rate. In the last decade, however, there has been an increasing demand for low power and low cost radios that can provide connectivity with small devices around us. They should be able to offer basic connectivity with a power consumption low enough to function extended periods of time on a single battery charge, or even energy scavenged from the surroundings. This work is focused on the design of ultra-low power receiver front-ends intended for a receiver operating in the 2.4GHz ISM band, having an active power consumption of 1mW and chip area of 1mm². Low power consumption and small size make it hard to achieve good sensitivity and tolerance to interference. This thesis starts with an introduction to the overall receiver specifications, low power radio and radio standards, front-end and LO generation architectures and building blocks, followed by the four included papers.
Paper I demonstrates an inductorless front-end operating at 915MHz, including a frequency divider for quadrature LO generation. An LO generator operating at 2.4GHz is shown in Paper II, enabling a front-end operating above 2GHz. Papers III and IV contain circuits with combined front-end and LO generator operating at or above the full 2.45GHz target frequency. They use VCO and frequency divider topologies that offer efficient operation and low quadrature error. An efficient passive-mixer design with improved suppression of interference, enables an LNA-less design in Paper IV capable of operating without a SAW-filter.
2013-10-10 Ex-jobbspresentation: Longyang Lin och Shuai Cheng
At 13.15, in E:2311, Longyang Lin and Shuai Cheng will present their master thesis entitled:
"Low Energy Data Compression for Brain Implant"
Abstract: A digital implementation of a data compressor for wireless brain machine interface with an attractive area and low energy cost is presented. This design consists of pre-processing filters, spike detectors, a spike compressor and Serial Peripheral Interface (SPI) IO protocol, targeting a 65 nm CMOS technology. Area and energy dissipation have been dramatically reduced by resources sharing, architectural optimizations and using a standard-cell based latch memory. Moreover, the compression ratio for each channel is adjustable by the users, which can vary from 4 to 64 based on the channel quality and the need of spike reconstruction accuracy. The reconstruction accuracy of the fixed-point digital implementation can reach more than 95%, similar to the full precision Matlab model (less than 0.1% loss). Additionally aggressive voltage scaling (down to the sub-VT region), clock gating and multiple clock domain have been performed resulting in a total die area of 900×500 μm2 for 16 channels. Energy dissipation in the sub-Vt region is estimated using a high-level sub-Vt energy model. The estimated value is 1.03 pJ/clock cycle, which is 30× improvement compared to the standard super-VT implementation without clock gating.
2013-10-03 Colloquim: Prof Suman Datta
On the 3rd of October at 15:15 we will arrange a colloquium with Professor Suman Datta from Pennsylvania State University, USA.
The title of his talk will be "Tunnel Transistors for Low Power Logic and Analog Applications"
The talk will be given in E-building, room E:C.
Professor Datta is a highly distinguished researcher and prior to joining Penn State, he was a Principal Engineer in the Advanced Transistor and Nanotechnology Group at Intel Corporation. He held the Joseph Monkowski Professorship for Early Faculty Career Development. He is exploring new materials, novel nanofabrication techniques, new classical and non-classical device structures for CMOS "enhancement" and CMOS "replacement" for future energy efficient, high performance and information processing systems.
He has published over 140 journal and conference articles and holds 145 US and international patents related to advanced process technologies and transistor architecture. He is an IEEE fellow and a distinguished lecturer of the IEEE Electron Devices Society.
He has also received numerous distinguished honors and awards such as:
IBM Faculty Award
Penn State Engineering Alumni Society Outstanding Research Award
Monkowski Career Development Professorship
Intel Achievement Award
Divisional Recognition Award from the Intel Logic Technology Development Group
PhD thesis defence Anil Dey
Date: 2013-10-04, at 10:00
Place: Lecture hall E:1406, E-building, Ole römers väg 3, Faculty of Engineering Lund University
Opponent: Prof. Suman Datta. Pennsylvania State University, USA
Low-Power Nanowire Circuits and Transistors
ISSN 1654-790X, No. 51, 2013.
Abstract: This thesis explores several novel material systems and innovative device concepts enabled by nanowire technology. State-of-the-art fabrication techniques such as electron beam lithography and atomic layer deposition are utilized to achieve high control and quality in the device fabrication. The devices in this thesis are based on two main types of design geometries, lateral and vertical, each of which have strengths and weaknesses. The first part of the thesis describes the goals of future metal-oxide-semiconductor field-effect transistors (MOSFETs) and discusses the ultimate scalability surrounding experimental results for 15-nm-diameter InAs nanowires and how they compare to other state-of-the-art transistors. The extracted on-resistance (Ron = 250 ?·µm) and drive currents (Ion = 1250 µA/µm) are comparable to state-of-the-art high-electron-mobility transistors (HEMTs) from MIT and quantum-well field-effect transistors from Intel. The outstanding performance is mainly attributed to the reduced access resistance achieved through an n+-i-n+ doping profile. The extracted mobilities also agree well with state-of-the-art and theoretical predictions for extremely scaled devices.
The second part of the thesis discusses how nanowires may be employed to enable III-V complementary metal-oxide-semiconductor (CMOS) digital logic. Nanowires enable the formation of both n-type semiconductors and p-type semiconductors, which are a requirement for CMOS, in a single nanowire and allow for integration on a Si platform. III-V MOSFETs are frequently employed for analog applications, but there is a disconnect regarding p-type devices, which are also required for digital logic. The individual segments of the nanowire are evaluated as well as the entire nanowire in an inverter configuration. This thesis then presents a strategy for matching the drive currents n- and p-type MOSFETs.
The final part of the thesis deals with a family of devices that operate according to principles fundamentally different from those of a traditional MOSFET, namely tunnel FETs (TFETs). There is a demand for steep-slope devices such as TFETs to enable supply-voltage scaling to reduce the power dissipation. Although devices have demonstrated <60 mV/decade operation, they commonly suffer from low on-currents. To maximize the drive current, the broken band gap alignment of GaSb/InAs is exploited to allow for a direct tunneling mechanism. The material system is first explored as Esaki diodes and in various doping profiles to understand the influence of doping on device performance. The devices are further evolved into TFETs by the addition of a high-k gate dielectric and an additional terminal. Experimental results display high on-currents of Ion = 310 µA/µm comparable to other state-of-the-art TFETs. Finally, an innovative design concept combining axial and radial heterostructures is utilized to design a radial TFET with a small footprint. A radial GaSb/InAs core/shell TFET provides an attractive way to increase the drive current of a TFET without compromising either device electrostatics or chip area. The functionality of radial Esaki diodes and TFETs is demonstrated and evaluated by the maximum peak currents, which are much improved as compared to their axial counterparts when normalized to the largest cross-sectional area of the nanowire, assuming a vertical device geometry, illustrating the advantage of a core--shell architecture. The dimensions of the InAs shells are below 15-nm and display clear quantization effects revealed in low-temperature electrical characterization.
2013-09-25 OBS! Ny sal och tid!!!! Ex-jobbspresentation: Oscar Sandström och Alfred Nyberg
On Wednesday September 25, Oscar Sandström and Alfred Nyberg will present their Master thesis entitled:
Review and upgrade of a sensor interface system for airfield automation / Undersökning och uppgradering av ett sensor interface-system för automation på flygfält
The presentation will be in room E:2349 at 10.15; you are all most welcome.
Andreas Sjölin at the company Safe Gate Group and I have been the supervisors and Henrik Sjöland is the examiner.
Increasing demand of power efficient products has led Safegate Group International AB to decrease system current from 6.6 A to 2.0 A . This decrease requires upgrade and review of their existing product portfolio. The master thesis focal point was to renew one of their products; a Sensor Interface Unit, a SIU, for use in the 2.0 A system. After examining the existing unit, the conclusion was that a new version of SIU had to be developed, not only to meet present specifications in the new 2 A system but also to satisfy future needs. Due to the scope of the project, this master thesis only focused on product planning, prototype development and prototype evaluation. An existing Safegate platform, an airfield light controller developed for the 2 A system, was used to create a new SIU prototype. SIU functionality was achieved by developing a hardware extension board including surge protection, filters, level conversions and a variable voltage power supply. Due to requests by Safegate, software was written in a platform and hardware independent fashion resulting in modular and encapsulated architecture. Evaluation of the prototype suggests that the concept, with a few adjustments, can be developed into a stable and useful product for Safegate Group International AB.
2013-09-23 Lund Circuit Design Workshop
Welcome to the Lund Circuit Design Workshop
September 23-24, 2013
Grand Hotel and Faculty of Engineering, Lund University Lund, Sweden
The workshop will offer an overview of the research activities in IC design at Lund University
Additionally, invited presentations will be given by outstanding experts from academia and industry.
The workshop is hosted by the VINNOVA Industrial Excellence Center in System Design on Silicon (SoS) and by the Swedish Foundation for Strategic Research projects:
Wireless Communication for Ultra Portable Devices (UPD)
Wireless with Wires (WWW)
Distributed antenna systems for efficient wireless systems (Distrant)
Digitally-Assisted Radio Evolution (DARE)
The workshop is free of charge.
Deadline: September 2, 2013
Contact: Pia Bruhn, Pia.Bruhn@eit.lth.se
Workshop website: http://cdworkshop.eit.lth.se/
Professor Jan Rabaey (UC Berkeley), advisor to SoS and Honorary Doctor of Lund University 2012, will give an invited talk at Innovation in Mind, an event in Lund September 18-20. Jan’s talk has the title The innovation is in the Mind – neuroscience, nanotechnology and information technology on a convergence path.
2013-09-19 Ex-jobbspresentation: Muris Sarajlic
On Thursday, September 19, at 13.15 in E:3139, Muris Sarajlic will present his MSc thesis on
"Power and performance trade-off in LTE receiver chains"
The thesis first investigates appropriate ways of modeling the effects of RF impairments and then finds a way to connect the tunable block parameters through RF impairments with the overall performance; this is accomplished by determining the Error Vector Magnitude (EVM) of each RF impairment source, calculating the total receiver EVM and using the information about thermal noise and EVM to determine the Signal-to-Noise-and-Distortion ratio (SNDR) which serves as the overall channel quality indicator. After this, the behavior of individual blocks of the direct-conversion receiver is analyzed in terms of a functional dependence between the value of tunable block parameters and power consumption; the analysis is performed through analyzing the trade-offs between power consumption and performance for each block, as well as for the entire receiver. Finally, using the knowledge about the power and performance behaviour of blocks due to the tuning of block parameters, a simple strategy of digital control is proposed and supported by initial calculations, which show that power consumption savings are possible when using a channel-adaptive, digitally tunable receiver.
2013-09-20 Ex-jobbspresentation: Teresia Olsson
On Friday, September 20, Teresia Olsson will present her thesis
"High-chromaticity Optics for the MAX IV 3 GeV Storage Ring"
The presentation takes place in the room Everest on the fifth floor of the E-building, at 10.15.
The MAX IV facility is currently under construction in Lund, Sweden and is going to be a 3rd generation state-of-the-art synchrotron light source.The MAX IV 3 GeV storage ring has a large negative natural chromaticity, and this has to be corrected to positive values to prevent head-tail instabilities. On the other hand, high linear chromaticity can lead to a large tune footprint which limits Touschek lifetime. Therefore, the linear chromaticity is corrected to +1 in both planes with sextupoles while octupoles are used to further reduce the tune footprint. Studies indicate this design leads to threshold currents for resistive wall and transverse mode coupling instabilities beyond what is expected during regular user operation. However, since these are only preliminary studies based on several approximations, the possibility of instability issues during commissioning needs to be considered. A short term solution is to operate the storage ring at a higher chromaticity. This thesis describes the development of a high-chromaticity optics for the MAX IV 3 GeV storage ring with linear chromaticity +4 in both planes. It focuses on reduction of chromatic and amplitude-dependent tune shifts to maximize dynamic aperture and Touschek lifetime. A comparison between the performance of the high-chromaticity optics and the design optics is also presented. The results show that the performance of the developed high-chromaticity optics is poorer than the performance of the design optics. Still, the performance of the high-chromaticity optics is expected to be sucient good to allow the optics to be applied in the MAX IV 3 GeV storage ring as a short term solution if instability issues should occur during commissioning.
2013-08-29 Ex-jobbspresentation: Ellinor Persson och Ahmed Almuqayad
On Thursday 29/8 at 13.15 in E:2349 the following master thesis will be presented:
Ellinor Persson and Ahmed Almuqayad:
"A study of the field distribution on finite array endfire antennas"
I have been the supervisor, and Mats Gustafsson is the examiner.
An endfire antenna is an array built up by smaller antennas, and is designed to direct the radiated power along the structure. In this thesis, we look at a very simple form, consisting of equally spaced monopoles over a perfectly conducting ground plane. After defining a unit cell, we set out to investigate whether a periodic structure model can be applied and used to calculate the electric field distribution on the antenna. The work is an initial study for a future research project on developing a more efficient algorithm for simulating very large array antennas. Hypotheses are tested in an iterative manner. It is concluded that the field distribution differs from the one on a passive periodic structure, and a correction is proposed. The correction gives a better fit, but is not enough to pick up all variations. Remaining problems are the dependence on the number of elements in the array and to find a way to calculate the coefficients of the correction in advance. Also, the hypothesis that the propagation constant is real could not be falsified with the method used.
2013-08-29 Ex-jobbspresentation: Simon Davani och Malda Said
On Thursday, Simon Davani and Malda Said will present their Master thesis entitled Cooperative Wireless Network Configuration Protocol Design in E:3139 at 10.15; you are all most welcome. I have been the supervisor and Ulf Körner is the examiner.
With the popularity of the IEEE 802.11 standard ever increasing, scanning and finding an access point is not an issue in medium to large cities as access points are deployed densely. Usually when searching for local access points in an area, it results in a number of available access points. This dense deployment of access points causes problems with just three non- overlapping frequency channels in the unlicensed 2.4 GHz band, especially in uncoordinated deployments. For example, residential users usually place their access points (routers) on locations according to their own contentment, and configure them without taking their neighbouring access points’ configuration into consideration. As a consequence the signal coverage areas of the different access points overlap which results in interference if they are running on the same channel. Therefore, we present an inter-access point protocol, working with the help of newly constructed text-based frames, which enables cooperation between access points and handles different functions to mitigate interference and improve the network performance. The protocol functions that have been considered and included in the protocol are; channel assignment, virtual TDMA between access points, balancing the load between access points and switching off interfering access points. Simulations were performed in ns-3, and it was concluded that by introducing the protocol functions the overall network throughput is enhanced.
2013-08-29 Ex-jobbspresentation: Mattias Frisk
Torsdagen den 29/8 kl 13.15 i E:3139 kommer Mattias Frisk att presentera sitt examensarbete "Matchningar av produkter mellan nätbutiker"
Onlinebutiker har oftast väldigt stora databaser över vilka produkter de säljer. I dessa kan det finnas dubbletter vilket såklart är oönskvärt. Att hitta dubbletter är nästintill omöjligt att göra manuellt på en databas som har tusentals poster. Det finns även marknad för företag som sammanställer försäljningsdata över produkter från flera olika onlinebutiker och det finns inte alltid unika ID för att trivialt kunna se vilka produkter som är dubbletter över olika databaser.
Målet med examensarbetet är att göra ett program som hittar dubbletter eller liknade objekt i en godtycklig samling av objekt. Programmet ska fungera oberoende på hur datan i objekten är strukturerad. Programmet ska även fungera om objekten i samma samling är strukturerade på olika sätt. Annat än funktionaliteten finns det inga krav på programmet. Fokus kommer ligga på att identifiera dubbletter och om tid blir över kommer det utbyggas för att gruppera objekt som är ”lika” varandra. Datan från olika databaser kommer konverteras till fritext (då detta är det mest generella) genom att t.ex. slå ihop alla tabellerna i en rad. Verifiering Till verifiering kommer det finnas ett antal databaser med EAN-koder. Man antar att två objekt är dubbletter om EAN-koden är samma. Programmet kommer att försöka matcha ihop objekt utan att ta hjälp från EAN-koden. Från detta kommer man kunna se hur mycket av de faktiskta dubbletterna programmet har lyckats matcha och hur många felaktiga matchningar det har gjort. Dessa databaser kommer skapas genom att crawla igenom onlinebutikers hemsidor.
2013-06-26 Ex-jobbspresentation: Pramod Ghimire
On Wednesday (26-06-2013) at 13:15 in E:2311 Pramod Ghimire will present his Master Thesis entitled:
"An Equalizer for an Integrated Optical Receivers in 65nm CMOS"
This thesis describes the design of an analog equalizer for a fully integrated optical receiver in 65nm CMOS process. An equalizer is designed which compensates for the limited bandwidth of n-well/p-sub photodiode with a roll-down of about 5dB/decade. A transimpedance amplifier is also designed which has transimpedance gain of 70dB and 3dB bandwidth of 1.3GHz. Equalization extends the bandwidth of the optical receiver to 1GHz from 6MHz. The design is simulated on different process corners to ensure robust operation. Total power consumption is 7mW.
2013-06-25 Ex-jobbspresentation: Jacob Ferm
Jacob Ferm will present his master thesis project
“Security analysis of WLAN roaming solutions”
Tuesday June 25, 10.15, room E:3139, E-building LTH
The project has been performed at Anyfi Networks AB
WLAN access has become not only a popular technology for Internet access in the private homes and residential environments but it also is gaining in popularity for use in public environments. For example, in several countries, e.g. the US, we see today that WLAN networks are deployed to offload the growing traffic demands in mobile networks. One of the problems for such WLAN networks is to arrange the operation of WLAN security. To secure that WLAN access is only available to authorized users and is subsequently protected through encryption and data integrity functions requires that proper credentials must be in place at the access points and the clients that request access. Today there are different solutions to solve this issue of secure access and credential handling. Anyfi Networks has developed a platform they call Anyfi.net, representing a quite novel approach. In Anyfi.net the access point itself does not have access to the authentication credentials or encryption keys for the connection. Instead the raw WLAN radio traffic is routed via knowledge provided by a discovery server to the access point where the original credentials reside. This means that the ordinary WLAN security mechanism protects the traffic end-to-end, all the way from the mobile device to an environment trusted by the user of the mobile device.
2013-06-20 Ex-jobbspresentation: Masoud Khazaee
Masoud Khazaee will present his master thesis
"Two-layer Topology Design of Wireless Optical Networks"
on Thursday 20/6 at 13.15 in E:3139
Michal Pioro has been his supervisor and Ulf Körner the examiner.
Free Space Optical (FSO) communication is an evolving wireless technology capable of providing point to point broadband links, using direct light-beam connections. Currently, FSO links have the range of up to several kilometers, and realize transmission rates of the order of 2.5 Gb/s. At the same time, FSO links do not interfere with each other and thus a set of FSO links can work simultaneously when organized in a wireless mesh network (WMN). Therefore, the resulting optical WMN (OWMN) is a promising networking solution for assuring the Internet access in metropolitan areas. High link bandwidth and lack of interference is an important advantage over traditional radio WMN (RWMN) solutions based on the IEEE 802.11 family standards. Still, FSO requires direct visibility between the end points of the installed optical links, and is more sensitive to weather conditions (e.g., to fog) than the radio links. Consequently we use mirror in order to conquer the difficulty of obstacles such as high building. The thesis consists of elaborating an optimization model of OWMN for two-layer topology design that includes the mirror-layer and the optical node layer. This thesis consists of three main parts. In the first part, we survey the development of the technology of FSO, describing the advantages and some drawbacks. Then, we set up a comprehensive mixed integer programing model for the topology design considering both optical nodes and mirror nodes. We aim to minimize the cost of deployment when some links are blocked by obstacles. At last, we make several numerical studies to show how the model works under different scenarios.
2013-06-20 Ex-jobbspresentation: Hanqing Zhang
On Thursday at 10:15 in E:2311 Hanqing Zhang will present his Master Thesis entitled:
"Design and Optimization of an Asynchronous MIPS Processor Based On De-synchronization"
With the improvement of Complementary Metal Oxide Semiconductor (CMOS) technologies in size, lots of interests are gained by asynchronous circuits for its advantages in robustness, potential for low power, low Electromagnetic Interfer- ence (EMI). Asynchronous approaches are capable of liberating Very-large-scale Integration (VLSI) designs from global clock tree problems. This trend reveals the need of modelling and design methodologies for asynchronous circuit. Moti- vated by the goal of creating a energy ecient, high performance processor for low power application , the design of 32-bit MIPS asynchronous processor based on desynchronization ow is presented. With a highly automated Tcl (Tool Command Language) scripts targeted at the processor implementation using Synopsys Design Compiler, the Hardware Description Language (HDL) specication of synchronous processor is desynchronized using bundled data model. Optimization methodology took advantage of instruction dependent delay line multiplexing along with ex- tended pipeline stages balanced for average performance. This strategy leads to a design of a nine-pipeline-stage MIPS processor. As a result, comparing with a typ- ical ve-pipe-line stage synchronous MIPS processor with 200MHz clock frequency under 65 nm CMOS technology, the de-synchronized nine-pipeline-stage processor can achieve at most 22% better equivalent clock frequency in performance as well as other nice properties of asynchronous circuits such as low peak power, robust to variations and ecient power management.
2013-06-19 Ex-jobbspresentation: Fredrik Karlsson och Daniel Perván
Fredrik Karlsson and Daniel Perván kommer att presentera sitt exjobb (på svenska) den 19/6 kl 13:15 i E3139
Titel: "Course Recommendation Using Aggregated Information Retrieval"
Study plan scheduling is a natural part of each student's life cycle at LTH. In the fourth and fth year, the students have to customize their education by selecting courses that will give them the possibility to graduate.The selection and scheduling is done ad hoc by the students and is most often a quite time consuming act; the information is spread across di erent web pages and a lot of values needs to be summarized manually by the student. In this master thesis we will examine the possibility of simplifying the process by implementing a prototype web based solution with aggregated searching of multiple sources and data structures. We will also try to enhance the result set by the use of statistical data, possibly from LADOK and data from CEQ at LTH. While some data may be mined directly from different databases or web pages, some may need to be produced through calculation. A web based tool for study plan scheduling, searching and browsing has been developed. It aggregates much of available data about the courses given at LTH. The system helps the students both by letting them create and validate their study plan, as well as easily explore suitable courses at LTH. Recommendation engines, based on course content and other statistics, are used to suggest suitable courses for the students.
2013-06-19 Ex-jobbspresentation: Yuqi Liu
On Wednesday June 19th at 10.15 Yuqi Liu will present his master thesis "Clock tree design in sub-VT circuits - Analysis on standard- and full-custom gates" in E:2311.
Process variation in the subthreshold (sub-VT) region is a key factor affects the functionality of clock tree. Standard-cell clock buffers experience a lot more performance variation among different design corners than that in the nominal supply. This variation effect should thus be reduced to ensure functionality of the clock tree. In this report, some aspects of constructing clock trees for sub-VT operation have been studied and compared. It has been confirmed that performing clock tree synthesis (CTS) at nominal supply and load lower supply libraries is more applicable than doing CTS directly at sub-VT voltage. Also a dual-VT customized buffer has been developed for constructing robust clock network. It has been approved in the simulation that this customized buffer could increase the performance of delay, skew and slew by a factor of 9.8X, 8.9X and 9.8X respectively at 400mV.
2013-06-18 Ex-jobbspresentation: Rasmus Wulff
On Tuesday 18th of June 10.15 Rasmus Wulff will present his master thesis, “Numerical Modeling of Nanowire Transistors”. The presentation takes place in E:2311.
Numerical modeling and simulations of wrap-gate InAs nanowire transistors are performed using Atlas from Silvaco. The Drift-Diffusion Transport Model, the Drift-Diffusion Mode-Space Method and the Non-Equilibrium Green’s Function (NEGF) Mode-Space Approach are used. The latter two simulation models take account of quantum mechanical effects, whereas the first one is a purely classical model. The NEGF mode-space model simulates ballistic transistors, where carriers are unaffected by scattering. The simulations are performed to investigate the effects of varying device parameters such as the NW radius, gate length and n-type doping. Transistors are assessed and compared using figures of merit like maximum transconductance, inverse subthreshold slope, DIBL and on-resistance. Some comparisons are made between the different models used. It is found that low channel-doping is preferable, and that the low-doped region should be aligned with, or within, the gate edges. The simulations show that it is possible to scale the gate length to ≤50nm for nanowire (NW) radiuses of ≤10nm, and still get reasonably good inverse SS. For these transistors it is less important that the doping in the channel is lower than in the rest of the NW.
2013-06-17 Disputation: Paul Stankovski
Monday June 17th, 2013, at 10.15 Paul Stankovski defends his doctoral dissertation “Cryptanalysis of Selected Stream Ciphers” in lecture hall E:1406, Ole Römers väg 3, S-223 63 Lund, Sweden.
Faculty opponent is Prof. Anne Canteaut, INRIA Paris-Rocquencourt, France.
The aim of this dissertation is to show some cryptanalytical results on a selection of stream ciphers. We have grouped theory and results into three main parts.
The first part focuses on the FCSR-based constructions X-FCSR and F-FCSR-H v3. For the X-FCSR family of stream ciphers we perform a severe state recovery attack. This attack works for both X-FCSR-128 and X-FCSR-256.
We then develop a generalized birthday algorithm for finding linear relations in FCSRs. This algorithm applies to the most recent and general FCSR architecture, the ring FCSR, so it can be used for analyzing the FCSR of any FCSR-based design. We apply the algorithm to produce an efficient distinguisher for F-FCSR-H v3, which was previously unbroken.
The second part of the dissertation covers topics related to the HC family of stream ciphers. First, a very general treatment of sampling methods is presented. Surprisingly, perhaps, a positive result is given. We prove that an efficient sampling method based on sampling vector weights is optimal in a given context. This sampling technique is employed to produce the best known distinguisher for HC-128.
We go on to show a few theoretical results on functions that use word rotation and xor. These results are applied to a modified variant of HC-128, and this application shows how the theory could be used in a cryptanalytical scenario. It also shows the important role of the addition operator in HC-128, without which the cipher would be much less secure.
In the third part of the dissertation we analyze stream ciphers, and block ciphers to a lesser extent, using algebraic methods. We develop a simple and intuitive greedy algorithm for automatic security testing of cryptographic primitives. This is done in a black box fashion, without using any information on the internal structure of the primitives. Despite this, it is shown how structural information is revealed very clearly under certain circumstances. The main features here are some nice results for the well-known stream ciphers Trivium, Grain-128 and Grain v1.
List of publications:
 P. Stankovski and M. Hell. An Optimal Sampling Technique for Distinguishing Random S-boxes. ISIT, July 2012, pp. 846-850, http://dx.doi.org/10.1109/ISIT.2012.6284680.
 P. Stankovski, M. Hell and T. Johansson. An Efficient State Recovery Attack on X-FCSR-256. FSE 2009, ed. by O. Dunkelman, vol. 5665, Lecture Notes in Computer Science, Springer-Verlag, 2009, pp. 23-37, ISBN: 978-3-642-03316-2, http://dx.doi.org/10.1007/978-3-642-03317-9_2.
 P. Stankovski, M. Hell and T. Johansson. An Efficient State Recovery Attack on the X-FCSR Family of Stream Ciphers, Journal of Cryptology, Online First TM (November 2012), pp. 1-22, ISSN: 0933-2790, http://dx.doi.org/10.1007/s00145-012-9130-9. Extended version of .
 P. Stankovski, M. Hell and T. Johansson. Analysis of Xorrotation with Application to an HC-128 Variant. ACISP12, ed. by W. Susilo, Y. Mu and J. Seberry, vol. 7372, LNCS, Springer Berlin Heidelberg, 2012, pp. 419-425, ISBN: 978-3-642-31447-6, http://dx.doi.org/10.1007/978-3-642-31448-3_31.
 P. Stankovski, S. Ruj, M. Hell and T. Johansson. Improved Distinguishers for HC-128. Designs, Codes and Cryptography 63 (Feb 2012), pp. 225-240, ISSN: 0925-1022, http://dx.doi.org/10.1007/s10623-011-9550-9.
 P. Stankovski. Greedy Distinguishers and Nonrandomness Detectors. Progress in Cryptology - INDOCRYPT 2010, ed. G. Gong and K. C. Gupta, vol. 6498, Lecture Notes in Computer Science, Springer-Verlag, 2010, pp. 210-226, http://dx.doi.org/10.1007/978-3-642-17401-8_16.
 H. Wang, P. Stankovski and T. Johansson. A Generalized Birthday Approach for Efficiently Finding Linear Relations in l-sequences. Accepted for publication in Designs, Codes and Cryptography on March 23rd, 2013.
2013-06-17 Ex-jobbspresentation: Dimitrios Vlastaras och Daniel Lestón
On Monday June 17th at 9.15 Dimitrios Vlastaras and Daniel Lestón will present their master’s thesis “Improved traﬃc safety by wireless vehicular communication” in E:2311. You are all welcome to attend.
Soon vehicles will have the ability to communicate and cooperate with each other in order to avoid collisions and provide useful informa- tion to each other. However, for this cooperation to be possible all vehicles will have to be equipped with compatible wireless 802.11p modules. In an early stage of the system there will be plenty of older vehicles without such equipment. This thesis addresses this problem by developing the hardware and software for a road side unit that detects older vehicles using a radar and forwards their position and speed vectors to newer vehicles using a compatible 802.11p module, avoiding potential accidents and saving human lives.
2013-06-14 Disputation: Kristin Persson
Friday 14th June, 2013, at 10.15 a.m. Kristin Persson defends her doctoral thesis "Radome Diagnostics: utilizing Source Reconstruction based on Surface Integral Representations" in lecture hall E:1406, Ole Römers väg 3, S-223 63 Lund, Sweden.
Faculty opponent is Prof. Margaret Cheney, Department of Electrical and Computer Engineering, Colorado State University, USA.
In this thesis, an inverse source reconstruction method with great potential in radome diagnostics is presented. A radome is a cover that encloses an antenna in order to protect it from environmental influences. Radome diagnostics are acquired in the design process, the delivery control, and in performance verification of repaired and newly developed radomes. A measured near or far field may indicate deviations, e.g., beam deflection, but the origins of the flaws are not uncovered. In this thesis, radome diagnostics is performed by imaging the tangential electromagnetic fields on radome surfaces, disclosing the radome influence on the electromagnetic fields as well as the positions and influences of defects.
The source reconstruction is based on a surface integral representation together with the extinction theorem. The extinction theorem and its associated surface integral equation ensure that the reconstructed tangential electromagnetic fields have their sources within the radome. The presence of axial symmetry in the measurement set-up enables usage of the fast Fourier transform to reduce the computational complexity. Furthermore, the problem is solved by an in-house body of revolution method of moments (MoM) code utilizing a singular value decomposition (SVD) for regularization. The reconstruction is performed on a fictitious surface in free space, located precisely outside the physical surface of the radome, i.e., no a priori information of the material of the radome is requested. Moreover, both synthetic and measured data are used to verify the method.
In Papers [1-3], the measurement set-up is a reflector antenna covered by a monolithic radome, and the near field is measured on a cylindrical surface. The height of the radome corresponds to 29-43 wavelengths in the frequency interval 8.0-12.0 GHz. The amplitude and phase of the tangential electromagnetic fields are reconstructed on the radome surface and the influence of the radome is investigated. Moreover, the alteration of the phase due to the transmission of the radome, the insertion phase delay (IPD), is imaged. Defects in the form of square copper patches, with an edge length corresponding to 1.6-2.4 wavelengths in the considered frequency interval, are attached to the radome wall. These might serve as a model for e.g., a lightning conductor or a Pitot tube. The attached patches alter the near field, and by applying source reconstruction, the disturbances of the patches are focused and detectable.
In Paper , the field is measured on a spherical sector in the far-field region at 10.0 GHz. Two set-ups with dielectric defects attached to the radome surface, are investigated. The aim is to investigate if variations in the electrical thickness of the radome wall can be detected. It is concluded that it is possible to discover dielectric patches of various edge sizes (0.5-2.0 wavelengths), and with the smallest thickness corresponding to a phase shift of a couple of degrees.
In Paper , a frequency selective (FSS) radome corresponding to a height of 51 wavelengths at the frequency 9.35 GHz is investigated. The electrical performance of an FSS radome depends on the periodic structure of the elements in the radome frame. The periodic structure of the investigated radome is disrupted by horizontal defects (vertical displacements of elements) and vertical defects (a column of missing elements). The far-field data is measured on a spherical sector, and the far-field data reveals that the radome changes the radiation properties. The tangential electromagnetic fields on the radome surface are reconstructed for several antenna illuminations to image the cause of these alterations. Furthermore, it is shown that the different components of the electromagnetic fields are affected differently by the defects, implying that both co- and cross-components of the electric and magnetic fields need to be considered. Moreover, the Poynting's vector is employed to visualize how the defects block the field from the antenna.
 K. Persson, M. Gustafsson, and G. Kristensson. Reconstruction and visualization of equivalent currents on a radome using an integral representation formulation. Progress In Electromagnetics Research, 20, 65-90, 2010.
 K. Persson and M. Gustafsson. Reconstruction of equivalent currents using a near-field data transformation - with radome applications. Progress in Electromagnetics Research, 54, 179-198, 2005.
 K. Persson and M. Gustafsson. Reconstruction of equivalent currents using the scalar surface integral representation. Technical Report LUTEDX/(TEAT- 7131)/1-25/(2005), Lund University, Department of Electrical and Information Technology, P.O. Box 118, S-221 00 Lund, Sweden, 2005.
 K. Persson, M. Gustafsson, G. Kristensson, and B. Widenberg. Radome diagnostics - source reconstruction of phase objects with an equivalent currents approach. Technical Report LUTEDX/(TEAT-7223)/1-22/(2012), Lund University, Department of Electrical and Information Technology, P.O. Box 118, S-221 00 Lund, Sweden, 2012. http://www.eit.lth.se.
 K. Persson, M. Gustafsson, G. Kristensson, and B. Widenberg. Source reconstruction by far-field data for imaging of defects in frequency selective radomes. IEEE Antennas and Wireless Propagation Letters, 12, 480-483, 2013.
2013-06-14 Ex-jobbspresentation: Zoraze Ali och Asif Qamar
Zoraze Ali and Asif Qamar are going to present their MSc project, Friday, June 14, 2013, at 13:15, in E:2311.
Title: “3GPP LTE versus IEEE 802.11p/WAVE: Competition or Coexistence?”
The cooperative awareness messages which referred as Beacons in cooperative intelligent transport system enable the cooperative safety applications. These safety applications require strict beacon delay and highly reliable communication of beacons. The technology that already been designed for vehicular communications is IEEE 802.11p/WAVE, this technology has some technical short comings in terms of the beacon delay and beacon loss ratio at congested traffic scenarios. The shortcomings of WAVE open the door to test new wireless communication technology for vehicular communications. The best available commercially launched wireless communication technology right now is 3GPP LTE. As compare to WAVE the LTE is infrastructure based technology. In this thesis, we are going to test the performance of LTE and WAVE in the rural highway traffic scenario. The evaluations will be done on the basis of supporting the strict cooperative vehicular safety applications in terms of beacon delay and beacon lost ratio. The simulations for both LTE and WAVE model for vehicular communications are done in NS-3. The effect of beaconing frequency and beacon size on the network is also analyzed. On the basis of our simulation results the comparative analysis was done between LTE and WAVE and in the end we concluded although the LTE has performed better than LTE but LTE primary responsibility is wireless telephony, therefore it is not wise to completely eradicate WAVE, so the best possible solution for the vehicular safety application supports would be coexistence of LTE and WAVE.
2013-06-13 Ex-jobbspresentation: Steffen Malkowsky och Christoph Müller
Steffen Malkowsky and Christoph Müller will present their master thesis on June 13th at 10:15 in E:2311
Title: "Optimisation of an Automated De-synchronisation Flow"
The vast majority of circuits today, is developed using an synchronous design approach. However, physical designs suer from process, voltage and temperature (PVT) variation, which can lead to malfunction. Moreover, when designing for ultra low-power, supply voltage scaling is applied to operate circuits near the threshold voltage or in subthreshold regime. In this region designs are even more prone to PVT variations. Asynchronous circuits are known to be potentially more resistant against variations, and, therefore, are a promising candidate for asynchronous design. To bypass the drawback of a direct asynchronous implementation, a de-synchronization flow capable of transforming synchronous designs to de-synchronized designs was developed in a 65-nm standard CMOS technology in a former Master's thesis. Based on this work, an optimized de-synchronization Flow was implemented which is presented in this Master's thesis. The optimized flow is applicable for the whole supply voltage range from super-VT down to sub-VT . Area and power overhead are addressed by combining standard cell library (SCL) gates with full custom gates. Moreover, the flow was made portable to other process technologies by introducing a library containing technology dependent data. Additionally, manually denable granularity oers a way for users to choose the best trade-o between elasticity and performance. Area and power improvements are proved by de-synchronizing dierent reference designs. Comparisons of the results of the initial and the optimized de-synchronization flow manifested a reduction of area and power overhead from 105% to 9% and 174% to 58% referred to the synchronous counterparts when operating in sub-VT .
2013-06-13 Ex-jobbspresentation: Pétur Hjartarson och David Dahlin
Pétur Hjartarson and David Dahlin will present their master project
"Protocol Analysis of Ubiquitous Sensor Networks with Health Care Monitoring and Surveillance"
Thursday 13/6 at 13.15 in E:3139
The project has been performed at Securitas Direct AB (Verisure).
As the population grows and people live longer, the demand of health care will increase. Visits to the hospital are exhausting and time consuming, especially for patients with long term diseases. The long term disease patients and elderly take up the largest part of the hospital resources and when the knowledge of sensor networks is improving and the communication channels are getting more reliable, the need for visiting the hospital in order to do a simple measurement can be reduced. Many measurements and even some monitoring activities can therefore be done at home instead of taking time up and resources from the hospitals. Not only can this prove to be bene cial in terms of cost, but also when it comes to produce new services and business models. This thesis is about implementing such a feature in an existing sensor network that is being used mainly for home security. It will present a test implementation, or a proof of concept that it is possible to send arbitrary data at the time being and it will also compare the existing sensor network protocol with other sub gigahertz sensor network protocols. Furthermore the thesis will address a system model based on the standard ISO/IEEE 11073 PHD to make it possible to co-exist with manufacturers of different health devices.
2013-06-12 Ex-jobbspresentation: Jingou Lai
On Wednesday, June 12 at 13.15 in E:2349, will Jingou Lai present his master’s thesis:
“Logarithm Implementation using Improved Parabolic Synthesis”
This thesis develops a design that approximates the fractional part of based two logarithm function by using Improved Parabolic Synthesis. Improved Parabolic Synthesis is a novel methodology in favor of implementing unary functions e.g. trigonometric logarithm, square root etc.in hardware. It is an evolved approach from Parabolic Synthesis by combining it with Second-Degree Interpolation. In the thesis, the design explores a simple and parallel architecture for fast timing and optimizes wordlengths in computing stages for a small design. The error behavior of the design is described and characterized for better reliability. This implementation is compared to other approaches e.g. Parablic Sythesis and CORDIC using 65nm standard cell libraries and it is proved to have better performance in terms of chip area, critical path, and power consumption.
2013-06-13 Ex-jobbspresentation: Tora Dunås och Madeleine Kildal Schilliger
Tora Dunås (Chalmers) and Madeleine Kildal Schilliger will present their master thesis
"Detection of traumatic intracranial bleedings in a realistic brain phantom using microwave technology"
on Thursday 13/6 at 14.15 in E:3139
The thesis is done in collaboration with Medfield Diagnostics. Stefan Candefjord and Stefan Kidborg are the supervisors and I am the examiner.
2013-06-10 Ex-jobbspresentation: Fredrik Stolt och Anders Skoog
Fredrik Stolt and Anders Skoog will present their master thesis on June 10th at 13:15 in E:3139.
The project was done at Böhmer Audio.
Title: Wi-Fi Based Remote Control of an Audio-Platform -A System Approach
2013-06-03--04: EIT hosts 9th Swedish National Computer Networking Workshop
SNCNW, the Swedish National Computer Networking Workshop, will be held at
the Dept. of Electrical and Information Technology (EIT).
The workshop is a creative and informal forum for researchers and practitioners in computer communication and networking.
The main objective of SNCNW is to give all research and development groups an opportunity to present their work.
The workshop also provides an excellent opportunity to discuss ideas among colleagues and establish contact between computer network researchers, practitioners and industry, and thus to help strengthen this field in Sweden.
Location: Room E:1406 in E-building, on the LTH campus.
2013-05-15 Ex-jobbspresentation: Alejandro Vazquez Bofill och Dalia Mihaiela Iurascu
On Wednesday May 15 at 13.15 in E:2349, will Alejandro Vazquez Bofill and Dalia Mihaiela Iurascu present their Master’s thesis entitled:
“Hardware Approximation of the Square Root Function”
You are all welcome!
The aim of this master thesis research work focus on the development of a revolutionary methodology in which Parabolic Synthesis and Second Degree Interpolation methods are used for designing and implementing a novel approximation method of the Square Root function. This algorithm is implemented in VHDL, synthesized and mapped to an ASIC, using 65 nm Standard VT technology library, with respect to timing, chip area and power consumption. This methodology is also suitable for calculating other different unary functions such as trigonometric, logarithmic and division functions and the main advantage of the approach is the simple hardware implementation using just elementary operation like addition, shifting and multiplication. The results reveal that the high proportion of parallelism used in the architecture of the design makes this algorithm implementation feasible for fast computation applications.
2013-05-06--07: 12th Swedish System-on-Chip Conference (SSOCC)
The SSOCC2013 will be hosted by the EIT this year.
The research seniors and PhD students in analog, RF, and digital design from the major Swedish universities gather for this conference. Confirmed invited speakers are Andreas Burg (EPFL, Lausanne, Switzerland), Fredrik Tillman (Ericsson research in Lund), and Ivan Jørgensen (DTU, Denmark)
The purpose of the conference is networking and presentation of current research work.
When: May 6-7
Place: Ystad Saltsjöbad
2013-04-17 Seminar: Francisco Monteiro
On Wednesday Francisco Monteiro from Instituto de Telecomunicações Lisbon will visit me and also give a seminar at 13.15 in E:2349.
Abstract and bio are pasted below.
Approaches to the closest vector problem in MIMO detection
Speaker: Francisco A. Monteiro (IT – Telecommunications Institute / University Institute of Lisbon)
This talk provides an overview of one of the central problems in communication engineering in the last dozen years, whose solution allows us now to reach the 1 Gbps frontier in wireless systems such as LTE Advanced and WiMax. The gains achieved by using multiple-input multiple output (MIMO) transmission links come at the expense of a much higher algorithmic complexity at the receiver side. The underlying detection problem is the closest vector problem (CVP) in a lattice. The talk will describe several approximate and exact solutions to CVP, emphasising the geometric manipulation of lattices that is carried out by the most relevant algorithms (maximum likelihood detection, zero-forcing, minimum mean square error, successive detection, sphere decoding and lattice reduction). A novel approach to the problem (by Monteiro and Kschischang) will also be presented, which maps the problem onto a graph-based path-minimisation problem. Speaker’s biography: Francisco Monteiro obtained the PhD in Engineering at the University of Cambridge (in the Computer Laboratory), UK, in 2011, the ECE degree from Instituto Superior Técnico (IST), Technical University of Lisbon, and a Masters in EEC degree also from IST in 2003. He has been awarded two best paper awards at IEEE conferences (in 2004 and in 2007) and a Innovation Young Engineer Prize (3rd place) from the Portuguese Engineers Institution in 2002. During his PhD he also spent 4 months as a visiting scholar at the ECE department of the University of Toronto, Canada, sponsored by the Royal Academy of Engineering, by the Gulbenkian Foundation, by Fitzwilliam College (Cambridge), and by the Cambridge Philosophical Society. In the past he was a teaching assistant at the ECE department of IST and presently he is a tenure-track Assistant Professor at the Department of Information Science and Technology at ISCTE- University Institute of Lisbon and a researcher in the Radio Systems Group of the Instituto de Telecomunicações. Francisco is keen in public understanding of science and is a member of several societies promoting those objectives. He is co-editing a book on MIMO-related techniques to be published by CRC Press by the end of 2013.
2013-04-10 Seminar: Grigori Kabatiansky
Grigori Kabatiansky, IITP RAS and PremoLab MIPT, Moscow, is currently visiting us. On Wednesday April 10, 9.15-10.00 in E:2311, he will give a seminar.
Title: Basic Facts about Compressed sensing: an overview and slightly beyond.
The term "Compressed sensing" was coined by D. L. Doloho in his article with the same title that appeared in IEEE Transactions on Information Theory (2006). He showed that n-dimensional vectors having few "significant" coordinates can be recovered from a small number of measurements. The topic was immediately pursued by E. J. Candes and Terence Tao (2006 Fields Medal), who showed the connection between this problem and the well-know problem on almost isometric embedding of normed spaces, and the work commenced. This topic become particularly popular because of the fact that one of its principal ingredients was optimization in L1, which has already become popular in various applications (thanks in particular to the LASSO method). We will provide an accessible review of the main results in this field and established some links with its discrete analogs.
2013-04-05 Ex-jobbspresentation: Viktor Liljegren
Viktor Liljegren will present his master thesis with the title “Evaluation of Circular Polarization Selective Surfaces for Space Applications”.
Time and location are
Friday April 5, at 9.15 in E:2349.
I have been the supervisor and Mats Gustafsson is the examiner. The work is related to an application we have recently made to ESA together with RUAG Space in Göteborg, and TU Eindhoven. The abstract is given below.
For space applications a circular polarization equivalent of a Dual Gridded Reﬂector (DGR) would be desirable. Such a structure could be created using Circular Polarization Selective Surfaces (CPSS). In this thesis, various CPSS designs are examined and evaluated for this purpose. The surfaces are evaluated based on how wide a bandwidth can be achieved while staying within requirements on Insertion Loss (IL), Return Loss (RL) and Axial Ratio (AR) as previously used by the European Space Agency (ESA). The report covers background information on circular polarization, scattering parameters and general CPSS notation that is needed to understand the work as well as more information on the structure requirements. Previously proposed CPSS concepts by Pierrot, Tilston and Morin are explained and a few diﬀerent implementations based on these concepts are simulated. The surfaces are designed at 10 GHz and simulated for 8−12 GHz. Each of the structures are simulated for normal incidence as well as for oblique incidence at up to 30 degrees. For the oblique incidence various incidence planes are also examined. The result for each of the simulated structures is presented. The results are summarized and based on the results it is concluded that the CPSS structures have quite a small bandwidth limited mainly by the AR and that for oblique angles greater than 10-20 degrees the performance of most of the structures is greatly reduced.
2013-04-05 Seminar: Professor Graeme Milton
Professor Graeme Milton, University of Utah, visits the Electromagnetic Theory Group on Thursday and Friday this week, and he gives a seminar on Friday.
Speaker: Distinguished Professor Graeme Milton, Department of Mathematics , University of Utah, USA
Title: Non-linear metamaterials from rods and hinges
A complete characterization is given of the possible macroscopic deformations of periodic nonlinear affine unimode metamaterials constructed from rigid bars and pivots. The materials are affine in the sense that their macroscopic deformations can only be affine deformations: on a local level the deformation may vary from cell to cell. Unimode means that macroscopically the material can only deform along a one dimensional trajectory in the six dimensional space of invariants describing the macroscopic deformation (excluding translations and rotations). We show by explicit construction that any continuous trajectory is realizable to an arbitrarily high degree of approximation provided at all points along the trajectory the geometry does not collapse to a lower dimensional one. In particular, we present two and three dimensional dilational materials having an arbitrarily large flexibility window. These are perfect auxetic materials for which a dilation is the only easy mode of deformation. They are free to dilate to arbitrarily large strain with zero bulk modulus.
2013-04-05 Seminar: Prof. Koichi Ito
A seminar by Prof. Koichi Ito of Chiba University, Chiba Japan, partly sponsored by the Communication Technology Portal at Lund University.
Title: Antennas for Body-Centric Wireless Communications
Date: Friday, 5 April 2013.
Time: 10.15 (approx. 1 hour + questions)
Buon Kiong Lau
Recently, a study on body-centric wireless communications has become an active and attractive area of research because of their various applications such as e-healthcare, support systems for specialized occupations, monitoring systems for elderly and handicapped people, entertainment, and so on. Whereas UHF bands are subjects of interest especially in Europe and USA, HF bands are of great interest especially in Japan. Hence, all of the prospective frequencies are in an extremely wide range, and an objective idea on how to select a right frequency band for individual applications is required. As for the antennas, many types of wearable (on-body) and implantable (in-body) antennas have been reported.
Currently in our laboratory, we have been studying on frequency dependence of basic characteristics of simple wearable antennas as well as body-centric wireless communication channels in the range of HF to UHF (3 MHz – 3 GHz).
In this presentation, firstly, a brief history of body-centric wireless communications is reviewed. Secondly, electric field distributions around the human body wearing a small top-loaded monopole antenna are numerically calculated and compared in a wide range of HF to UHF bands. Then, received open voltages at receiving antennas which are equipped at several different points on the human body are numerically investigated. The received open voltages are also numerically calculated and compared with several different postures of the human body. Finally, a few examples of practical wearable antennas are introduced including a dual-mode wearable antenna which can be used at 10 MHz (on-body) as well as at 2.45 GHz (off-body).
Koichi Ito was born in Nagoya, Japan, and received the B.S. and M.S. degrees from Chiba University, Chiba, Japan, in 1974 and 1976, respectively, and the D.E. degree from Tokyo Institute of Technology, Tokyo, Japan, in 1985, all in electrical engineering. From 1976 to 1979, he was a Research Associate at the Tokyo Institute of Technology. From 1979 to 1989, he was a Research Associate at Chiba University. From 1989 to 1997, he was an Associate Professor at the Department of Electrical and Electronics Engineering, Chiba University, and is currently a Professor at the Department of Medical System Engineering, Chiba University. From 2005 to 2009, he was Deputy Vice-President for Research, Chiba University. From 2008 to 2009, he was Vice-Dean of the Graduate School of Engineering, Chiba University. Since April 2009, he has been appointed as Director of Research Center for Frontier Medical Engineering, Chiba University. In 1989, 1994, and 1998, he visited the University of Rennes I, France, as an Invited Professor. He has been appointed as Adjunct Professor to the University of Indonesia since 2010.
His main research interests include analysis and design of printed antennas and small antennas for mobile communications, research on evaluation of the interaction between electromagnetic fields and the human body by use of numerical and experimental phantoms, microwave antennas for medical applications such as cancer treatment, and antenna systems for body-centric wireless communications.
Professor Ito is a Fellow of the IEEE, a Fellow of the Institute of Electronics, Information and Communication Engineers (IEICE) of Japan, a member of the American Association for the Advancement of Science, the Bioelectromagnetics Society (BEMS), the Institute of Image Information and Television Engineers of Japan (ITE) and the Japanese Society for Thermal Medicine. He served as Chair of the Technical Group on Radio and Optical Transmissions, ITE, from 1997 to 2001, Chair of the Technical Committee on Human Phantoms for Electromagnetics, IEICE, from 1998 to 2006, Chair of the Technical Committee on Antennas and Propagation, IEICE, from 2009 to 2011, Chair of the IEEE AP-S Japan Chapter from 2001 to 2002, Vice-Chair of the 2007 International Symposium on Antennas and Propagation (ISAP2007), General Chair of the 2008 IEEE International Workshop on Antenna Technology (iWAT2008), Co-Chair of ISAP2008, an AdCom member for the IEEE AP-S from 2007 to 2009, an Associate Editor for the IEEE Transactions on Antennas and Propagation from 2004 to 2010, a Distinguished Lecturer for the IEEE AP-S from 2007 to 2011, and General Chair of ISAP2012. He currently serves as Chair of the IEEE AP-S Committee on Man and Radiation (COMAR), a member of the Board of Directors, BEMS, and a Councilor to the Asian Society of Hyperthermic Oncology (ASHO). He has been appointed as a member of the IEEE Life Sciences New Initiative (LSNI) Project Team. He has been elected as a delegate to the European Association on Antennas and Propagation (EurAAP) since 2012.
2013-03-27 Seminar: Dr. Goutam Paul
Dr. Goutam Paul is visiting the Crypto and Security group during two weeks. He will give a seminar.
Time and location: Wednesday March 27, 15.15-16 in E:3139
Title: "Biases and Distinguishers of the RC4 Stream Cipher"
Speaker: Goutam Paul
Stream cipher, a symmetric key cryptography primitive, is essentially a pseudo-random number generator (PRNG) that generates a long stream of data (called keystream) based on a short seed (the secret key). Till date, RC4 remains the most popular and widely deployed byte-oriented software stream cipher despite the existence of many short-term and a few long-term biases involving the secret key, the internal state and the keystream. Amongst these, the biases involving only the keystream lead to some interesting distinguishers. In this talk, we discuss the most important biases and distinguishers of RC4. Short bio of speaker: Dr. Goutam Paul did his Ph.D. at the Cryptology Research Group (CRG) of Indian Statistical Institute, Kolkata in 2009. Currently, he is an Assistant Professor in the Department of Computer Science & Engineering of Jadavpur University, Kolkata, India, and visiting RWTH Aachen, Germany as an Alexander von Humboldt Fellow. Apart from design and cryptanalysis of stream ciphers, his other research interests include steganography, privacy preservation for location-based services and quantum cryptography.
2013-03-21 Seminar: Fredrik Brännström
Fredrik Brännström from Chalmers will visit EIT and give a seminar on labelings for coded and uncoded modulations. The seminar is on March 21 at 10.15 in E:3139.
"Labelings for Coded and Uncoded Modulations – Bit-Error Rate, Mutual Information, and Design"
You are all welcome to attend the seminar!
Coded modulation is today used in most communication systems and it has been studied for a long time by many researchers. However, until recently, fundamental properties like the analytical expression of the uncoded bit-error rate (BER) and the behavior of the mutual information (MI) in the high-SNR regime were not known – even for simple constellations and labelings. In this tutorial talk, some closed-form expressions for the uncoded BER for 8PAM is presented, which before was only known for 4PAM. Newly developed asymptotic expressions for the MI of one-dimensional constellations in the high-SNR regime are also presented. These expressions are then used to formally prove the long standing conjecture that Gray codes are the binary labelings that maximizes the MI for bit-interleaved coded modulation. Finally, an approach to find optimal TCM schemes are presented that minimizes the BER and FER in contrast to Ungerboeck’s TCM schemes, which only optimized the minimum Euclidean distance.
Fredrik Brännström is an Assistant Professor at the Department of Signals and Systems, Chalmers University of Technology, Gothenburg, Sweden. He received his Ph.D. degree in communication theory from the Department of Computer Engineering, Chalmers, in 2004. From 2004 to 2006, he had a Postdoctoral position in the Communication Systems Group at the Department of Signals and Systems, Chalmers. From 2006 to 2010 he was a Senior Algorithm Engineer and Principal Design Engineer at Quantenna Communications, Fremont, CA. At Quantenna he was part of the system team that developed the PHY layer for the world’s first WiFi 802.11n chip with 4x4 MIMO using dynamic digital beamforming together with LDPC codes operating at 600 Mbps. His research interests in communication theory and information theory include code design, coded modulation, bit-to-symbol mappings, and efficient iterative processing, as well as MIMO signal processing and algorithm design for vehicular communication systems.
2013-03-21 Project: Algorithms in Signal Processors
On Thursday March 21th, at 15.15-17 the projects in our project course Algorithms in Signal Processors ETIN80 will be presented and
demonstrated in E:2311. The projects of this year have been related to audio processing on Texas Instruments DSP C6713; the students have implemented a vocoder effect, a karaoke effect, pitch correction, sound detection, echo cancellation and speech recognition.
You are all welcome to attend!
2013-03-15 Professorsinstallation, Björn Landfeldt
Björn Landfeldt installeras som professor vid Lunds universitet fredagen den 15 mars, i Universitetshusets aula.
Installeras. Björn Landfeldt är en av LU:s nyaste professorer.
Följande text är hämtad från ett häfte som ges ut i samband med professorsinstallationen och som kan läsas i sin helhet här.
Att vara ständigt uppkopplad mot “nätet” är en självklarhet för många idag, men det var inte så länge sedan endast ett fåtal visste om att det fanns ett Internet. Den explosion av dataanvändning vi nu ser kommer också att medföra att de sätt vi bygger trådlösa nät på inte klarar av datamängden, och det är nödvändigt att para ihop olika teknologier som inte alls fungerar på samma sätt och få dem att samarbeta. I den trådlösa framtiden kommer även fler och fler ting och produkter att kunna kommunicera med varandra, eftersom det nästan alltid finns fördelar att dra av informationsutbyte och samarbete.
Idag vet vi väldigt lite om hur vi skall konstruera system som klarar av att hantera tusentals eller miljoner gånger fler användare än dagens Internet, speciellt eftersom olika framtida teknologier kommer att fungera väldigt olika. Ett viktigt nytt sätt att bygga dessa system är genom att låta alla deltagare samarbeta och gemensamt hitta den bästa lösningen för tillfället. Nu arbetar jag med grunderna för att förstå hur detta ska gå till.
Samtidigt kommer denna trådlösa revolution att medföra att olika system som kan kommunicera kommer att fungera som en kombination av ett fysiskt system och ett “cybersystem”. Till exempel kommer trådlösa nätverk mellan alla bilar att medföra att bilar (förare) kan samarbeta för att förkorta köer eller att undvika kollisioner. Ett stort problem är dock att själva bilsystemet, hur bilarna rör sig, kraftigt påverkar hur det trådlösa nätverket fungerar. Vi behöver komma fram till nya teorier om hur denna kombination av system fungerar för att kunna förstå dem bättre och utveckla nya tjänster.
När allt pratar med allt blir möjligheterna svindlande, men vi måste först lägga den vetenskapliga grunden för att förstå hur vi ska bygga systemen.
Foto: Johan Cedervall
PhD thesis defence Adnan Prlja
Date: 2013-02-28, at 10:15
Place: Lecture hall E:1406, E-building, John Ericssons väg 4, Lund University Faculty of Engineering
Opponent: Professor Joachim Hagenauer. Technical University of Munich (TUM), Germany
Reduced Receivers for Faster-than-Nyquist Signaling and General Linear Channels
Series of licentiate and doctoral theses, Vol. no. 48, ISSN 1654-790X, Tryckeriet i E-huset, Lund, 2013.
Abstract: Fast and reliable data transmission together with high bandwidth efficiency are important design aspects in a modern digital communication system. Many different approaches exist but in this thesis bandwidth efficiency is obtained by increasing the data transmission rate with the faster-than-Nyquist (FTN) framework while keeping a fixed power spectral density (PSD). In FTN consecutive information carrying symbols can overlap in time and in that way
introduce a controlled amount of intentional intersymbol interference (ISI). This technique was introduced already in 1975 by Mazo and has since then been extended in many directions.
Since the ISI stemming from practical FTN signaling can be of significant duration, optimum detection with traditional methods is often prohibitively complex, and alternative equalization methods with acceptable complexity-performance tradeoffs are needed. The key objective of this thesis is therefore to design reduced-complexity receivers for FTN and general linear channels that achieve optimal or near-optimal performance. Although the performance of a detector can be measured by several means, this thesis is restricted to bit error rate (BER) and mutual information results. FTN signaling is applied in two ways: As a separate uncoded narrowband communication system or in a coded scenario consisting of a convolutional encoder, interleaver and the inner ISI mechanism in serial concatenation. Turbo equalization where soft information in the form of log likelihood ratios (LLRs) is exchanged between the equalizer and the decoder is a commonly used decoding technique for coded FTN signals.
The first part of the thesis considers receivers and arising stability problems when working within the white noise constraint. New M-BCJR algorithms for turbo equalization are proposed and compared to reduced-trellis VA and BCJR benchmarks based on an offset label idea. By adding a third low-complexity M-BCJR recursion, LLR quality is improved for practical values of M. M here measures the reduced number of BCJR computations for each data symbol. An improvement of the minimum phase conversion that sharpens the focus of the ISI model energy is proposed. When combined with a delayed and slightly mismatched receiver, the decoding allows a smaller M without significant loss in BER.
The second part analyzes the effect of the internal metric calculations on the performance of Forney- and Ungerboeck-based reduced-complexity equalizers of the M-algorithm type for both ISI and multiple-input multiple-output (MIMO) channels. Even though the final output of a full-complexity equalizer is identical for both models, the internal metric calculations are in general different. Hence, suboptimum methods need not produce the same final output. Additionally, new models working in between the two extremes are proposed and evaluated. Note that the choice of observation model does not impact the detection complexity as the underlying algorithm is unaltered.
The last part of the thesis is devoted to a different complexity reducing approach. Optimal channel shortening detectors for linear channels are optimized from an information theoretical perspective. The achievable information rates of the shortened models as well as closed form expressions for all components of the optimal detector of the class are derived. The framework used in this thesis is more general than what has been previously used within the area.
2013-02-27 Ex-jobbspresentation: Robin Larsson
Robin Larsson will present his master thesis "Detection of user patterns in live streaming media
during the London Olympic Games" on Wednesday 27/2 at 15.15 in E:2349.
Robin has analysed data from the SVT Play service. The project has been performed within the EFRAIM project in collaboration with Acreo AB.
You are all welcome to attend!
As the internet continues to grow so does the need for smarter networks that can cater the ever-growing amount of traffic. Streaming media has in just a few years become a major part of the internet's data flow, but the distribution of streaming media is still done in the same, inefficient way traditional WWW content is distributed. This leads to a potential decrease in both quality of service and quality of experience. In order to make networks smarter, it is important to make them able to understand their users. By dividing users into smaller clusters based on their viewing preferences it will be easier to predict which content will become popular, which in turn can be used for increasing the users' quality of experience e.g. by caching the content in a closely located distribution server. In this thesis, an analysis of the Swedish state-owned streaming video service SVT Play is done based on data traffic collected in a Swedish city during the 2012 London Olympics (2012-07-27 -- 2012-08-12). More general analyses to understand user behaviour are done separately for general SVT Play data and data related to the Olympic Games. Methods for clustering users on a content access basis are then suggested.
2013-02-22 Seminar: Nathan Woollett
Electromagnetic Theory research seminar: Nathan Woollett, Dept Physics, Lancaster University UK.
Title: Hidden Sector Photon Searches
Friday February 22 at 15.15 in E:2349.
2013-02-19: Azzam Al-Nahari
Tomorrow afternoon, 19/2 2013, our Erasmus PostDoc, Azzam Al-Nahari, will give a seminar on physical layer security.
Title: Some techniques for enhancing the physical layer security
Place: E:3139 (E-building, Ole Römers väg 3)
Physical layer security has recently attracted a considerable attention in the context of information-theoretic point of view. The goal of physical layer security is to allow the source to confidentially communicate with the legitimate destination while the eavesdropper cannot interpret the source information, without any assumption for the eavesdropper nodes such as computational power and available information. In this presentation, I will give an introduction to the subject, and present some of the results of our research in this area. There are two main parts that will be considered. First, the application of cooperative communication techniques to enhance the physical layer security. Secondly, the application of the massive MIMO for detection of the active eavesdropper in the network.
2013-02-12 Leif Willhelmsson: In-Device Multi-Standard Co-Existence
With the introduction of LTE, two new frequency need to be supported in mobile terminals, namely Band 40 (2.3-2.4 GHz) intended for TDD and Band 7 (2.5-2.57 GHz (UL) + 2.62-2.96 GHz (DL)) intended for FDD.
This presentation will be given as part of the ISSCC Forum Advanced RF Transceiver Design Techniques in San Francisco, www.isscc.org ISSCC is the flagship conference of the IEEE Solid-State Circuits Society, and is the premier forum for the presentation of advances in solid-state circuits and systems-on-a-chip.
Abstract: With the introduction of LTE, two new frequency need to be supported in mobile terminals, namely Band 40 (2.3-2.4 GHz) intended for TDD and Band 7 (2.5-2.57 GHz (UL) + 2.62-2.96 GHz (DL)) intended for FDD.
This poses a real coexistence challenge as Bluetooth and WiFi, supported in virtually every mobile terminal, use the ISM band (2.4-2.4835 GHz). Consequently, there is no guardband between Band 40 and the ISM band, and only 16.5 MHz guardband between the ISM band and the DL for Band 7. Several use cases, however, require simultaneous operation of these different technologies.
This talk will describe state-of-the art filter technologies and analyze how effects like compression, TX noise, reciprocal mixing, and intermodulation impacts performance.
The talk will also give an overview of the standardization work done within Bluetooth and 3GPP to address the problem.
2013-02-07 Cake Seminar with Rebecca Seviour
The first Cake Seminar of the spring will take place Thursday February 7th. The title of the seminar is ”Particle Accelerators".
Speaker is Lise Meitner-professor Rebecca Seviour and her abstract is:
There are over 30,000 particle accelerators in use across the work today, yet many people are unaware of how vital this technology is to our lives, and how particle accelerators can help address many of the problems faced by society today.
In this talk I present an overview of how particle accelerators work, current applications and possible future application of particle accelerators. In addition I will outline some of the challenges that need to be addressed to enable accelerators to solve some of the biggest issues facing humanity over the next 50 years.
As usual, we start at 14.06 in room E1406.
The board of the E-building greets welcome!
2013-01-31 Ex-jobbsprestentation: Jolina Haberkamm
On Thursday Jan 31 at 13.15 in E:3139, Niclas Unnervik and Jolina Haberkamm will present
their master project on
"Community clustering analysis using Facebook content demand patterns".
The project has been performed in cooperation with Acreo AB within the projects IPNQSIS and EFRAIM.
You are all welcome to attend!
Community clustering analysis using Facebook content demand patterns
In order to increase Quality of Service and Quality of Experience in mobile and residential broadband access networks it is important to understand user pat- terns. This understanding could make the implementation of cost effective per- formance optimizing functionality possible rather than expanding the networks physical layer. This thesis aims to develop and evaluate methods for determining Facebook user behavior based on Facebook content demand patterns. The problem is ap- proached by analyzing data from two large municipal network dumps performed on different geographic locations, during different time spans. This study shows that community clustering using content demand patterns is possible and that users with similar user behavior can be grouped together in user clusters. This is supported by a highly configurable Trust-function that build a graph G = (V, E) where E denotes similarity between the connected users and the clustering algorithm Chinese Whispers. It was also discovered that different user activities and certain terminal usage patterns are strongly correlated, showing that user patterns differs between devices. Geographical location is discarded as a parameter for potential future methods since all results from the two networks are highly aligned and no deviation in user behavior is detected due to location.
2013-01-28 Ex-jobbspresentation: Joel Andersson
Joel Andersson will present his master's thesis entitled Electron gun investigations at MaxIV on Monday, Jan 28, 10.15 in E2349.
Two electron guns for the MAX IV laboratory were investigated, one thermionic gun and one photocathode gun. The thermionic gun will be the electron source for the new 3.0 GeV and 1.5 GeV synchrotron rings at MAX IV. The photo- cathode gun was a prototype of the gun that will be used for the short pulse facility that is under construction.Investigations were done with regards to cavity parameters and the electric eld and both structures were simulated using Supersh and Comsol multi- physics. The simulated results were compared with measurements of the struc- tures. It was found that the simulated electric elds in both guns agree well to the measured eld. Based on the simulated electric elds, the eld on the cathode surfaces was analysed. Using the operating data for the cathodes, the electron emission from the cathodes was investigated with regard to thermionic emission in the thermionic gun and with regards to dark current in the photocathode gun. Particle tracking was done for the thermionic gun. No measurement data was available for comparison at the time of the thesis.
2013-01-18 Ex-jobbspresentation: Andreas Ericsson
On Friday 18/1 at 10.15 in E:2349, Andreas Ericsson will present his master thesis entitled
"Self-Supported Hat Feed Reflector Antenna for SATCOM Applications"
The work was performed in collaboration with TICRA in Denmark.
Supervisors: Cecilia Cappellin (TICRA) and Mats Gustafsson
Examiner: Daniel Sjöberg
The hat feed antenna is a rear radiating self-supported compact reflector antenna commonly used for SATCOM (satellite communications) applications. In this thesis the fundamental hat feed design presented in 1987 is studied analytically. A literature survey is performed of all hat feed reflector antenna designs presented up to date. Finally different hat feed antenna designs are implemented and optimized in CHAMP, a mode matching/method of moments optimization software developed by TICRA. The same improvements of the hat antenna performance as have been presented in the last 25 years are observed when comparing the results of different implemented designs. By adding more degrees of freedom to the feed design the cross polarization component of the radiated signal is reduced and the return loss bandwidth is drastically increased. The best antenna design optimized in CHAMP just about fulfils radiation envelopes defined by FCC for satellite communication in the KU-band at the frequencies of 11.5-12.5 GHz. This design have a return loss bandwidth of 18% and an aperture efficiency of 69.4%. It is concluded that CHAMP is a powerful reflector antenna optimization tool and that further optimizations can most likely produce an antenna compliant with the high standards of satellite communication in a wider bandwidth.
2013-01-17 Ex-jobbspresentation: Olof Olsson
Olof Olsson will present his master's thesis titled: "Flexible software solution for automatic maintenance of thick-client kiosk systems"
on Thursday Jan 17 10:15 in E:3139.
Thesis background Many kiosk systems are implemented with so called thin-clients. This makes them fairly easy to update since the application is stored on a central server. When on the other hand a kiosk system need to access local hardware a thick-client based system must be used. To keep control over thick client systems, it is necessary to automate the update and control procedure with software. This software should provide an easy overview of all systems and an accessible way to remotely interact with them. The software should also be able to update and control the behavior of unique installations. The software needs to be flexible in its design to allow support for multiple system types.
Thesis purpose The purpose of this project is to propose a design for a solution to enable costeffective maintenance and control of thick-client kiosk systems. The design will be verified by implementing and setting up a proof-of-concept application for a specific case.
• Find the requirements and necessary algorithms for a software that will solve the problem of updating and controlling thick-client kiosk systems.
• Implement and setup a proof-of-concept solution.
• Evaluate the implemented solution and propose improvements if necessary.