Personal tools
Home People Industry Industry Relations Annual Research Review 2011 ARR Abstracts 2011 ARR Abstracts

2011 ARR Abstracts

Technical Program

  1. New Solutions to Communication Problems
  2. Emerging Electromagnetic Techniques
  3. New Trends in Power-Aware Design
  4. Innovative Devices
  5. Next-Generation CAD for Circuit Design
  6. Novel Circuits for Cognitive Radios
  7. Emerging Sensor and Wireless Networks
  8. Novel Antennas
  9. Advanced Algorithms for Speech and Signal Processing
  10. High-Performance Oscillators and PLLs
  11. Innovative Analog and RF Design
  12. Next-Generation Communications
  13. Novel Photonic and High-Energy Techniques
  14. High-Performance Coding Applications

Session 1: New Solutions to Communication Problems

"Secure State-Estimation for Dynamical Systems under Active Adversaries"
Hamza Fawzi, Paulo Tabuada, and Suhas Diggavi 
Abstract: We consider the problem of state-estimation of a linear dynamical system when some of the sensor measurements are corrupted by an adversarial attacker. The errors injected by the attacker can be arbitrary and do not follow a particular model. We first characterize the maximum number of errors that can be corrected by any decoding algorithm. We then propose a specific computationally feasible decoding algorithm and we characterize the number of errors it can correct. For this, we use ideas from compressed sensing and error correction over the reals. We show using examples that the decoder performs very well in practice. 
"On Information Sharing and the Predicament of Privacy" 
Supriyo Chakraborty and Mani B. Srivastava
Abstract not available
"To the New frontier of Multi-User Communications - an Approach Based Repeated Games with Intervention" 
Yuanzhang Xiao and Mihaela van der Schaar 
Abstract: There have been extensive works that modeled communication scenarios with selfish users as one-shot games and proposed incentive schemes to achieve desirable point on the Pareto boundary as equilibrium. However, in many communication scenarios, the Pareto boundary of the one-shot games is inefficient. In this talk, we propose a generalized repeated game model with intervention to enlarge the set of equilibrium payoffs and loosen the conditions under which it can be achieved. We study the protocol design problem of maximizing the social welfare subject to QoS guarantees, and show the huge performance gain (improvements by a factor of up to 3) obtained by using repeated games with intervention. We also derive the minimum intervention capability required to achieve the social optimum, and design equilibrium strategies to achieve it. The generalized repeated game model with intervention applies to a large variety of communication scenarios, including power control, medium access control, and flow control. Importantly, these new game theoretic - design concepts can also be applied to many emerging fields - network security, social networks, energy markets etc. 
"Maximizing Profit on User-Generated Content Platforms" 
Shaolei Ren and Mihaela van der Schaar 
Abstract: In this talk, I present a unified payment scheme that can maximize the profit of a user-generated content platform. First, I study how the content viewers’ attention is spread across content available on the platform. Then, by modeling the content producers as self-interested agents making independent production decisions, I analyze the equilibrium state in which no content producers can gain additional benefits by deviating. Next, the optimal payment scheme is derived based on the well-known Dixit-Stiglitz utility function. Discussions regarding whether and when the intermediary should charge or reward the content producers are also provided. The proposed payment scheme can be applied to maximize the profit by online content platforms such as YouTube, Yahoo!, and Facebook. 

Session 2. Emerging Electromagnetics Techniques

"Dual Band Isolation Circuits Based on CRLH Transmission Lines for Triplexer Applications" 
Hanseung Lee and Tatsuo Itoh 
Abstract: This presentation shows a novel dual band isolation network using a Composite Right/Left Handed (CRLH) transmission line (TL) and its application to a triplexer. This triplexer consists of three dual band isolation networks composed of a CRLH transmission line, and each network is connected to a bandpass filter. This triplexer has two significant advantages. First, there is no restriction in choosing filters. Any filters can be used in this concept and it is not necessary to modify characteristics of the filters. Second, the design process of the triplexer is straightforward and no complex numerical analysis process is needed.
"Smoothed Sigmoid Circularly Polarized Waveguide Septum Design using Particle Swarm Optimization" 
Ilkyu Kim, Joshua Kovitz, and Yahya Rahmat-Samii 
 Abstract not available 
"Near-field based Electrical and Communication Systems" 
Umar Azad, Crystal Jing and Yuanxun Ethan Wang
Abstract: To evaluate the performance of near-field energy transfer links, a near-field power transfer equation for an inductively coupled near-field system, analogous to Friis transmission equation for far-field communication, is derived. Experimental results show the proposed near-field coupling equation is trustworthy as it correctly predicts the transferred power versus distance relationship for different values of loaded quality factors at the transmitter and the receiver. Capacity performance of near-field communication (NFC) links is analyzed for noise limited and interference limited scenarios. The analytical results provide guidelines for design of inductively coupled antenna systems as the power and capacity budget of the link is carried out. 
"Stacked Tunable Filter Bank with Very Wide Tuning Range" 
Jim S. Sun, Noriaki Kaneda, and Tatsuo Itoh 
Abstract: Tunable filters with wide tuning range are desirable in several wireless applications. We propose a partially shielded tunable filter structure using varactor-loaded split-ring resonator. The physical topology of the filter is suitable for vertical stacking of filters and enables a compact multi-layer tunable filter for applications requiring very wide tuning range. A prototype is developed with 5:1 frequency tuning capability, and design method is also presented for further improvement of the prototype or re-design at another frequency range.

"Total System Power Minimization of Microprocessors using Refrigerated Systems for Electronic Cooling"
Wonho Park and Ken Yang 
Abstract: Power dissipation and thermal problems have become a growing issue for scaled technology. This phenomenon drives the need for advance cooling systems. It is well-known that cooling the operating temperature results in reduced electric power and/or speed gains. However, total power dissipation includes both electric power and the refrigeration power. A refrigeration system is developed and experimentally tested to demonstrate that cooling the high performance microprocessor can lead to overall system power improvement. A processor that dissipates 175.4W of maximum power with 30% electronic leakage power operating at 97A°C is cooled using our refrigeration system. Our measurement results show that with a minimum refrigeration coefficient of performance of 2.7, the processor operates with junction temperature <40A°C and offers a 25% total system power reduction over the non-refrigerated design. This experiment is the first demonstration of active cooling that lead reduced total wall power. The experiment also indicate that the compressor requires high COP across a broad range of cooling capacity in order for the system to obtain total power savings across different range of processor utilization. In such case, our analysis shows that at least >12% of total power is saved across the entire range of processor utilization. 
"Practical Energy-Aware Link Adaptation for MIMO-OFDM Radios" 
Eren Eraslan, Chao-Yi Wang, and Babak Daneshrad 
Abstract: Circuit designers have put substantial effort to reduce the energy consumption of specific blocks in wireless communication devices. However, much more significant increase in energy efficiency can be achieved at the system level by proper choice of transmission parameters. Choosing the best mode to transmit given the channel conditions in a wireless link is referred to as link adaptation. In this talk, we present a novel energy-aware link adaptation protocol for MIMO OFDM systems, which chooses the transmission parameters, such as transmit power, number of transmit/receive antennas, modulation and coding scheme, to minimize the total energy consumption of the link while satisfying the application's QoS requirements. Through simulation and experimental results, we show that it provides orders of magnitude gain in energy efficiency of the communication link. 
"Efficiency Enhancement of a Class E Power Amplifier under Back-off" 
Nitesh Singhal and Sudhakar Pamarti 
Abstract: As modern wireless communication technologies increasingly use non-constant envelope modulation schemes, power amplifiers (PAs) that can maintain close to peak efficiency over the dynamic range of such modulation schemes become critical. This talk focuses on novel methods of maintaining high efficiency under output power back-off using a class E amplifier. A Zero voltage switching technique which can ideally maintain 100% efficiency at all power is discussed. Another technique which involves modulating the load in an optimal way is also presented. The application of the techniques is illustrated by the design and measurement on discrete and IC based prototypes. 
"A Behavioral Algorithm for State of Charge Estimation" 
Ayca Balkan, Min Gao, Paulo Tabuada, and Lei He 
Abstract: Estimating the state of charge (SOC) of electrochemical batteries is an important, yet challenging task. In our work, we propose a new algorithm to estimate the SOC using the behavioral framework proposed by Jan Willems. Our method, which only requires terminal voltage and terminal current data, does not postulate any model of a battery unlike most of the previous work. We compare the performance of our algorithm with existing methods and demonstrate comparable results. 

"Voltage-Induced Switching of CoFeB-MgO Magnetic Tunnel Junctions" 
Juan G. Alzate, Pedram Khalili Amiri, Sergiy Cherepov, Jian Zhu, Pramey Upadhyaya, Mark Lewis, Ilya Krivorotov, Jordan Katine, Jurgen Langer, Kosmas Galatsis, and Kang L. Wang 
Abstract: We present voltage-induced switching of magnetic states in nanoscale elliptical magnetic tunnel junctions (MTJs). In this work, we demonstrated experimentally voltage-driven magnetization switching, assisted by magnetic fields, in a 190 nm by 60 nm structure by exploiting the voltage-controlled perpendicular anisotropy in ultra-thin ferromagnetic metallic layers. The two stable equilibrium states had an MR ratio ~15%, and the coercivity was 120 Oe. The measured quasi-static R-V curves demonstrate very low switching voltages of ~0.6 V for both directions of switching. This voltage-driven magnetization switching can be exploited in ultralow-power memory and logic applications. 
"Nanowire Fabric Devices and VLSI Technology for Nanoarchitectonics" 
Jorge Kina, Kyeong-Sik Shin, and Chi On Chui 
Abstract: Over the past few years, several unconventional circuit architectures based on nanoscale device fabrics have been proposed as an alternative to future generations of CMOS. One promising candidate is the Nanoscale Application Specific Integrated Circuit (NASIC), which is based on semiconductor nanowires and targeting data-paths with built-in fault resilience modules. The NASIC fabric is built on a 2-D semiconductor nanowire grid with crossed nanowire field-effect transistors (xnwFET) at selected crosspoints. Our research has been focusing on the experimental developments of xnwFET devices and fabric formation technology. In this talk I will present several device innovations as well as VLSI-ready fabric manufacturing technology towards the demonstration of a fully functional NASIC prototype. 
"CMOS-Compatible Surface-Micromachined RF-Relay" 
 Jere Harrison and Rob Candler 
Abstract: The search for increased isolation, reduced insertion loss, and improved system linearity has led to a great deal of academic and industry interest in RF-MEMS switches. Traditional semiconductor switches have isolation limited by capacitive coupling through the high dielectric constant semiconductor material, and the modulation of channel carrier density results in nonlinear channel conductance for large amplitude signals. RF-MEMS switches can directly address these limitations. A magnetically actuated waveguide can produce linear response with a free-space throw of several tens of microns, producing very high isolation. These devices can be fabricated with a low-temperature back-end CMOS-compatible process. 

"Defect-Aware Floorplanning for EUV Masks" 
 Abde Ali and Puneet Gupta
Abstract: Extreme ultraviolet (EUV) lithography is the most promising technology solution replace to 193nm lithography. Apart from source and resist issues, fabricating defect-free mask blanks remains a major "show-stopper" for the adoption of EUV lithography. One promising approach to alleviate this problem is smart reticle floorplanning to minimizing the design impact of buried defects. We propose a simulated annealing based gridded floorplanner for single project reticles that minimizes the design impact of these buried defects. Our simulation results show a substantial improvement in mask yield: for a mask with 40 defects, our approach can improve mask yield from 53% to 94%.
"Stochastic Simulation for Analog Circuit and Stochastic Yield for Memory" 
Fang Gong and Lei He 
Abstract: Two research threads related to stochastic modeling and simulation are presented. Considering device mismatch, thermal noise and flick noise, we develop a transient noise analysis over 100X faster than the existing work and apply it to high precision analog circuits. Considering rare event failure rate for memory, we introduce several methods including cross-entropy minimization within the framework of importance sampling and apply them for memory yield optimization. 
"DRE: A Framework for Efficient and Systematic Co-Evaluation of Design Rules, Technology Choices, and Layout Methodologies" 
Rani S. Ghaida and Puneet Gupta
Abstract: Design rules have been the primary contract between technology developers and designers. While current approaches for defining design rules are largely empirical, this talk offers a framework for fast and systematic evaluation of design rules in terms of area, yield, and variability. The framework essentially creates a virtual cell library and performs the evaluation based on the virtual layouts. Due to the focus on exploration of rules at an early stage of technology development, we use first order models of variability/yield and topology/congestion-based area estimates. The framework can be used to co-evaluate and co-optimize rules, patterning technologies, and layout methodologies. 

"A Single-LC-Tank 5-10GHz Quadrature Local Oscillator for Cognitive Radio Applications" 
Jianhua Lu, Ning-Yi Wang and Frank Chang 
Abstract: This talk presents a local oscillator (LO) that converts oscillation frequencies of 13.3-20GHz from a single-LC-tank VCO to the intended 5-10GHz with continuous frequency coverage. A 4-stage differential injection-locked ring oscillator (ILRO) is used after the latch-based divider to produce quadrature output phases without requiring 50% duty cycle of input signals as those of conventional divide-by-2 approaches. When implemented in 65nm CMOS, the prototype LO consumes 22mA at 1V supply and is able to exhibit a worst-case phase noise of -102dBc/Hz at 1MHz offset across the entire 5-10GHz band. 
"Low-Power Broadband Frequency Synthesis for Cognitive Radios" 
Marco Zanuso and Behzad Razavi 
Abstract: Cognitive radios communicate on any standard across two or three frequency decades. A key challenge in the design of such radios is the synthesis of the carrier frequency for a broad range. A new frequency synthesizer architecture is proposed that leverages the low phase noise of LC oscillators and the wide tuning range of ring oscillators. The synthesizer generates a carrier frequency from 10 MHz to 10 GHz and its phase noise satisfies mobile communication standards. The power consumption is lower than 20 mW. 
"A 7.4mW 120MS/s Wideband Spectrum Sensing Digital Baseband Processor for Cognitive Radios" 
Tsung-Han Yu, Danijela Cabric and Dejan Markovic
Abstract: A digital baseband cognitive radio spectrum sensing processor with channel-specific threshold and sensing time is integrated in 1.64mm2 in 65nm CMOS. The processor achieves detection probability .9 and false-alarm probability .1 for 5dB within a 50ms sensing time. The power and area of are minimized by jointly considering algorithm, architecture, and circuit parameters. The chip dissipates 7.4mW for a 200MHz sensing bandwidth. A 22x reduction in power per sensing bandwidth is achieved compared to prior work. 

"Variability-Aware Tasking for Sensor Networks" 
Lucas Wanner, Rahul Balani, Puneet Gupta, and Mani Srivastava 
Abstract: Instance and temperature-dependent power variation has a direct impact over quality of sensing for battery powered, long running sensing applications. We introduce variability-aware operating system abstractions that allow sensing applications to specify minimum lifetime and quality requirements for individual tasks, so that the system can dynamically adjust and activate tasks, and overall quality of service is maximized in the presence of power variability. We show that variability-aware task scheduling yields an average improvement of 6x in total active time over schedules based on worst-case estimations of power for a contemporary embedded processor. 
"Context Guided and Personalized Activity Classification System" 
James Y. Xu and Gregory Pottie
Abstract: Continued rapid progress in the development of embedded motion sensing enables wearable devices that provide fundamental advances in the capability to monitor and classify human motion, detect movement disorders, and estimate energy expenditure. We present a novel end-to-end system that provides context guided personalized activity classification. The system introduces interface models that feature a context classification committee, the concept of context specific activity classification, the ability to manage sensors given context, and the ability to operate in real time through web services. We also present an implementation that demonstrates accurate context classification and accurate activity classification using context specific models. 
"Energy Aware Multicasting for Wireless Networks" 
Jeffrey Tan, Reuven Cohen, and Izhak Rubin 
Abstract: We consider a wireless network, such as a public safety network or an LTE-based cellular system, in which multiple base stations (or access points) coordinate their operations to efficiently utilize the wireless downlink channels in distributing multicast messages to interested user clients. We develop adaptive power scheduling algorithms that determine which base stations will simultaneously multicast messages to their associated user clients in each time slot, and set the transmit power levels employed by these scheduled base station. Our aim is to present and study such algorithms that enable the multicasting of outstanding load of packets to all identified multicast group clients (or to as many as feasible) located in the area of operation within a short period of time. Thus achieving a high receive multicast throughput rate. We first develop a mathematical formulation for minimizing the schedule length required to complete the transmission of the multicast messages. We show that the problem can be represented as a mixed-integer linear programming model, and prove that the joint scheduling and power control problem is NP-hard. Consequently, we develop and evaluate three heuristic algorithms of polynomial complexity for solving the problem in a timely and practical manner. Heuristic Algorithm 1 is a centralized algorithm that is based on the construction of a power controlled multicast interference graph. We then introduce Heuristic Algorithm 2, which is of lower computational complexity. It operates by iteratively selecting for activation, in a given time slot, a base station transmission that covers the highest residual number of multicast clients. 
"Efficient Sampling and Inferencing in Wireless Health Systems"
Zainul Charbiwala and Mani Srivastava 

"Directivity Enhancement of Offset-Fed Reflectarray Antennas Employing Dipole Type Radiating Elements" 
Jordan Budhu and Yahya Rahmat-Samii 
Abstract: Microstrip Reflectarray Antennas are becoming more widely used in applications where a traditional parabolic reflector would be. In order to provide comparable performance, the Reflectarray antenna must provide similar pattern characteristics. When an offset feeding arrangement is chosen in conjunction with dipole type radiating elements, several factors must be considered in the design in order to achieve optimum pattern performance. This presentation will overview these design choices, the impact they have on pattern performance and overall directivity, and provide guidelines which can be used in the initial design planning stage of these types of antennas. 
"Dual-Band CRLH Phased-Array Feed Network"
Jun Choi and Tatsuo Itoh 
Abstract not available
"C-shaped, U-shaped and E-shaped Patch Antennas in Wireless Application- A comparative Study" 
Shubhendu Bhardwaj and Yahya Rahmat-Samii 
Abstract: Slotted patch antennas have gained their popularity in wireless communication frequency band (1.9GHz-2.4GHz), due to their high bandwidth and miniaturized size. These antennas can have varying characteristics depending on the position and orientation of slot. Here, three popular types of slotted antennas are compared with respect to their RF mechanism, radiation mechanism, bandwidth, size and cross-polarization characteristics for same operational frequency. It is found that C-shaped and double C-shaped antennas, even though miniaturized, suffer from narrow bandwidth and high cross-polarization. U-slotted and E-shaped patch antennas can achieve wider bandwidth with minor penalty in size and cross-polarization. 
"MEMS Reconfigurable E-Shaped Patch Design using Particle Swarm Optimization" 
Joshua Kovitz, Harish Rajagopalan, and Yahya Rahmat-Samii 
Abstract: A septum is an effective polarizer to generate circular polarization from a linear excitation in a rectangular waveguide. Using full-wave simulations, a graphical visualization is presented to aid in better understanding the fundamental principles of a stepped septum circular polarizer. Previous designs for septum polarizers have used simple trial and error methods, which are not always optimal for multi-objective operations. In this research, we propose a novel septum design through the use of Particle Swarm Optimization (PSO), a global nature-inspired optimization technique. This design utilizes modified Sigmoid functions to define its contour within the discontinuous region. These functions allow another degree of freedom to control each step’s smoothness in order to achieve better impedance matching, axial ratio, and bandwidth.

"Noise Robust Fundamental Frequency (F0) Estimation for Speech using SNR-weighted Summary Correlograms from Multi-band Comb Filters" 
Lee Ngee Tan and Abeer Alwan 
Abstract: A noise-robust, signal-to-noise ratio (SNR)-weighted correlogram-based F0 estimation algorithm is proposed. A comb-filterbank operates in each frequency band (low, mid and high). Correlograms are obtained by applying autocorrelations on all comb-filtered channel outputs. An SNR-weighting scheme is used for channel selection to yield a summary correlogram from each correlogram. These summary correlograms are averaged and time-smoothed before peak extraction is performed. The proposed algorithm is evaluated on the Keele corpus with additive white or babble noises. In comparison with other widely-used F0 estimation algorithms, the proposed algorithm has the lowest overall gross pitch error, especially in low SNR cases. 
"Real-time optogenetic functional magnetic resonance imaging (rt-ofMRI) using graphic processing unit (GPU) based parallel computation" 
Zhongnan Fang and Jin Hyung Lee 
Abstract not available 
"Cooperative Prey Herding Based on Diffusion Adaptation" 
Sheng-Yuan Tu and Ali H. Sayed
Abstract: Mobile adaptive networks consist of a collection of nodes with learning and motion abilities that interact with each other locally in order to solve distributed processing and distributed inference problems in real-time. In this talk, we develop adaptation algorithms that exhibit self-organization properties and apply them to the model of cooperative hunting among predators. The results help provide an explanation for the agile adjustment of network patterns in the interaction between fish schools and predators. 
"Noise Robust Automatic Speech Recognition Using a Novel Approach to Log-Spectral Enhancement"
Julien van Hout and Abeer Alwan 
Abstract: This talk describes a technique for enhancing the Mel-filtered log spectra of noisy speech, with application to noise robust speech recognition. We first compute an SNR-based soft-decision mask in the Mel-spectral domain as an indicator of speech presence. Then, we exploit the known time-frequency correlation of speech by treating this mask as an image, and performing median filtering and blurring to remove the outliers and to smooth the decision regions. This mask constitutes a set of multiplicative coefficients (ranging in [0,1]) that are used to discard the unreliable parts of the Mel-filtered log-spectrum of noisy speech. Finally, we apply Log- Spectral Flooring [Borgstrom&Alwan, 2010] on the liftered spectra of both clean and noisy speech so as to match their respective dynamic ranges and to emphasize the information in the spectral peaks. The noisy MFCCs computed on these modified log-spectra show an increased similarity with their corresponding clean MFCCs. Evaluation on the Aurora-2 corpus shows that the proposed approach provides slight improvements in accuracy when compared to state-of-the-art front-ends, like ETSI-AFE, MVA or PNCC, for a low computational cost. 

"Analysis of Phase Noise in Phase/Frequency Detectors" 
Ali Homayoun and Behzad Razavi 
Abstract: The mechanisms giving rise to the phase noise for a PFD in a phase-locked loop are described. The phase noise is calculated based on the accumulation of phase noise in successive transitions around the PFD circuit. Each transition is similar to the transition in an inverter, whose phase noise due to both white and flicker noise is modeled and computed. The proposed phase noise calculations are validated by spectre simulations of static and dynamic PFDs realized in 65-nm CMOS technology. 
"A V-band Voltage Controlled Oscillator with Greater than 18GHz of Continuous Tuning-Range Based on Orthogonal E mode and H mode Control" 
Alborz Jooyaie and Frank Chang
Abstract: A technique to achieve an extended continuous tuning range for Voltage Controlled Oscillators (VCO) is presented. The technique is scalable and the theory could be applied to achieve wide tuning range VCOs operating at arbitrary center frequency; however, it is more desirable at mm-wave regime (V-Band in this case) as it alleviates the need for switches and big varactor banks. The technique incorporated here relies on separate E and H mode excitation of the resonator, while avoiding the Q-degrading switches. The standing-wave V-band VCO reported here is implemented in 65-nm CMOS technology and achieves a continuous tuning range from 58 GHz to 76.2 GHz, with an average phase noise of -89.5 dBc/Hz at 1 MHz offset across the entire band, consumes an average of 5.8 mW (excluding the output buffers), and thus achieves a record FoM.
"A 2.8 to 3.2 GHz fractional-N digital PLL with ADC-assisted TDC and inductively coupled fine-tuning DCO" 
Chih-Wei Yao and Alan Willson 
Abstract: A 2.8 to 3.2 GHz fractional-N digital PLL implemented in 0.18-um CMOS is presented. An ADC is employed to boost TDC resolution by five times to achieve 2 ps effective resolution. A dither-less DCO with an inductively coupled fine-tune varactor bank improves tuning step-size to 20 kHz. A divider with two-stage retiming improves linearity to reduce fractional spurs without increasing the in-band noise floor. The prototype is expected to achieve better in-band phase noise, better FOM, and smaller area than the state-of-art fractional-N charge-pump PLLs. 
"A 25-Gb/s 5-mW CMOS CDR Circuit" 
Jun Jung and Behzad Razavi
Abstract: Low-power solutions for high-speed wireline circuits have seen a resurgence in demand. This presentation introduces a charge-steering circuit technique that considerably reduces the power consumption of circuits such as latches and flipflops. A half-rate CDR circuit employing this technique and including a 2:4 DMUX consumes less than 5 mW from a 1-V supply in 65-nm CMOS technology. 

"V-band Self-Healing Power Amplifier with Adaptive Feedback Bias Control in 65nm CMOS"
Jenny Yi-Chun Liu, Adrian Tang, Ning-Yi Wang, Qun Jane Gu, Roc Berenguer, and Frank Chang
Abstract: A self-healing two-stage 60 GHz power amplifier (PA) with amplitude/phase compensation is realized in 65 nm CMOS. An adaptive feedback bias scheme with three control knobs is proposed to extend the linear operating region and enhance chip-to-chip performance yield; allowing a 5.5 dB improvement of the output 1-dB compression point (P1dB) and a less than 2% chip-to-chip gain variation. At a 1 V supply, the fully differential PA achieves a saturation output power (Psat) of 14.85 dBm with a peak power-added-efficiency (PAE) of 16.2%. With the on-chip amplitude compensation, the P1dB is extended to 13.7dBm. With the on-chip phase compensation, the output phase variation is minimized to less than 0.5 degree. The PA delivers a linear gain of 9.7 dB and has a 7 GHz bandwidth from 55.5 to 62.5 GHz with a very compact area of 0.042 mm2. 
"Wide-bandwidth open loop phase modulator" 
Nitin Nidhi and Sudhakar Pamarti 
Abstract: Emerging wireless communication standards require the transmitter to be wide-bandwidth and power efficient. Polar and out-phasing are two promising candidates for such applications. Both of these architectures require a wide-bandwidth phase modulator. Open loop phase modulation presents a viable solution for achieving wide-bandwidth operation. An on-chip calibration technique, which can attain high precision in the measurement of digital-to-phase characteristics of a phase interpolator, is proposed. The technique makes use of the time-to-digital converter (TDC) in a digital PLL, while avoiding the impact of its non-idealities. 
"Gain-Enhanced Distributed Amplifier-Based CRLH-Leaky Wave Antenna for Quasi-Resonant Power Recycling Scheme" 
Chung-Tse Michael Wu and Tatsuo Itoh 
Abstract: A distributed amplifier combined with CRLH-leaky wave antennas (LWAs) for a new resonant type power recycling scheme is proposed. By connecting a closed loop to the LWA on the drain side of the distributed amplifier, we can obtain gain enhancement compared to the one directly terminated without the loop. Such enhancement fluctuates with respect to the frequency. At some specific frequencies, the enhancement reaches a maximum. The phenomenon will be explained theoretically and validated through the measurement and simulation. 
"Signal-processing techniques for wideband data converters"
Abhishek Ghosh and Sudhakar Pamarti 
Abstract: This talk will focus on signal conditioning techniques for voltage-controlled oscillator (VCO) ring A/D converters. VCOs promise an inexpensive, power-efficient way to implement A/D converters of moderate to high bandwidths. The frequency of the ring oscillator output is changed proportional to the input signal that needs to be quantized; quantization is achieved by simply counting the number of rising/falling edges of the ring oscillator output(s) in a given period of time. However, the non-linear VCO tuning curve poses severe limitations on the ADC dynamic range. To allay this problem, novel dithered signal-conditioning techniques to scramble the VCO-tuning errors are proposed. The oversampled nature of the system enables pushing out the error power out of the signal band, thereby attaining a robust SNDR at a minimal power expense. The scheme is able to achieve a resolution >12 bits at a nominal power-consumption of 5mW for signals having bandwidths in the 20MHz range. 

"Detecting Stumbles Using Accelerometers" 
Nabil Hajj Chehade and Gregory Pottie
Abstract: Falls are a major health problem for the elderly, and stumbles are good indicators for a fall. In this talk, we describe an approach for the detection of stumbles with a new personal activity monitoring system. Our system consists of low cost triaxial accelerometers that may be worn by patients and are convenient for a wide range of subjects. We use machine learning and data mining techniques to detect and count the stumbles in the acceleration data. We also validate our system with data collected from 10 subjects.
"Cyclostationary Feature Detection from Sub-Nyquist Samples" 
Eric Rebeiz and Danijela Cabric 
Abstract: Wideband spectrum sensing which requires detecting the presence or absence of signals in a wideband channel faces multiple practical issues. Current bandwidth limitations of state-of-the-art analog to digital converters require alternative approaches to be considered for wideband sensing. Cyclostationary feature detection is a promising sensing tool which is robust to noise, and takes advantage of the noise stationarity. In this talk, we propose a cyclostationary feature detector that operates on sub-Nyquist samples obtained via either multicoset sampling or the modulated wideband converter analog front-end, and present the receiver. 
"A practical approach to interference suppression using multiple antennas" 
Gaelen Pereira and Babak Daneshrad 
Abstract: Interference suppression has been the subject of extensive research in multi-user communication, with several strategies having emerged to address this issue. However, most multi-antenna based approaches assume knowledge of channel estimates, which are hard to obtain in practice. We propose a practical two step solution that consists of a spatial filter applied directly to the incoming signal, followed by an MMSE decoder, and thus does not require channel estimates or changes to the existing receiver in order to suppress interference. The talk will also cover implementation challenges such as AGC control in a MIMO OFDM radio prototype, and we show through experimental results robust performance at a SIR of -10dB.

"Lensfree Optical Tomographic Microscopy" 
Serhan O. Isikman, Waheb Bishara, Sam Mavandadi, Steve Feng,Frank Yu, and Aydogan Ozcan
Abstract: Tomographic imaging of biological specimen provides detailed volumetric information regarding their internal structure. Nevertheless, existing three-dimensional microscopy (3D) modalities are relatively costly and bulky, and they can probe limited imaging volumes. To provide an alternative microscopy tool that enables depth-resolved imaging of orders of magnitude larger samples in a compact and simple architecture, we have recently developed lensfree optical tomography. In this platform, holographic shadow images of micro-objects are recorded for different illumination angles to compute slice images (tomograms) with micrometer-scale 3D resolution. Lensfree optical tomography can be a particularly useful sectional-imaging tool for lab-on-a-chip applications and telemedicine microscopy. 
"Developing a Compact Source of High Energy Protons for Cancer Therapy"
Dan Haberberger, Sergei Tochitsky, Chao Gong, Chan Joshi, Warren Mori, and Frederico Fiuza 
Abstract: Over the past decade, laser driven ion acceleration (LDIA) has drawn great interest from the scientific community due to the promise of a cheap and compact source of high quality ion beams for a myriad of applications including hadron cancer therapy. Towards this end, at the UCLA Neptune Laboratory we have investigated LDIA using a high-power CO2 laser pulse in a H2 gas jet. This unique interaction produces a shock wave which accelerates protons to energies up to 22MeV contained within an energy spread of E/EFWHM~1%.With a modest extrapolation of state-of-the-art laser technology, it may be possible to reach 100-200 MeV proton beams that are needed for the treatment of many types of cancerous tumors. 
"Lensfree Fluorescence Microscopy" 
Ahmet F. Coskun, Ting- Wei Su, Ikbal Sencan, and Aydogan Ozcan 
Abstract: We review a recently introduced wide-field on-chip fluorescence microscopy platform that can *simultaneously* monitor fluorescent micro-objects or labeled cells/model animals over >0.6-8 cm2 imaging field-of-view without the use of any lenses, thin-film interference filters or mechanical scanners. In this platform, the fluorescence emission from the objects is collected by a fiber-optic faceplate and is delivered to an opto-electronic sensor-array (e.g., a CCD chip). These recorded lensfree fluorescence images are then rapidly decoded using a compressive sampling algorithm to achieve ~4µm spatial resolution over entire chip area (e.g., >0.6-8cm2). Such an on-chip lensfree imaging platform could be very useful for high-throughput cytometry, rare-cell analysis, and microarray research. 
"Active terahertz quantum-cascade leaky-wave antenna"
Amir A. Tavallaee, Benjamin S. Williams, Philip W. C. Hon, Tatsuo Itoh, and Qi-Sheng Chen 
Abstract: We present an active leaky-wave metamaterial antenna realized in terahertz quantum-cascade (QC) structures that exhibits frequency-dependent direction of radiation. The metamaterial antenna is fed by a master oscillator QC-laser with a mode that propagates with an effective phase index smaller than unity such that it radiates in the surface direction due to a leaky-wave mechanism. The direction of emission of main beam is governed by the antenna dispersion characteristic. 25 degrees of beam steering is observed as the lasing frequency of the QC-laser is varied from 2.65-2.81 THz. 
"Automated On-Chip Semen Analysis using a Handheld Lensfree Holographic Microscope" 
Ting-Wei Su, Anthony Erlinger, Derek Tseng, and Aydogan Ozcan 
Abstract: We demonstrate automated semen analysis using a lensfree on-chip microscope. This compact holographic microscope weighs ~46 grams and does not require any lenses, lasers or other bulky optical components to achieve phase and amplitude imaging of sperms over ~24 mm^2 field-of-view with a numerical aperture of ~0.2. Such a compact and light-weight automated semen analysis platform that can investigate both sperm concentration and motility over a large field-of-view is especially important for fertility clinics, personal male fertility tests, as well as for field use in veterinary medicine such as in stud farming and animal breeding applications. 

"Soft Information for LDPC Decoding in Flash: Mutual-Information Optimized Quantization" 
Jiadong Wang, Thomas Courtade, Hari Shankar and Richard Wesel 
Abstract: High-capacity NAND flash memory can achieve high density storage by using multi-level cells (MLC) to store more than one bit per cell. Although this larger storage capacity is certainly beneficial, the increased density also increases the raw bit-error-rate (BER), making powerful error correction coding necessary. Traditional flash memories employ simple algebraic codes, such as BCH codes, that can correct a fixed, specified number of errors. This talk investigates the application of low-density parity-check (LDPC) codes which are well-known for their ability to approach capacity in the AWGN channel. We obtain soft information for the LDPC decoder by performing multiple cell reads with distinct word-line voltages. The values of the word-line voltages (also called reference voltages) are optimized by maximizing the mutual information between the input and output of the multiple-read channel. Our results show that using this soft information in the LDPC decoder provides a significant benefit and enables us to outperform BCH codes over a range of block error rates. 
"Extending the Lifetime of Flash Memory Using Coding Techniques" 
Ryan Gabrys and Lara Dolecek 
Abstract: Flash memories promise faster data access, less power consumption, and greater durability than traditional storage mediums. However, these benefits are accompanied with greater lifetime variability. In this talk, we will demonstrate how a novel coding methodology can be used with great success to mitigate device wear-out and to substantially extend memory lifetime. Coding methods that opportunistically exploit the nature of intracell variability of dense Flash memories will be discussed.
"Superposition Coding for Constrained Modulations" 
Thomas Courtade and Richard Wesel 
Abstract: In this talk, we consider a network of n nodes, each initially possessing a subset of packets. Each node is permitted to broadcast functions of its own packets and the messages it receives to all other nodes via an error-free channel. We provide an algorithm that efficiently solves the Weighted Universal Recovery Problem and the Secrecy Generation Problem for this network.
"Protograph-Based Raptor-Like LDPC Codes for Rate Compatibility with Short Blocklengths" 
Tsung-Yi Chen, Dariush Divsalar, Jiadong Wang and Richard Wesel 
Abstract: In this presentation we will introduce a new class of rate-compatible LDPC codes, protograph-based Raptor-like (PBRL) LDPC codes. The proposed PBRL codes are jointly decodable with an iterative belief propagation decoder. As with Raptor codes, additional parity bits can be easily produced by exclusive-or operations on the precoded bits, providing extensive rate compatibility. We will present a design procedure that optimizes this class of rate-compatible LDPC codes. The new PBRL codes outperform 3GPP rate-compatible turbo codes with the same short blocklength at high SNR and show no sign of an error floor at the FER region of $10^{-7}$.
Document Actions