A Reliability Improvement Technique in Severe

0 downloads 0 Views 957KB Size Report
Nima Soltani, Miaad Aliroteh, Roman Genov. University of Toronto, Canada. C4L-A.2. Mechanical Power Assessment of Fast Eye Motions for Energy Harvesting.
A Reliability Improvement Technique in Severe Process Variations and Ultra Low Voltages Mohsen Radfar , Kriyang Shah and Jack Singh Centre for Technology Infusion La Trobe University Melbourne, Australia Abstract—Yield and reliability reduction at sub/nearthreshold voltage domains, caused by the severe process variations in this voltage region, is a challenging characteristic of implantable medical and ultra low power sensory silicon devices. Using a variation sensitive design, this paper proposes a novel technique capable of sensing and responding to process variations by providing an appropriate forward body bias (FBB) so that the delay variations and reliability of whole system are improved. Theoretical analysis for the error probability is confirmed by post-layout HSPICE simulations for an 8-bit Kogge-Stone adder and shows considerable reduction of the error rate from 50% to 1% at 0.4V.

I.

INTRODUCTION

With the introduction of 65nm technologies, reliability of circuits started to challenge the transistor scaling, an essential trend for continuation of performance, area and energy improvements in the silicon industry, especially in ultra low power medical or sensory applications [1]. Higher inaccuracy in the fabrication process as a result of aggressive transistor scaling is the source of this unreliability.

Authors in [7] also studied the capability of ABB techniques to address these variations and implemented a subthreshold processor to show its effectiveness. In this paper, the purpose is addressing the challenge of reliability reduction at subthreshold voltages. An extreme process variation sensitive FBB circuit is proposed capable of addressing process variations while VDD scaling by tuning P to N ratios in different process corners. This technique helps improve the system reliability by applying the FBB to NMOS and/or PMOS networks at lower voltages and/or lower temperatures and applying appropriate FBB to slower devices depending on the process variation. The rest of this paper describes how FBB circuit can achieve this by firstly mathematically analysing how circuit works and why it reduces error rate in section II, and then, in section III, providing the HSPICE simulation results obtained from post-layout Monte Carlo runs to back the theoretical findings. II.

THEORETICAL ANALYSIS OF THE PROPOSED CIRCUIT

On the other hand, as the subthreshold voltage domain offers minimum energy consumption, it is very attractive for designers of the implantable medical electronics [2]. However, unreliability challenges are pronounced even more seriously at the subthreshold voltages as process variations rise exponentially with the voltage scaling. This results in a dramatic uncertainty and subsequent functional failure that can be very disastrous in medical applications. This impact still urges designers to employ adaptive body-bias (ABB) techniques [3] despite the fact that technology scaling counteracts the body biasing effect.

Figure 1 shows the schematic of the proposed FBB generator introduced in [8] which was used to apply FBB to the MOS devices so that the Energy Delay Product (EDP) of system was improved and process, voltage and temperature variations were addressed while system was working in subthreshold voltages.

Although efficient in the superthreshold region, the impact of the body biasing is especially sensed at the subthreshold voltages because of the exponential increase in the sensitivity of devices to the threshold voltage. For example in a typical 90nm technology, if threshold voltage is changed by 50mV at a 1V supply voltage, delay varies 13% whereas it results in a 55% delay increase at a 0.45V supply voltage [4].

A. Proposed Forward Body Bias Circuit If a device is operating in subthreshold region, its drain source current can be modelled by the well known equation [9]:

As PMOS and NMOS transistors can be controlled independently using body biasing techniques, this opens many opportunities for designers to optimally tune the β-ratio and preventing VTH mismatch problems. For example, authors in [5] [6] used VTH balancing schemes which enabled them to implement a supply voltage scaling from superthreshold to subthreshold voltages. Their body biasing technique adapts P/N-ratio dynamically while voltage scaling.

978-1-4799-1471-5/13/$31.00 ©2013 IEEE

Here, however, this technique is mathematically analysed and its novel aspects and applications in reliability and error rate reduction are revealed and discussed. First a quick background is given in subsection A, and then the error rate is worked out in subsection B.

= ℎ

e

1−e =µ

2

(1)

where Vth signifies the threshold voltage, γ the body bias coefficient and m the slope factor of transistor. For I0, µ is the

210

2≤ =

1 2

2

=

2





(3)

in which Erfc(x) is the complementary error function. B. Effect of FBB circuit on error rate While [8] showed the improved EDP resulted from this FBB technique, here the main focus is to show how this technique was designed to take advantage of process variations to improve the reliability. Like the assumption made in [8], as FBB is cancelled at superthreshold voltages, it can be assumed that no improvement or degradations happens while operating in superthreshold region. Therefore, results of the following equations are valid for superthreshold region too. When output of an inverter in the subthreshold region switches from 0 to 1, PMOS is the device which is playing the main role in the delay of the gate, therefore it can be written [11]:

Figure 1. Body Bias generators for a) PMOS network and b) NMOS network

mobility, W the width, L the length, Φs the surface potential, vT the thermal voltage (=kBT/q), NDEP the doping concentration in the channel and εsi the permittivity of silicon for this device. Transistors in first and second stages of these BB generators are always kept off and hence the current passing through them is a subthreshold current. Authors in [10] showed that process variations, mainly affecting channel length and Vth, do not control the reference voltage VN1. It is usually assumed that all process variations can be abstracted in Vth variation [11]. Assuming Vthp0 is a specific threshold voltage of MA2 in which VN2 is equal to VDD/2 and VN2 and VTHP represent random variables for variable VN2 (output voltage of the second stage) and variable Vthp (threshold voltage of MA2), respectively, then VTHP has a known probability density function (PDF), as variations in Vthp presumably follow a Normal Distribution. Vthp0 can be found by letting ≤ [8] which results in: =



+

+

ln

= ℎ

=

(4) 1−

Error rate is modelled as the probability of an erroneous event in a data-path [14] and can be approximated by the probability of error in an inverter which has to have a delay of less than t0 to meet the required speed. Given this assumption and by defining Td as the random variable of td, the probability of error or simply error rate can be acquired by the following equation and using the law of total probability: (

>

(2) =

Sized to its feature size, MA2 is prone to process variations while MB2 is sized large enough (especially in channel length) to be resistant to these variations [12-13]−here standard deviation of VTHN decreased to 0.01V while VTHP standard deviation is 0.04V. As the PDF of VTHP is already known, then the probability of VBSP=VDD or VN2≤VDD/2 can be found by:

1 2



)=

=

+

|

=

>

ln

|

>

> ln

>

+ ≤



.

+

(5) .

Using above equation and a few mathematical derivations, final probability of error is worked out in (6) for FBB circuit and in (7) for zero body bias (ZBB) case1. 1

Using (2), (3), (4), and (5) and the fact that 1 − ≈ 1 for VDD>0.3V, D0 can be substituted in above equations resulting in:

211

Equation (6) has been sketched in Figure 2.a by substituting variables with real world numbers extracted by simulations. Figure 2.a shows that higher supply voltages or looser delay constraints bring about zero probability of error. It can be observed that there is a concavity, when in subthreshold voltage domain, which means tighter delay constraints can be tolerated with respect to ZBB inverter. This is clearer when compared with ZBB inverter (using expression P(TdZBB>t0)-P(Td>t0) or equation (7) - equation (6)) in Figure 2.b which shows the improvement (reduction) in probability of the error in a ZBB inverter after the FBB technique is applied. As it can be seen in Figure 2.b, improvement in the probability of error in the superthreshold voltage region tends Figure 2. a) Probability of error in FBB inverter for mp=1.7, mn=1.48, vT=0.026V, σVTHP=0.04V, µVTHP=0.5V, µVTHN=0.45V, η=2.1, and Cs=1pF b) to zero. This is again due to FBB cancellation in this region. In the subthreshold voltage region, however, when t0 or the delay constraint on the inverter’s output is very tight and close to zero, the probability of delay being larger than the expected constraint is the same for both ZBB and FBB designs and equal to one which results in zero probability improvement. This is true for very loose delay constraints too as both ZBB and FBB inverters can meet the constraint and having the same probability of error equal to zero results in zero improvement too. But delay constraints are determined by clock frequency and are set to practical limits which are neither too tight damaging the production yield nor too loose not meeting the required performance, but they are set to the tightest error free condition. For example, in Figure 2.a this delay constraint (or clock frequency for a data-path) is just set to the points where the surface is about to rise from zero to one. On the other hand, Figure 2.b suggests that improvements in the error rate is possible if FBB applied (here maximum error rate reduction is 0.35 for a theoretical inverter) and this helps tighten the delay constraints even more to achieve higher frequencies. Although delay variation analysis for a single inverter does not represent the exact effect of variations on a data-path comprised of many various cells, but as showed in [11], a path made up from different serried cells will improve variation effects which means the approximation of ( =

⎧1 ⎪ ⎪2

)

>

⎨ ⎪ ⎪ ⎩

ln 2 1 2

ln 2

⎧1 ⎪ ⎪2

2

(1 + ) −

+

2

⎫ .⎪ ⎪



+



⎫ .⎪ ⎪

⎬ ⎪ ⎪ ⎭

(6)

2 ⎨ ⎬ − 1 ⎪ ⎪ ⎪ ⎪ − 2 2 ⎩ ⎭ A ZBB inverter with TdZBB as the random variable for output delay has no FBB and hence a zero VBSP and therefore will have the error rate of ( > ) +

=

1 2

ln 2

2

+



improvement (reduction) in error probability of ZBB after FBB is applied

improvement will be even more experienced in a data-path as simulations prove later. III.

SIMULATION RESULTS AND DISCUSSIONS

In this section, as well as investigating how the proposed technique reacts to the process changes, previously obtained equations are examined through simulations. An 8-bit Kogge-Stone adder was chosen to verify this FBB technique. All simulations were performed using Low Power 65nm TSMC technology model. On this adder, 1000 Monte Carlo (MC) runs were executed to simulate the process variations for each supply voltage ranging from 0.3V to 0.8V. Foundry provided global process variation model was exploited to produce the most accurate possible results in simulating inter-die process variations. This model contained all foundry proven Gaussian distributions which led to the most practical and realistic Monte Carlo simulation results possible. Voltage was also swept to resemble VDD scaling. Simulations for the adder also verify the predictions of equations (6) and (7) abstracted in Table I and Figure 3. As suggested by Figure 2.b, there is a maximum improvement curve for Kogge-Stone adder data-path which, as explained, is expected to be higher (here 0.52) than what was predicted by the theoretical inverter. As this adder clearly has higher delay than the analysed single inverter in the previous section, its delay constraints are also higher than the inverter. It can be seen, in both Table I and Figure 3, that although error rate reduction is maximum along this curve, the FBB error rate is not satisfactory even for subthreshold voltage domain. For example, Table I shows the error probability of 0.138 for VDD of 0.3V when the maximum error rate reduction is sought which is 0.528. In addition, Table I shows that, if FBB is applied, ~2.8x delay constraint relaxation can also be achieved, compared to a ZBB data-path with the same error rate of 0.138. But choice of maximum error rate improvement leads to error rate of 13.8% or higher (across different voltages) in the data-path which is very yield damaging and, in the superthreshold region, this choice has no benefit in terms of the error rate reduction.

(7)

212

technique will be more beneficial in this situation. Here, the best situation, which has the benefit of more delay constraint relaxation, lower error rate, and a reasonable error rate reduction, happens at 0.4V when error rate is 1% with error reduction of 0.495 meaning 50% error rate before the FBB application. IV. CONCLUSION As reliability degradation is a critical challenge when it comes to subthreshold implantable electronics, this paper proposed a technique that dramatically dropped the error rate in an examined pipelined adder, a basic component of every processor. This technique provided FBB based on process variations and the operating voltage. It was concluded from the results that the proposed technique was able to improve the error rate from 50% in the ZBB adder down to 1% in the FBB adder at 0.4V. REFERENCES Figure 3. Probability of error for FBB and ZBB data-path and the maximum gained improvement for 1K MC simulations at 25oC

It is error rate that is often set as a design goal for which case Table I shows examinations of the error rate constraint of 1%. Clearly, as limit on final error rate (probability of error) in the FBB data-path is tightened, error rate reduction declines as well. This is due to distancing from the maximum reduction point, which is experienced at higher error rates. Despite this decline, the error reduction is still significant, even for the expected error rate of 1%. And with this error rate, the delay constraint can be relaxed more, as a result of FBB application and compared to the case of maximum error rate reduction. This is as a result of the FBB design novelty which 1) not only improves the performance of data-path, by providing FBB when required, 2) but also reduces variations in data-path delay. The former feature can be observed in Figure 3 in which, as delay constraint reduces (tightens), probability of error raises to one in ZBB earlier than FBB does. This demonstrates the FBB supremacy in performance because it now enables the data-path to perform faster as delay constraints can be tougher. And the latter is apparent in the slope of error rate curve in Figure 3 when rising up from zero to one with steeper curves belonging to the FBB due to its lower delay variation. As a result, when error rate is reduced to lower percentages (like 1% here), a ZBB data-path will fail more and more as extreme corner variations play an important role in this case and satisfying such extreme process corners will need looser delay constraints and hence this FBB TABLE I. design goal→ voltage↓

ERROR RATE IMPROVEMENTS

1% error rate

maximum improvement

error rate reduction/delay constraint relaxation(x)

error rate (%)

0.3V

0.296/3.3113

0.528/2.8184

13.8

0.4V

0.495/3.3884

0.589/2.6303

13.9

0.5V

0.256/1.9953

0.394/1.5849

18.8

0.6V

0.012/1.0471

0.048/1.0233

26.3

[1] M. Porter, et al., "Reliability considerations for implantable medical ICs," in Reliability Physics Symposium, 2008. IRPS 2008. IEEE International, 2008, pp. 516-523. [2] P. Gerrish, E. Herrmann, L. Tyler, and K. Walsh, "Challenges and constraints in designing implantable medical ICs," Device and Materials Reliability, IEEE Transactions on, vol. 5, pp. 435-444, 2005. [3] M. Radfar, K. Shah, and J. Singh, "Recent Subthreshold Design Techniques," Active and Passive Electronic Components, vol. 2012, p. 11, 2012. [4] J. Rabaey, Low power design essentials: Springer Verlag, 2009. [5] Y. Pu, J. Pineda de Gyvez, H. Corporaal, and Y. Ha, "An Ultra-LowEnergy Multi-Standard JPEG Co-Processor in 65 nm CMOS With Sub/Near Threshold Supply Voltage," Solid-State Circuits, IEEE Journal of, vol. 45, pp. 668-680, 2010. [6] M. E. Hwang and K. Roy, "ABRM: Adaptive $ beta $-Ratio Modulation for Process-Tolerant Ultradynamic Voltage Scaling," Very Large Scale Integration (VLSI) Systems, IEEE Transactions on, vol. 18, pp. 281-290, 2010. [7] S. Hanson, et al., "Ultralow-voltage, minimum-energy CMOS," IBM journal of research and development, vol. 50, pp. 469-490, 2006. [8] M. Radfar, K. Shah, and J. Singh, "A highly sensitive and ultra lowpower forward body biasing circuit to overcome severe process, voltage and temperature variations and extreme voltage scaling," Int. J. Circ. Theor. Appl.. doi: 10.1002/cta.1935, 2013. [9] W. Liu, et al., "BSIM4. 6.4 MOSFET Model." [10] M. SEOK, D. SYLVESTER, D. BLAAUW, S. HANSON, and G. CHEN, "REFERENCE VOLTAGE GENERATOR HAVING A TWO TRANSISTOR DESIGN," ed: WO Patent 2,010,151,754, 2010. [11] B. Zhai, S. Hanson, D. Blaauw, and D. Sylvester, "Analysis and mitigation of variability in subthreshold design," presented at the Proceedings of the 2005 international symposium on Low power electronics and design, San Diego, CA, USA, 2005. [12] A. Srivastava, R. Bai, D. Blaauw, and D. Sylvester, "Modeling and analysis of leakage power considering within-die process variations," presented at the Proceedings of the 2002 international symposium on Low power electronics and design, Monterey, California, USA, 2002. [13] S. Narendra, V. De, S. Borkar, D. Antoniadis, and A. Chandrakasan, "Full-chip sub-threshold leakage power prediction model for sub0.18µm CMOS," presented at the Proceedings of the 2002 international symposium on Low power electronics and design, Monterey, California, USA, 2002. [14] S. Bhunia, S. Mukhopadhyay, and K. Roy, "Process Variations and Process-Tolerant Design," in VLSI Design, 2007. Held jointly with 6th International Conference on Embedded Systems., 20th International Conference on, 2007, pp. 699-704.

213

BioCAS 2013 Table of Contents Expand the bookmarks and navigate to the desired Date and/or Session Name. Scroll to the desired paper and select a Blue title link to open the paper. After viewing, use the bookmarks to return to the Table of Contents or another section. A1P-C

Live Demonstrations of Biomedical Circuits and Systems

Time: Place: Chair(s):

Thursday, October 31, 2013, 17:10 - 19:00 16th floor Andreas Demosthenous, University College London Wouter Serdijn, Delft University of Technology

A1P-C.1

A 6.7µW CMOS Bioamplifier for Active Electrode with DC Rejection ......................1 Shi Huang1, Jinyong Zhang2, Lei Wang1 1 Shenzhen Institute of Advanced Technology, China; 2Shenzhen Institute of Advanced Technology & University of Hong Kong, China

B1P-C

Biomedical Circuits and Systems I

Time: Place: Chair(s):

Friday, November 1, 2013, 10:10 - 11:30 16th floor Marc Notten, Philips Research Bin-Da Liu, National Cheng Kung University

B1P-C.1

An on-Chip Learning, Low-Power Probabilistic Spiking Neural Network with Long-Term Memory.......................................................................................................5 Hung-Yi Hsieh, Kea-Tiong Tang National Tsing Hua University, Taiwan

B1P-C.2

A Bio-Inspired Feedforward System for Categorization of AER Motion Events.............................................................................................................................9 Bo Zhao2, Qiang Yu3, Hang Yu2, Shoushun Chen2, Huajin Tang1 1 Agency for Science, Technology and Research, Singapore; 2Nanyang Technological University, Singapore; 3National University of Singapore, Singapore

B1P-C.3

A 32-Channel Neural Recording System with a Liquid-Crystal Polymer MEA ......13 Seung-In Na, Susie Kim, Taehoon Kim, Hyongmin Lee, Hyunjoong Lee, Suhwan Kim Seoul National University, Korea, South

B1P-C.4

Bio-Mimetic Gyroscopic Sensor for Vestibular Prostheses....................................17 Charalambos Andreou, Yiannis Pahitas, Evdokia Pilavaki Pilavaki, Julius Georgiou University of Cyprus, Cyprus

B1P-C.5

Low-Power Low-Noise ECG Acquisition System with DSP for Heart Disease Identification ................................................................................................................21 Bo-Yu Shiu, Shuo-Wei Wang, Yuan-Sun Chu, Tsung-Heng Tsai National Chung Cheng University, Taiwan

B1P-C.7

Proto-Object Based Visual Saliency Model with a Motion-Sensitive Channel ......25 Jamal Molin2, Alexander Russell2, Stefan Mihalas1, Ernst Niebur2, Ralph EtienneCummings2 1 Allen Institute for Brain Science, United States; 2Johns Hopkins University, United States

B1P-C.8

An Embedded Probabilistic Neural Network with on-Chip Learning Capability .....................................................................................................................29 Jen-Huo Wang, Kea-Tiong Tang, Hsin Chen National Tsing Hua University, Taiwan

B1P-C.9

Electroporation System Generating Wide Range Square- Wave Pulses for Biological Applications...............................................................................................33 Voitech Stankevic, Vitalij Novickij, Saulius Balevicius, Nerija Zurauskiene, Algirdas Baskys, Aldas Dervinis, Vytautas Bleizgys Vilnius Gediminas Technical University, Lithuania

B1P-C.11 A Piecewise Linear Approximating ISFET Readout .................................................37 Mohammadreza Sohbati, Pantelis Georgiou, Christofer Toumazou Imperial College London, United Kingdom B1P-C.12 A Novel Near-Infrared Array Based Arterial Pulse Wave Measurement Method..........................................................................................................................41 Wei-Chin Huang, Hsiang-Wen Hou, Ching-Ju Cheng, Shih-Yang Wu, Tien-Ho Chen, Wai-Chi Fang National Chiao Tung University, Taiwan B1P-C.13 A Portable Hardware Implementation for Temporal Laser Speckle Imaging.........45 Abhishek Rege3, Betty M Tyler1, M Jason Brooke3, Kartikeya Murari2 1 Johns Hopkins University, United States; 2University of Calgary, Canada; 3 Vasoptic Medical, Inc., United States

B1P-C.14 A Wideband CMOS Current Driver for Bioimpedance Applications with Output DC Regulation.................................................................................................49 Loucas Constantinou, Andreas Demosthenous University College London, United Kingdom B1P-C.15 Wireless Sensing Framework for Long-Term Measurements of Electric Organ Discharge .........................................................................................................53 Michael Harris, Evan Salazar, Robert Güth, Vishal Nawathe, Mahmoud Sharifi, Wei Tang, Satyajayant Misra New Mexico State University, United States B1P-C.16 Non-Contact ECG Employing Signal Compensation ...............................................57 Guochen Peng, Mark Bocko University of Rochester, United States B1P-C.17 A Training System for the MyoBock Hand in a Virtual Reality Environment.........61 Go Nakamura2, Taro Shibanoki1, Keisuke Shima5, Yuichi Kurita1, Masaki Hasegawa4, Akira Otsuka4, Yuichiro Honda2, Takaaki Chin3, Toshio Tsuji1 1 Hiroshima University, Japan; 2Hyogo Rehabilitation Center, Japan; 3Hyogo Rehabilitation Center Hospital, Japan; 4Prefectural University of Hiroshima, Japan; 5 Yokohama National University, Japan B1P-C.18 Floating-Gate Capacitance Sensor Array for Cell Viability Monitoring..................65 Timir Datta, Emily Naviasky, Pamela Abshire University of Maryland, United States B1P-C.19 Skin Insertion Mechanisms of Microneedle-Based Dry Electrodes for Physiological Signal Monitoring ................................................................................69 Conor O'Mahony1, Francesco Pini2, Liza Vereschagina1, Alan Blake1, Joe O'Brien1, Carlo Webster1, Paul Galvin1, Kevin McCarthy2 1 Tyndall National Institute, University College Cork, Ireland; 2University College Cork, Ireland B1P-C.21 Estimating the Instantaneous Wrist Flexion Angle from Multi-Channel Surface EMG of Forearm Muscles .............................................................................77 Bence József Borbély, Péter Szolgay Pázmány Péter Catholic University, Hungary

B2L-A

SPECIAL SESSION: Implantable Electronics for Neural Recording and Stimulation

Time: Place: Chair(s):

Friday, November 1, 2013, 11:30 - 13:00 16th floor Manuel Delgado, IMSE-CNM Sylvie Renaud, Université de Bordeaux

B2L-A.1

Energy-Efficient High-Voltage Compliant Implantable Brain-Machine Interfaces .....................................................................................................................81 Md. Hasanuzzaman2, Rabin Raut1, Mohamad Sawan2 1 Concordia University, Canada; 2École Polytechnique de Montréal, Canada

B2L-A.2

Output Stage of a Current-Steering Multipolar and Multisite Deep Brain Stimulator.....................................................................................................................85 Virgilio Valente2, Andreas Demosthenous2, Richard Bayford1 1 Middlesex University, United Kingdom; 2University College London, United Kingdom

B2L-A.3

A CMOS Neurostimulator with on-Chip DAC Calibration and Charge Balancing .....................................................................................................................89 Elliot Greenwald1, Cheng Chen1, Nitish Thakor1, Christoph Maier2, Gert Cauwenberghs2 1 Johns Hopkins University, United States; 2University of California, San Diego, United States

B2L-A.4

A Transistor-Only Power-Efficient High-Frequency Voltage-Mode Stimulator for a Multichannel System..........................................................................................93 Marijn van Dongen, Wouter Serdijn Delft University of Technology, Netherlands

B2L-A.5

A Compact Recording Array for Neural Interfaces ..................................................97 Lieuwe Leene, Yan Liu, Timothy Constandinou Imperial College London, United Kingdom

B2L-B

Lab-on-Chip

Time: Place: Chair(s):

Friday, November 1, 2013, 11:30 - 13:00 15th floor Timothy G Constandinou, Imperial College London Tsung-Yi Ho, National Cheng Kung University

B2L-B.1

Automatic Synthesis of Microfluidic Large Scale Integration Chips from a Domain-Specific Language ......................................................................................101 Jeffrey McDaniel, Christopher Curtis, Philip Brisk University of California, Riverside, United States

B2L-B.2

Multiplexed Detection of Waterborne Pathogens with an Array of Microfluidic Integrated High-Sensitivity Organic Photodiodes ............................105 Nuno Miguel Matos Pires, Tao Dong Vestfold University College, Norway

B2L-B.3

A 50µm Pitch, 1120-Channel, 20kHz Frame Rate Microelectrode Array for Slice Recording .........................................................................................................109 Ben Johnson, Shane Peace, Thomas Cleland, Alyosha Molnar Cornell University, United States

B2L-B.4

Development of Low-Cost Plastic Microfluidic Sensors Toward Rapid and Point-of-Use Detection of Arsenic in Drinking Water for Global Health ..............113 Unyoung Kim, Jessica Vandergiessen, Benjamin Demaree, Mary Reynolds, Kyle Perricone Santa Clara University, United States

B2L-B.5

Integrated Microcapillary System for Microfluidic Parasite Analysis ..................118 Andras J. Laki2, Gabor Zs. Nagy2, Kristóf Iván2, Péter Fürjes1, Olga Jacsó4, Éva Fok4, Pierluigi Civera3 1 Institute for Technical Physics and Materials Science of the Hungarian Academy of Sciences, Hungary; 2Pázmány Péter Catholic University, Hungary; 3Polytechnic University of Turin, Italy; 4Szent István University, Hungary

B3L-A

Biomedical Sensors

Time: Place: Chair(s):

Friday, November 1, 2013, 14:10 - 15:40 16th floor Mohamad Sawan, École Polytechnique de Montréal Pau-Choo Chung, National Cheng Kung University

B3L-A.1

Non-Invasive Tumor Detection Using NIR Light.....................................................122 Yung-Chi Lin1, Sheng-Hao Tseng1, Pau-Choo Chung1, Ching-Fang Yang1, MingHan Wu1, Shoko Nioka1, Yong-Kie Wong2 1 National Cheng Kung University, Taiwan; 2Taichung Veterans General Hospital, Taiwan

B3L-A.2

A Miniaturized System for Imaging Vascular Response to Deep Brain Stimulation.................................................................................................................126 Xiao Zhang, M. Sohail Noor, Clinton B. McCracken, Zelma H.T. Kiss, Orly YadidPecht, Kartikeya Murari University of Calgary, Canada

B3L-A.3

Nanogap-Based Enzymatic-Free Electrochemical Detection of Glucose ............130 Ismael Rattalino1, Paolo Motto2, Irene Taurino1, Fernando Cortes-Salazar1, Gianluca Piccinini2, Danilo Demarchi2, Giovanni de Micheli1, Sandro Carrara1 1 École Polytechnique Fédérale de Lausanne, Switzerland; 2Politecnico di Torino, Italy

B3L-A.4

Non-Contact ECG Recording System with Real Time Capacitance Measurement for Motion Artifact Reduction ..........................................................134 Tom Torfs, Refet Firat Yazicioglu IMEC, Belgium

B3L-A.5

A 0.8V 8-Bit Low-Power Asynchronous Level-Crossing ADC with Programmable Comparison Windows ....................................................................138 Yongjia Li, Duan Zhao, Wouter Serdijn Delft University of Technology, Netherlands

B3L-B

Biomedical Signal and Image Processing I

Time: Place: Chair(s):

Friday, November 1, 2013, 14:10 - 15:40 15th floor Viktor Owall, Lunds Tekniska Högskola Julius Georgiou, University of Cyprus

B3L-B.1

Time-Lapse Imaging of Human Heartbeats Using UWB Radar.............................142 Sverre Brovoll1, Tor Berger1, Yoann Paichard1, Oyvind Aardal1, Tor Sverre Lande3, Svein-Erik Hamran2 1 Norwegian Defence Research Establishment, Norway; 2Norwegian Defence Research Establishment & Univerity of Oslo, Norway; 3University of Oslo, Norway

B3L-B.2

Harmonic Path (HAPA) Algorithm for Non-Contact Vital Signs Monitoring with IR-UWB Radar....................................................................................................146 Van Nguyen, Abdul Q. Javaid, Mary Ann Weitnauer Georgia Institute of Technology, United States

B3L-B.3

Energy-Efficient Two-Stage Compressed Sensing Method for Implantable Neural Recordings ....................................................................................................150 Yuanming Suo, Jie Zhang, Ralph Etienne-Cummings, Trac Tran, Sang Chin Johns Hopkins University, United States

B3L-B.4

Highly Energy-Efficient Compressed Sensing Encoder with Robust Subthreshold Clockless Pipeline for Wireless BANs ............................................154 Yi Li, Xu Cheng, Yicheng Zhang, Weijing Shi, Jun Han, Xiaoyang Zeng Fudan University, China

B3L-B.5

Joint Analog-to-Information Conversion of Heterogeneous Biosignals..............158 Valerio Cambareri2, Mauro Mangia2, Riccardo Rovatti2, Gianluca Setti1 1 Università degli Studi di Ferrara, Italy; 2Università di Bologna, Italy

B4L-A

Biosensors

Time: Place: Chair(s):

Friday, November 1, 2013, 16:00 - 17:30 16th floor Sandro Carrara, EPFL Bin-Da Liu, National Cheng Kung University

B4L-A.1

Multi-Electrode Amperometric Biosensor for Neurotransmitters Detection .......162 Genevieve Massicotte1, Mohamad Sawan1, Giovanni De Micheli2, Sandro Carrara2 1 École Polytechnique de Montréal, Canada; 2École Polytechnique Fédérale de Lausanne, Switzerland

B4L-A.2

Fabrication and Packaging of a Fully Implantable Biosensor Array ....................166 Camilla Baj-Rossi1, Enver G. Kilinc1, Sara S. Ghoreishizadeh1, Daniele Casarino2, Tanja Rezzonico Jost3, Catherine Dehollain1, Fabio Grassi3, Laura Pastorino2, Giovanni De Micheli1, Sandro Carrara1 1 École Polytechnique Fédérale de Lausanne, Switzerland; 2Universitá degli Studi di Genova, Italy; 3Università della Svizzera italiana, Switzerland

B4L-A.3

An EMG Readout Front-End with Automatic Gain Controller for HumanComputer Interface ...................................................................................................170 Hyeon-Cheon Seol, Young-Cheon Kwon, Seong-Kwan Hong, Oh-Kyong Kwon Hanyang University, Korea, South

B4L-A.4

Implementation of a Neuromorphic Vestibular Sensor with Analog VLSI Neurons......................................................................................................................174 Giovanni Passetti1, Federico Corradi2, Marco Raglianti1, Davide Zambrano1, Cecilia Laschi1, Giacomo Indiveri2 1 Scuola Superiore Sant'Anna, Italy; 2University of Zürich and ETH Zürich, Switzerland

B4L-A.5

On-Chip Base Sequencing Using a Two-Stage Reaction-Control Scheme: 3.6-Times-Faster and 1/100-Reduced-Data-Volume ISFET-Based DNA Sequencer ..................................................................................................................178 Yoshimitsu Yanagawa, Naoshi Itabashi, Sonoko Migitaka, Takahide Yokoi, Makiko Yoshida, Takayuki Kawahara Hitachi, Ltd., Japan

B4L-B

Biomedical Signal and Image Processing II

Time: Place: Chair(s):

Friday, November 1, 2013, 16:00 - 17:30 15th floor Pau-Choo Chung, National Cheng Kung University Marc Notten, Philips Research

B4L-B.1

Efficient Scene Preparation and Downscaling Prior to Stimulation in Retinal Prosthesis ..................................................................................................................182 Walid Al-Atabany1, Patrick Degenaar2 1 Helwan University, Egypt; 2Newcastle University, United Kingdom

B4L-B.2

Framework for Evaluating EEG Signal Quality of Dry Electrode Recordings .....186 Alexandra-Maria Tautan1, Wouter Serdijn1, Vojkan Mihajlovic2, Bernard Grundlehner2, Julien Penders2 1 Delft University of Technology, Netherlands; 2Holst Centre / IMEC, Netherlands

B4L-B.3

FPGA Design of a Pulse Encoder for Optoelectronic Neural Stimulation and Recording Arrays ......................................................................................................190 Musa Al-Yaman1, Arfan Ghani1, Alex Bystrov1, Patrick Degenaar1, Pleun Maaskant2 1 Newcastle University, United Kingdom; 2University of Cork, Ireland

B4L-B.4

An Active TX/RX NMR Probe for Real-Time Monitoring of MRI Field Imperfections.............................................................................................................194 Jonas Handwerker2, Maurits Ortmanns2, Jens Anders2, Martin Eschelbach1, Paul Chang1, Anke Henning1, Klaus Scheffler1 1 Max Planck Institute for Biological Cybernetics, Germany; 2Universität Ulm, Germany

B4L-B.5

Towards Automatic Sleep Staging via Cross-Recurrence Rate of EEG and ECG Activity...............................................................................................................198 Nicoletta Nicolaou, Julius Georgiou University of Cyprus, Cyprus

C1P-C

Biomedical Circuits and Systems II

Time: Place: Chair(s):

Saturday, November 2, 2013, 10:10 - 11:30 16th floor Andreas Demosthenous, University College London Julius Georgiou, University of Cyprus

C1P-C.1

A Charge-Balanced 4-Wire Interface for the Interconnections of Biomedical Implants......................................................................................................................202 Jiawei Yang, Shun Bai, Nhan Tran, Hosung Chun, Omid Kavehei, Yuanyuan Yang, Efstratios Skafidas, Mark Halpern, David Ng, Vijay Muktamath University of Melbourne, Australia

C1P-C.2

Externally-Coupled Transcutaneous Energy Transmission for a Ventricular Assist Device -Miniaturization of Ferrite Core and Evaluation of Biological Effects Around the Transformer- .............................................................................206 Takehiro Shibuya, Kenji Shiba Tokyo University of Science, Japan

C1P-C.3

A Reliability Improvement Technique in Severe Process Variations and Ultra Low Voltages ....................................................................................................210 Mohsen Radfar, Kriyang Shah, Jack Singh La Trobe University, Australia

C1P-C.4

A 1-Wire® Communication Interface Between a Control Hub and Locally Powered Epidural Stimulators .................................................................................214 Clemens Eder, Virgilio Valente, Nick Donaldson, Andreas Demosthenous University College London, United Kingdom

C1P-C.5

An Implantable Bio-Micro-System for Drug Monitoring ........................................218 Sara S. Ghoreishizadeh, Enver G. Kilinc, Camilla Baj-Rossi, Catherine Dehollain, Sandro Carrara, Giovanni De Micheli École Polytechnique Fédérale de Lausanne, Switzerland

C1P-C.6

Single-Wire RF Transmission Lines for Implanted Devices..................................222 Jordan Besnoff, Matthew Reynolds Duke University, United States

C1P-C.7

Ultra-Low Power Frequency and Duty-Cycle Modulated Implantable Pressure-Temperature Sensor.................................................................................226 Ahmad Khairi1, Chenyang Wu1, Yoed Rabin1, Gary Fedder1, Jeyanandh Paramesh1, David Schwartzman2 1 Carnegie Mellon University, United States; 2University of Pittsburgh Medical Center, United States

C1P-C.8

Selecting a Safe Power Level for an Indoor Implanted UWB Wireless Biotelemetry Link ......................................................................................................230 Kerron Duncan, Ralph Etienne-Cummings Johns Hopkins University, United States

C1P-C.9

A Fully Reconfigurable Low-Noise Biopotential Sensing Amplifier.....................234 Tzu-Yun Wang2, Min-Rui Lai2, Christopher Twigg1, Sheng-Yu Peng2 1 Binghamton University, United States; 2National Taiwan University of Science and Technology, Taiwan

C1P-C.10 Area-Power-Efficient 11-Bit SAR ADC with Delay-Line Enhanced Tuning for Neural Sensing Applications....................................................................................238 Teng-Chieh Huang2, Po-Tsang Huang2, Shang-Lin Wu2, Kuan-Neng Chen2, JinChern Chiou2, Kuo-Hua Chen1, Chi-Tsung Chiu1, Ho-Ming Tong1, Ching-Te Chuang2, Wei Hwang2 1 Advanced Semiconductor Engineering Group, Taiwan; 2National Chiao Tung University, Taiwan

C1P-C.11 A Fully Analog Low-Power Wavelet-Based Hearing Aid Front-End .....................242 Jose E. G. Medeiros2, Lucas A. P. Chrisóstomo3, Gabriela Meira3, Yuri C. R. Toledo1, Matheus Pimenta1, Sandro Haddad2 1 DFchip, Brazil; 2Universidade de Brasília, Brazil; 3University of Brasilia, Brazil C1P-C.13 New Non Invasive Doppler Technology for Carotid Stenosis Assessment.........246 Daniele Righi1, Gabriele Ciuti1, Walter Dorigo1, Leonardo Forzoni2, Sara D'Onofrio2, Piero Tortoli3 1 Azienda Ospedaliero–Universitaria Careggi - Firenze, Italy; 2Esaote S.p.A., Italy; 3 Università degli Studi di Firenze, Italy C1P-C.14 A Complete 256-Channel Reconfigurable System for In Vitro Neurobiological Experiments...................................................................................250 Miroslaw Zoladz1, Zoladz Kmon1, Jacek Rauza1, Pawel Grybos1, Tomasz Kowalczyk2, Bartosz Caban2 1 AGH University of Science and Technology, Poland; 2University of Lódz, Poland C1P-C.15 Modeling a Safety-Related System for Continuous Non-Invasive Blood Pressure Monitoring .................................................................................................254 Huiyun Sheng1, Michael Schwarz2, Josef Börcsök1 1 Universität Kassel, Germany; 2University Kassel, Germany C1P-C.17 Motion Artifact Reduction in EEG Recordings Using Multi-Channel Contact Impedance Measurements .......................................................................................258 Alexander Bertrand2, Vojkan Mihajlovic1, Bernard Grundlehner1, Chris Vanhoof1, Marc Moonen2 1 Holst Centre / IMEC, Netherlands; 2Katholieke Universiteit Leuven, Belgium C1P-C.20 Automatic Detection of Sleep Spindles Using Teager Energy and Spectral Edge Frequency ........................................................................................................262 Syed Anas Imtiaz, Siavash Saremi-Yarahmadi, Esther Rodriguez-Villegas Imperial College London, United Kingdom C1P-C.21 Design, Realisation and Validation of Microfluidic Stochastic Mixers Integrable in Bioanalytical Systems Using CFD Modeling ....................................266 Eszter Tóth3, Kristóf Iván3, Péter Fürjes2, Zoltán Fekete2, Eszter Holczer1 1 Budapest University of Technology and Economics, Hungary; 2Institute for Technical Physics and Materials Science of the Hungarian Academy of Sciences, Hungary; 3Pázmány Péter Catholic University, Hungary C1P-C.22 Low-Distortion Super-GOhm Subthreshold-MOS Resistor for CMOS Neural Amplifiers...................................................................................................................270 Hossein Kassiri, Karim Abdelhalim, Roman Genov University of Toronto, Canada

C2L-A

Implantable Electronics

Time: Place: Chair(s):

Saturday, November 2, 2013, 11:30 - 13:00 16th floor Julius Georgiou, University of Cyprus Shuenn-Yuh Lee, National Chung Cheng University

C2L-A.1

An Implantable Microsystem for Long-Term Study on the Mechanism of Deep Brain Stimulation.............................................................................................274 Yu-Po Lin, Hung-Chih Chiu, Pin-Yang Huang, Zong-Ye Wang, Hsiang-Hui Cheng, Po-Chiun Huang, Kea-Tiong Tang, Hsi-Pin Ma, Hsin Chen National Tsing Hua University, Taiwan

C2L-A.2

A Signal-Specific Approach for Reducing SAR-ADC Power Consumption.........278 Ken Chiang, N. Sertac Artan, H. Jonathan Chao Polytechnic Institute of New York University, United States

C2L-A.3

Miniature Implantable Telemetry System for Pressure-Volume Cardiac Monitoring..................................................................................................................282 Kyle Fricke, Robert Sobot Western University, Canada

C2L-A.4

A CMOS MedRadio-Band Low-Power Integer-N Cascaded Phase-Locked Loop for Implantable Medical Socs .........................................................................286 Yu-Yu Liao, Wei-Ming Chen, Chung-Yu Wu National Chiao Tung University, Taiwan

C2L-A.5

Amplitude-Engraving Modulation (AEM) Scheme for Simultaneous Power and High-Rate Data Telemetry to Biomedical Implants .........................................290 Reza Erfani, Amir Masoud Sodagar K.N. Toosi University of Technology, Iran

C2L-B

Neuromorphic Circuits and Systems

Time: Place: Chair(s):

Saturday, November 2, 2013, 11:30 - 13:00 15th floor Kea-Tiong Tang, National Tsing Hua University Timothy G Constandinou, Imperial College London

C2L-B.1

Computation Using Mismatch: Neuromorphic Extreme Learning Machines ......294 Enyi Yao, Shaista Hussain, Arindam Basu, Guang-Bin Huang Nanyang Technological University, Singapore

C2L-B.2

A Spiking Neural Network Architecture for Visual Motion Estimation.................298 Garrick Orchard2, Ryad Benosman3, Ralph Etienne-Cummings1, Nitish Thakor2 1 Johns Hopkins University, United States; 2National University of Singapore, Singapore; 3Vision Institute, University Pierre and Marie Curie, France

C2L-B.3

Hardware Efficient, Neuromorphic Dendritically Enhanced Readout for Liquid State Machines ..............................................................................................302 Subhrajit Roy, Arindam Basu, Shaista Hussain Nanyang Technological University, Singapore

C2L-B.4

Real-Time Motion Estimation Using Spatiotemporal Filtering in FPGA...............306 Garrick Orchard2, Nitish Thakor2, Ralph Etienne-Cummings1 1 Johns Hopkins University, United States; 2National University of Singapore, Singapore

C2L-B.5

A Dual Operation Mode Bio-Inspired Vision Sensor..............................................310 Juan Antonio Leñero-Bardallo, Philipp Häfliger University of Oslo, Norway

C3L-A

Biomedical Circuits

Time: Place: Chair(s):

Saturday, November 2, 2013, 14:10 - 15:40 16th floor Timothy G Constandinou, Imperial College London Herming Chiueh, National Chiao Tung University

C3L-A.1

A Low-Power, Low-Noise, and Low-Cost VGA for Second Harmonic Imaging Ultrasound Probes ....................................................................................................314 Peng Wang2, Trond Ytterdal2, Thomas Halvorsrød1 1 GE Vingmed Ultrasound AS, Norway; 2Norwegian University of Science and Technology, Norway

C3L-A.2

A Compact Gm-C Filter Architecture with an Ultra-Low Corner Frequency and High Ground-Noise Rejection...........................................................................318 Yu-Chieh Lee2, Wen-Yang Hsu2, Tai-Ting Huang1, Hsin Chen2 1 Industrial Technology Research Institute, Taiwan; 2National Tsing Hua University, Taiwan

C3L-A.3

A 430nW 64nV/Square Root Hertz Current-Reuse Telescopic Amplifier for Neural Recording Applications................................................................................322 Shuang Song, Michaël Rooijakkers, Pieter Harpe, Chiara Rabotti, Massimo Mischi, Arthur van Roermund, Eugenio Cantatore Eindhoven University of Technology, Netherlands

C3L-A.4

An Energy Efficient Inverter Based Readout Circuit for Capacitive Sensor........326 Thanh Trung Nguyen, Philipp Häfliger University of Oslo, Norway

C3L-B

Stimulation and Sensing

Time: Place: Chair(s):

Saturday, November 2, 2013, 14:10 - 15:40 15th floor Ralph Etienne-Cummings, Johns Hopkins University Manuel Delgado, IMSE-CNM

C3L-B.1

Bio-Feedback Iontophoresis Patch for Controllable Transdermal Drug Delivery ......................................................................................................................330 Kiseok Song, Unsoo Ha, Jaehyuk Lee, Hoi-Jun Yoo Korea Advanced Institute of Science and Technology, Korea, South

C3L-B.2

Optimization of Neural Stimulation in a Device for Treating Urinary Incontinence ..............................................................................................................334 Arsam N. Shiraz, Anne Vanhoestenberghe, Andreas Demosthenous University College London, United Kingdom

C3L-B.3

Characterization of a Non Linear Fractional Model of Electrode-Tissue Impedance for Neuronal Stimulation.......................................................................338 Florian Kölbl1, Jocelyn Sabatier2, Gilles N'Kaoua1, Frederic Naudet1, Emilie Faggiani1, Abdelhamid Benazzouz1, Sylvie Renaud1, Noëlle Lewis1 1 Université de Bordeaux, France; 2University of Bordeaux, France

C3L-B.4

A CMOS Micromachined Capacitive Tactile Sensor with Compensation of Process Variations ....................................................................................................342 Hao-Cheng Tsai, Tien-Keng Wu, Tsung-Heng Tsai National Chung Cheng University, Taiwan

C3L-B.5

Design Optimization of the Optical Receiver in Transcutaneous Telemetric Links ...........................................................................................................................346 Tianyi Liu, Jens Anders, Maurits Ortmanns Universität Ulm, Germany

C4L-A

Wireless Systems

Time: Place: Chair(s):

Saturday, November 2, 2013, 16:00 - 17:30 16th floor Ge Tong, Nanyang Technological University Viktor Owall, Lunds Tekniska Högskola

C4L-A.1

Cellular Inductive Powering System for Weakly-Linked Resonant Rodent Implants......................................................................................................................350 Nima Soltani, Miaad Aliroteh, Roman Genov University of Toronto, Canada

C4L-A.2

Mechanical Power Assessment of Fast Eye Motions for Energy Harvesting in Autonomous Intraocular Implants.......................................................................354 Daniel Laqua, Stefan Pollnow, Miriam Stadthalter, Jan Fischer, Martin Weis, Peter Husar Technische Universität Ilmenau, Germany

C4L-A.3

A Sub-Threshold Voltage Ladder Rectifier for Orthogonal Current-Reuse Neural Amplifier.........................................................................................................358 Changhyuk Lee, Ben Johnson, Alyosha Molnar Cornell University, United States

C4L-A.4

Design and Evaluation of a Novel Wireless Reconstructed 3-Lead ECG Monitoring System ....................................................................................................362 Yishan Wang, Ralf Wunderlich, Stefan Heinen Rheinisch-Westfälische Technische Hochschule Aachen, Germany

C4L-A.5

A Low-Power Robust GFSK Demodulation Technique for WBAN Applications...............................................................................................................366 Pengpeng Chen1, Bo Zhao1, Rong Luo1, Yong Lian2, Huazhong Yang1 1 Tsinghua University, China; 2York University, Canada

C4L-B

Biomedical Systems

Time: Place: Chair(s):

Saturday, November 2, 2013, 16:00 - 17:30 15th floor Julius Georgiou, University of Cyprus Wai-Chi Fang, National Chiao Tung University

C4L-B.1

Fuzzy Logic, an Intermediate Description Level for Design and Simulation in Synthetic Biology ......................................................................................................370 Yves Gendrault, Morgan Madec, Vincent Wlotzko, Christophe Lallement, Jacques Haiech Laboratoire ICube, France

C4L-B.2

EDA Inspired Open-Source Framework for Synthetic Biology .............................374 Morgan Madec2, François Pecheux1, Yves Gendrault2, Loic Bauer2, Jacques Haiech2, Christophe Lallement2 1 Laboratoire d’Informatique Paris VI, France; 2Laboratoire ICube, France

C4L-B.3

Predicting Effective Drug Combinations via Network Propagation .....................378 Balázs Ligeti2, Roberto Vera1, Gergely Lukács2, Balázs Gyórffy3, Sándor Pongor2 1 International Centre for Genetic Engineering and Biotechnology, Italy; 2Pázmány Péter Catholic University, Hungary; 3Research Laboratory of Pediatrics and Nephrology, Hungarian Academy of Sciences, Hungary

C4L-B.4

An Efficient Lossless Data Compression Method Based on ExponentialGolomb Coding for Biomedical Information and its Implementation Using ASIP Technology.......................................................................................................382 Shoko Nakatsuka, Takashi Hamabe, Yoshinori Takeuchi, Masaharu Imai Osaka University, Japan

C4L-B.5

New Bipolar and Hybrid Argon Plasma Coagulation Technologies Enable Improved Electrosurgical Results ...........................................................................386 Daniel Friedrichs, James Gilbert, Joe Sartor Covidien Ltd., United States