Interuniversity Micro Electronics Center

Leuven, Belgium

Interuniversity Micro Electronics Center

Leuven, Belgium
SEARCH FILTERS
Time filter
Source Type

Bellanger P.,University of Lisbon | Brito M.C.,University of Lisbon | Pera D.M.,University of Lisbon | Costa I.,University of Lisbon | And 4 more authors.
IEEE Journal of Photovoltaics | Year: 2014

The SLIM-cut technique provides a way to obtain thin silicon foils without a standard sawing step, thus avoiding kerf losses. This process consists of three steps: depositing a stress-inducing layer on top of the silicon surface; stress activation by heating and cooling, resulting in crack propagation in the silicon and detachment of a thin silicon layer; and a chemical cleaning to remove the stress-inducing layer. This paper describes a new stress activation method using Ag/Al and epoxy stress-inducing layers. The crack propagation is controlled along the sample length in order to avoid unwanted additional crack formation and interaction with other crack fronts. Silicon foils with thickness ranging between 50 and 130 μm were obtained with effective lifetimes between 1 and 81 μs. © 2014 IEEE.


Stucchi M.,Interuniversity Micro Electronics Center | Perry D.,Qualcomm | Katti G.,Institute of Microelectronics, Singapore | Dehaene W.,Catholic University of Leuven | Velenis D.,Interuniversity Micro Electronics Center
IEEE Transactions on Semiconductor Manufacturing | Year: 2012

3-D chip stacking using through-silicon vias (TSVs) requires accurate characterization of the TSV, the thinned silicon, and the stacked dies. This paper proposes a set of test structures specifically designed to address the electrical characterization of TSV in terms of resistance, capacitance, leakage, yield, and their impact on the 2-D interconnects of the stacked dies. Examples of the use of these structures are presented, and the observed electrical behaviors are explained with the support of FIB cross-section images. © 1988-2012 IEEE.


Vanwinckelen G.,Catholic University of Leuven | Van Otterlo M.,Catholic University of Leuven | Driessens K.,Maastricht University | Pollin S.,Interuniversity Micro Electronics Center
2011 IEEE International Symposium on Dynamic Spectrum Access Networks, DySPAN 2011 | Year: 2011

An important aspect of spectrum sharing is reliable protection of licensed, primary, users from interference by unlicensed, secondary, users. In this paper we investigate the reliability of the iterative power adjustment algorithm proposed by Pollin, Adams and Bahai (2008). The goal of this transmission power control algorithm is to allow a static secondary transmitter to maximize its power without interfering with primary users. A distributed flooding algorithm is used to detect primary users and estimate the distance to the primary propagation contour. The secondary transmitter makes a local channel estimation with a moving least squares algorithm to average out noise. The metric used to estimate interference is the propagation contour-contour distance between the secondary and primary transmitters. In our first contribution we investigate the reliability of this metric by computing the location probability, a new FCC proposed metric for configuring Digital Terrestrial Television networks. We show that the propagation contour-contour distance is correlated with the location probability. In a second contribution we make the flooding algorithm more cost efficient by reducing communication. We then study the influence of the number of flooding messages on the performance of the iterative power adjustment algorithm in terms of location probability and number of iterations. © 2011 IEEE.


Horlin F.,Roosevelt University | Fickers J.,Roosevelt University | Emplit P.,Roosevelt University | Bourdoux A.,Interuniversity Micro Electronics Center | Louveaux J.,Catholic University of Louvain
Optics Express | Year: 2013

In order to improve the spectral efficiency of coherent optical communication systems, it has recently been proposed to make use of the orthogonal frequency-division multiplexing offset quadrature amplitude modulation (OFDM-OQAM). Multiple optical channels spaced in the frequency domain by the symbol rate can be transmitted orthogonally, even if each channel overlaps significantly in frequency with its two adjacent channels. The solutions proposed until now in the literature unfortunately only address a single polarization communication, and therefore do not benefit from the capacity gain reached when two polarizations are used to transmit independent information signals. The aim of the present paper is to propose a receiver architecture that can decouple the two polarizations. We build an equalizer per channel at twice the symbol rate and optimize it based on the minimum mean square error (MMSE) criterion. We demonstrate the efficiency of the resulting system compared to the Nyquist wavelength-division multiplexing (N-WDM) system both in terms of performance and complexity. We also assess the system sensitivity to transmit synchronization errors and show that system can even work under significant synchronization errors. © 2013 Optical Society of America.


Redant T.,Catholic University of Leuven | Dehaene W.,Catholic University of Leuven | Dehaene W.,Interuniversity Micro Electronics Center
IEEE Transactions on Circuits and Systems II: Express Briefs | Year: 2013

This brief presents an approach to compensate time-of-arrival (ToA) estimation errors caused by the effect of signal depending propagation delays in receivers for ranging applications. As a straightforward reasoning, reducing the ToA error results in increased power consumption in the analog-frontend blocks of such receivers. In the proposed system topology, the asynchronous receiver's analog-front-end blocks contain an array of identical continuous-time comparators, which makes it suitable for a joint modeling approach and today's low-supply voltages. Attention is paid to tuning the nonstrobed comparator topology to make it suitable for this joint modeling. Using the comparator model in a least-squares-based algorithmic approach can reduce the ToA error (accuracy improvement) to nearly zero. A ToA systematic error reduction of 28 is faced for the ToA figure. Furthermore, energy consumption is more related to the activity of the ranging events, which makes this an interesting approach for ranging measurement rates of less than 1 MHz. © 2013 IEEE.


Lee B.G.,IBM | Schow C.L.,IBM | Rylyakov A.V.,IBM | Van Campenhout J.,IBM | And 9 more authors.
Journal of Lightwave Technology | Year: 2011

A custom 90-nm bulk digital CMOS switch driver is codesigned and integrated with a silicon photonic switch. A photonic device model is created within the electronic design environment, facilitating driver optimization and performance evaluation prior to fabrication. The fabricated drivers implemented in two variations produce transition times as low as 50 ps and generate open eye diagrams using supply voltages ranging from 0.8 to 5 V. The driver is hybrid integrated with a broadband low-power silicon photonic 2 × switch, based on a modified Mach-Zehnder interferometer. The switch has demonstrated operation over 100 nm of spectral bandwidth with less than-17-dB crosstalk, greater than 25 °C tolerance to temperature variations, and compatibility with 1-V driving signals. The integration demonstrates the interoperability of the switch and driver, which together achieve transition times below 4 ns and average power consumption of 2 mW. Finally, throughput bandwidth of 160 Gb/s is demonstrated for all switch configurations via eye diagrams and bit error rate curves. © 2011 IEEE.


Timmers M.,Interuniversity Micro Electronics Center | Timmers M.,Catholic University of Leuven | Pollin S.,Interuniversity Micro Electronics Center | Pollin S.,University of California at Berkeley | And 5 more authors.
IEEE Transactions on Vehicular Technology | Year: 2010

A cognitive radio (CR) network should be able to sense its environment and adapt communication to utilize the unused licensed spectrum without interfering with licensed users. In this paper, we look at CR-enabled networks with distributed control. As CR nodes need to hop from channel to channel to make the most use of the spectrum opportunities, we believe distributed multichannel medium access control (MAC) protocols to be key enablers for these networks. In addition to the spectrum scarcity, energy is rapidly becoming one of the major bottlenecks of wireless operations and has to be considered as a key design criterion. We present here an energy-efficient distributed multichannel MAC protocol for CR networks (MMAC-CR). Simulation results show that the proposed protocol significantly improves performance by borrowing the licensed spectrum and protects primary users (PUs) from interference, even in hidden terminal situations. Sensing costs are evaluated and shown to contribute only 5% to the total energy cost. © 2009 IEEE.


Redant T.,Catholic University of Leuven | Daniels J.,Catholic University of Leuven | Steyaert M.S.J.,Catholic University of Leuven | Dehaene W.,Catholic University of Leuven | Dehaene W.,Interuniversity Micro Electronics Center
IEEE Transactions on Circuits and Systems I: Regular Papers | Year: 2011

A pulse digitizing approach for time-of-arrival pulse radio based ranging is introduced. It is based on a bank of time-to-digital converter (TDC) cores. A comparator bank triggers these multiple TDCs. This multiple event approach has advantages over classic single TDC solutions when facing unknown channel gains, noise corruption, and strong fading channel behavior. Pulses are digitized in a way that is superior in terms of performance versus power to classic A/D conversion. A power effort figure ξ and a new SNDR metric are introduced, easing performance comparison of pulse digitizers. A low power 8 channel digitizing system with a resolution of δtring=62.5ps is presented for a cm accurate ranging application. The asynchronous, event-based nature of the architecture requires nonstrobed comparators to fire value crossing events. A dynamic range of 800:1 is realized. The digitization device is designed for 130 nm standard CMOS. An analog-baseband front-end I-Q energy detection and comparator threshold level configuration D/As are added to the design. The complete system is designed to consume 4 mW. © 2011 IEEE.


Daniels J.,Catholic University of Leuven | Dehaene W.,Catholic University of Leuven | Dehaene W.,Interuniversity Micro Electronics Center | Steyaert M.S.J.,Catholic University of Leuven | Wiesbauer A.,Infineon Technologies
IEEE Transactions on Circuits and Systems I: Regular Papers | Year: 2010

An analog-to-digital conversion (ADC) scheme based on asynchronous ΔΣ modulation and time-to-digital conversion is presented. An asynchronous ΔΣ modulator translates the analog input to an asynchronous duty-cycle modulated signal. Next, the edge locations are digitally measured using a time-to-digital converter (TDC). This information is then digitally processed into a conventional digital signal. The performance of this novel ADC scheme is theoretically analyzed and verified with simulations. With the proposed digital demodulation algorithm, 11-bit resolution can be obtained with an overcycling ratio (OCR) of only four, which is suitable for high bandwidth applications such as very high bit-rate digital subscriber line (VDSL). When a higher OCR can be tolerated, a gated ring-oscillator (GRO) TDC with an inherent first-order noise shaping property is suggested in combination with a digital continuous-time moving-average (CTMA) filter. This allows for resolutions in excess of 13 bits, which is suitable for ADSL2+. The proposed technique shifts the complexity toward the digital domain, leading to more compact ADC and reduced power consumption, and is, therefore, particularly suited for ADC in ultralow-voltage nanometer technologies that are used for high-speed data communication applications. © 2006 IEEE.

Loading Interuniversity Micro Electronics Center collaborators
Loading Interuniversity Micro Electronics Center collaborators