2366 IEEE JOURNAL OF SOLID-STATE CIRCUITS, VOL. 45, NO. 11, NOVEMBER 2010

Transcription

1 2366 IEEE JOURNAL OF SOLID-STATE CIRCUITS, VOL. 45, NO. 11, NOVEMBER 2010 A Low-Energy Inductive Coupling Transceiver With Cm-Range 50-Mbps Data Communication in Mobile Device Applications Seulki Lee, Student Member, IEEE, Kiseok Song, Student Member, IEEE, Jerald Yoo, Member, IEEE, and Hoi-Jun Yoo, Fellow, IEEE Abstract A low-energy inductive coupling transceiver is proposed for Cm-range multimedia data transmission in mobile device applications. The Transmission Time Control (TTC) scheme is proposed to reduce the transmitter energy consumption to pj/b, and the Adaptive Gain Control (AGC) scheme is adopted to make the receiver energy consumption be pj/b. The planar-type inductor with self-resonance frequency of about 200 MHz fabricated on the flexible substrate achieves a data rate over 50 Mbps. To compensate for the weakly coupled channel, the receiver sensitivity is enhanced by the differential detection method (DDM) of the nodal voltages across the receiver inductor. With this method, the communication distance is increased up to 7 cm, and channel misalignment tolerance is enhanced up to 2 cm. The proposed transceiver is implemented within mm 2 in m CMOS process and operates with 1-V supply. Index Terms Inductive coupling, low-power transceiver, mobile device communication, short-range wireless communication. I. INTRODUCTION T HESE days, personal mobile devices such as cellular phones or PDAs are widely used in daily life, and proximity communication between them also becomes popular to conveniently transmit video, audio, graphics, or text-type data. Since their data file size is usually large over 10 MB, the proximity communication in mobile devices should satisfy more stringent requirements. It should support high-speed communication because people are usually impatient with long data transfer time. Also, low energy consumption is highly required because personal mobile devices are battery-powered. In particular, with standard battery capacitance of 1000 mah in current cellular phones, less than 2 pj/b energy consumption is recommended not to affect the battery lifetime even for transmission of 100 MB data. Finally, centimeter-range (Cm-range) distance between mobile devices should be guaranteed to Manuscript received February 05, 2010; revised May 06, 2010; accepted June 24, Date of current version October 22, This paper was approved by Guest Editor Mototsugu Hamada. This work was supported by the Midcareer Researcher Program through the National Research Foundation of Korea (NRF) under Grant funded by the Korean government (MEST). S. Lee, K. Song, and H.-J. Yoo are with the Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon , Korea ( sklee@eeinfo.kaist.ac.kr; sks8795@eeinfo.kaist.ac.kr; hjyoo@ee.kaist.ac.kr). J. Yoo is with Microsystems Engineering, Masdar Institute, Abu Dhabi, United Arab Emirates, and also with the Microsystems Technology Laboratory, Massachusetts Institute of Technology, Cambridge, MA USA ( jerald@mit.edu). Digital Object Identifier /JSSC achieve both reliable and safe communication in this application. In general, Infrared Data Association (IrDA) or Bluetooth is used for phone-to-phone proximity data communication. Although IrDA recently achieved high-speed data communication up to 1 Gbps [1], it still suffers from a line-of-sight (LoS) problem. The two communicating mobile devices should be fixed in their locations during data transmission, making the immediate usage very inconvenient. Bluetooth consumes more than 10 mw, which is not proper for a mobile large-file data communication. Since these conventional techniques consume battery power rapidly to transmit large data, it is reasonable to use inductive coupling communication for Cm-range proximity communication. Some inductive coupling transceivers were introduced previously for Cm-range data communication [2] [5], but none of them showed either low energy consumption of few pj/b or high data rate over tens of Mbps. Therefore, we propose an inductive coupling transceiver for low energy and high data rate. In the proposed transceiver, the transmission time control (TTC) scheme in the transmitter and the adaptive gain control (AGC) scheme in the receiver are adopted to achieve the low energy consumption of and pj/b, respectively. The data rate can be reliably increased up to 50 Mbps by adopting the planar inductor with high self-resonance frequency of about 200 MHz on a flexible substrate as a communication channel. However, since the inductance value of this inductor is smaller than that of [3] and [4], the differential detection method (DDM) of the nodal voltages across the receiver inductor is proposed in order to compensate for the weak coupling strength in the communication channel. Due to this method, the communication becomes more robust with enhanced receiver sensitivity. The remainder of this paper is organized as follows. In Section II, channel inductor selection for the proposed transceiver will be discussed. Section III describes the TTC transmitter, and Section IV explains the AGC receiver. The proposed DDM for high sensitivity will be explained in Section V. Then, the implementation results will be shown in Section VI. Finally, conclusions will be made in Section VII. II. CHANNEL INDUCTOR SELECTION The communication channel of the proposed transceiver consists of a pair of inductors. Since the frequency characteristics of the channel mainly affect the maximum data rate of the transceiver, the channel inductor should be selected to have high self-resonance frequency. Especially in pulse-based inductive /$ IEEE

2 LEE et al.: LOW-ENERGY INDUCTIVE COUPLING TRANSCEIVER WITH CM-RANGE 50-MBPS DATA COMMUNICATION IN MOBILE DEVICE APPLICATIONS 2367 of the channel between two mobile devices. From (1), the following condition is derived: are all in millimeters (2) Fig. 1. Planar inductor with its four design parameters. coupling transmission, the data recovery can be correct when the self-resonance frequency of the channel inductor is at least twice the data rate [6]. Moreover, mobile devices are becoming thinner and thinner so that its size will come to be as thick as a credit card and it is fabricated on a flexible substrate [7]. To satisfy this trend in the form factor of mobile devices, a planar-type inductor formed thinly on the flexible substrate is adopted in this work. A. Planar Inductor Fig. 1 shows the octagonal structure of the planar inductor. It has four design parameters: wire width, wire spacing, the number of turns, and inner diameter of the inductor. Other general parameters such as outer diameter, average diameter, or fill ratio [2], [8] can be expressed using,,, and as in According to (1), there is a correlation between each parameter of the inductor. For example, the number of turns cannot be decreased without increasing if,, and are fixed; or, once is determined to have a specific value, should be increased in order to get smaller, i.e., higher self-resonance frequency [8], with fixed and. The size of the channel inductor for mobile device communication is determined by two reasons: the first is communication distance, and the other is mobile device size. Since the inductor size should be increased to the same order of communication distance to achieve higher received signal strength [9], it needs cm-order size. Also, with the consideration of the current mobile phone size, the parameter of the channel inductor is determined to be 3 cm in this work. If the inductor size is much smaller than the mobile device size, the immediate usage is disturbed by misalignment (1) Fig. 2 shows the measurement result of the relationship between self-resonance frequency and design parameters for the planar inductor. The curve in Fig. 2(a) is measured with the case of 30, 0.5, and 0.5 mm for,, and, respectively, while the values of two turns, 0.5 and 0.5 mm, are used for,, and in Fig. 2(b). When,, and are fixed, self-resonance frequency of the inductor is inversely proportional to both and, specifically, the square of [10]. Since and the square of affects the self-resonance frequency [10], reduces it more significantly than. In other words, minimizing would be more effective to achieve higher self-resonance frequency. Also, more than 100 nh of inductance is needed for reliable data transmission. This is because, with less than a 100-nH inductor, the data cannot be recovered at the receiver distant over the cm-range due to its weak coupling strength. The relationship between inductance and is shown in Fig. 3, and it is obvious that a one-turn inductor cannot give enough inductance value. Therefore, and in this work are set to 2 and 29 mm, respectively. Fig. 4 shows the inductor used in this work and its impedance characteristics in the MHz frequency range. Its inductance value is measured as 0.33 H, and MHz of the self-resonance frequency is achieved, which can give enough bandwidth for over 50 Mbps of data transaction. B. Inductor Channel Two similar inductors in Fig. 4 are employed to form a communication channel. The transfer characteristics of the channel as a function of the frequency are measured as shown in Fig. 5. Since the communication between mobile devices is assumed in this work, the channel medium for data transmission will become the case material, which is plastic, not air. Thus, at first, the transfer characteristics within the plastic medium are measured, and air medium is measured as a reference. From the measurement, the channel loss in plastic medium is 1 2 db more than air medium in a low-frequency region of less than 220 MHz, but not affecting that much to the frequency characteristics. Also, in a high-frequency region over 250 MHz, the channel loss in air medium is more than 5 db larger than that of the plastic medium. Since our frequency region of interest is a low-frequency region below 200 MHz, the transceiver for this application should support higher sensitivity than the air-medium case for reliable communication. Also, to enhance the sensitivity, the DDM is proposed in the receiver, as will be described in Section V. III. TTC TRANSMITTER Fig. 6 shows the architecture of the proposed transmitter. It consists of the conventional H-bridge circuit and the proposed TTC circuit which is connected in parallel with an H-bridge circuit. Since the most important issue for transmitter design in mobile device applications is low energy consumption, the H-bridge architecture with the TTC scheme is proposed in order

3 2368 IEEE JOURNAL OF SOLID-STATE CIRCUITS, VOL. 45, NO. 11, NOVEMBER 2010 Fig. 2. Measurement result of the relationship between self-resonance frequency (f ) and design parameters. (a) Number of turns n. (b) Inner diameter d. Fig. 3. Relationship between inductance and the number of turns (n). Fig. 5. Frequency characteristics of the air and plastic medium channel (measured at 3-cm channel distance). Fig. 4. Planar inductor employed in this work. (a) Photograph and design parameters. (b) Impedance characteristics. Fig. 6. Transmitter architecture. to increase the data rate and reduce power consumption at once. The H-bridge architecture is generally used in high-speed transmitters for inductive coupling communication [2], [3], [11], but it consumes relatively high power compared with a carrier-modulated transmitter [4]. The most power-consuming part of an H-bridge-based transmitter is the direct path formation from

4 LEE et al.: LOW-ENERGY INDUCTIVE COUPLING TRANSCEIVER WITH CM-RANGE 50-MBPS DATA COMMUNICATION IN MOBILE DEVICE APPLICATIONS 2369 Fig. 7. Operation of the TTC scheme. power supply to ground whenever data are transmitted. To suppress this unwanted power consumption, several techniques like pulse generation [2] or pulsewidth control [11] were proposed. However, it still consumes a few milliwatts, which is too much for mobile applications. In the inductive coupling communication, the received voltage is expressed by transmitted current and mutual inductance of the channel equation. From this relationship and [11], the received voltage magnitude is not changed if the pulse slew rate is kept the same. Therefore, if the transmission time is reduced without degrading the slew rate of the transmitted pulse, lower power consumption can be obtained without any attenuation of the received voltage in the receiver. Fig. 7 shows the operation of the proposed TTC scheme. When the signal across the transmit inductor becomes larger than the predetermined reference value of the comparator, the TTC scheme is enabled so that both sides of the transmit inductor are shorted together. At the same time, the control signal of nmoss in the H-bridge circuit, which is the output of the three-input OR gate, transits from 1 to 0 due to the output of the comparator in the TTC circuit. As a result, the time for direct path formation from power supply to ground can be shortened. This affects both power reduction and intersymbol interference (ISI) reduction, and ISI reduction leads to the high data rate. For correct operation, the reference voltages of the comparator in the TTC circuit of Fig. 6 should be varied, and Fig. 8 shows the circuits in detail. By using the magnitude difference between and, the output of the comparator is determined with (3). Fig. 9 summarizes the values of and for this comparator according to the transmit data, TXDATA. In the case of data 1 transmission, the TTC circuit should be enabled when the signal becomes larger than the positive reference value. In the case of data 0 transmission, it should be enabled when the signal becomes smaller than the reference value which is negative. All reference voltage generators used in this work are modified from those of [12]. To implement the bias circuit with low power consumption, the transistors are operated in the subthreshold region. Its current level is very low, about 1 A. There are four reference voltage generators in this work, two for the transmitter and another two for the receiver. Since this voltage generator circuit also has the supply-independent characteristics, it can support the more reliable operation to the proposed transceiver. By using the TTC scheme, the maximum data rate can be increased up to 50 Mbps, which is five times higher than that of [3]. Also, power consumption is reduced from 1 mw [2] to W at a 50-Mbps data rate. In total, 95.7% and 83.6% of energy is reduced compared with [2] and [3], respectively. Although the lowest energy inductive coupling transceiver is reported in [11], it is excluded from the performance comparison since its chip-stacking application and micrometer-range communication are totally differed from this work. IV. AGC RECEIVER In the mobile device applications like in this work, one of the most important issues for the user is convenience. For example, users want to carry them freely, and the data transaction must be accomplished even though there is some misalignment between two devices. To achieve more convenience, the transceiver should offer high sensitivity since misalignment or distance between two mobile devices reduces the received signal strength. Fig. 10 shows the architecture of the proposed receiver. It consists of a gain amplifier, a sense amplifier (SA), and the AGC circuit. Sensitivity of the proposed receiver is increased by the AGC scheme in order to make continuous communication possible even if the devices are slightly moved by users during the data transaction. To optimize both power consumption and communication reliability, two receiver modes are proposed: one is the low-power (LP) mode, and the other is the high-sensitivity (HS) mode as shown in Fig. 11. In LP mode operation, the AGC scheme disables the gain amplifier, and the direct path from the receiver inductor to the SA is formed to achieve low power consumption. By contrast, the AGC scheme enables the gain amplifier in HS mode. The received signal magnitude becomes larger since all of them are passed through the gain amplifier, but the power consumption is also increased compared with the LP mode. Compared with [2] and [3], with the help of the AGC, the proposed receiver can detect much a smaller signal in HS mode without increasing its power consumption in the LP mode. The operation mode change of the proposed receiver is shown in Fig. 12. At first, the receiver always operates in the LP mode. As the communication distance between the transmitter and receiver inductor becomes longer or the misalignment between two inductors becomes larger, the maximum signal amplitude across the receiver inductor,, is gradually decreased. At a distance of 5 cm, the value of becomes smaller than the predetermined reference value. Thus, the operation mode

5 2370 IEEE JOURNAL OF SOLID-STATE CIRCUITS, VOL. 45, NO. 11, NOVEMBER 2010 Fig. 8. Comparator in the TTC circuit. However, both the communication distance and misalignment tolerance are worse than the HS mode, since its sensitivity is not sufficient to recover a very small signal. In the HS mode, although the receiver consumes slightly more power, 65.2 W, the communication distance and misalignment tolerance can be increased up to 7 and 2 cm, which are 29% and 25% enhancements compared with the LP mode, respectively, due to its higher sensitivity. Fig. 9. Summary of reference values for TTC circuit. Fig. 10. Receiver architecture. changes into the HS mode, and AGC enables the gain amplifier. The data can be transmitted correctly even for the switching period from the LP mode to the HS mode because is initially set to a larger value than the operating limitation of the sense amplifier. Since the value of is also decreased with increasing communication distance in the HS mode, the signal can be recovered correctly within 7 cm of communication distance, even in the HS mode. By using these two operation modes in the receiver, it can achieve either LP consumption or HS according to the communication environment. In the LP mode, the receiver power consumption is W, which is a 37% reduction from HS mode. V. HS DDM As explained in Section II, the channel medium for the mobile device communication is determined by the case material of the devices. Also, since the channel loss of the plastic medium channel is slightly higher than that of the air medium channel in our frequency region of interest below 200 MHz, the DDM is proposed for higher sensitivity in the receiver. The DDM scheme can be implemented by connecting two inputs of the SA to the receiver inductor nodes in the LP mode or the outputs of the gain amplifier in the HS mode as shown in Fig. 11. Fig. 13(a) shows the SA circuit which is used in the proposed DDM scheme. It basically operates just as a conventional SA, but it adopts some logic gates in order to reduce power consumption. When the clock signal CLK goes low, the two differential outputs of the sense amplifier, and,go to the precharge period so both values are 1. When CLK is high, the outputs are determined by the magnitude relationship between two inputs, and. Once values of and are determined, the cross-coupled structure in the sense amplifier stops its operation. Therefore, a XOR gate and an AND gate are connected as shown in Fig. 13(a) in order to reduce the power consumption after the full development of outputs. Since and always have the opposite values each other, the output of XOR gate must be 1 whenever the output determined. Thus, the proposed SA circuit detects the time when the determination is finished and turns off the nmos transistor in cross-coupled structure. Its final outputs are and, which are the outputs of two D-flip-flops. They are edge-triggered and their values are not changed even though nmos is turned off while CLK is low. Fig. 13(b) shows

6 LEE et al.: LOW-ENERGY INDUCTIVE COUPLING TRANSCEIVER WITH CM-RANGE 50-MBPS DATA COMMUNICATION IN MOBILE DEVICE APPLICATIONS 2371 Fig. 11. Two modes of the proposed receiver. (a) LP mode. (b) HS mode. Fig. 12. Operation mode change of the proposed receiver. the output determination table according to, and CLK. As mentioned above, both and are high if CLK is low (precharge period). And the data determination is performed while only CLK is high. If is larger than at the point of comparison, then becomes high while becomes low. In contrast, if is larger than, then becomes high and becomes low. Fig. 13(c) shows the timing diagram of the proposed SA. This proposed DDM scheme replaces the Schmitt triggers used in [2] and [3], and it shows higher sensitivity. Since the Schmitt trigger uses absolute reference voltage to separate data 0 and 1 and it is just fixed before data communication, it cannot deal with longer communication distance in which the received signal for data 1 is smaller than reference voltage. However, the proposed scheme can recover the received data correctly if the difference between two nodes of the received signal is larger than operation limitation of the SA. Thus, the sensitivity can be improved over 16% compared with the simulation result of [2]. VI. IMPLEMENTATION RESULTS Fig. 14 shows the chip micrograph of the proposed inductive coupling transceiver. The die size is 1.5 mm 2.37 mm including pads in a m CMOS technology. Successful data transaction of the proposed transceiver at 50-Mbps data rate is shown in Fig. 15. The communication distance between transmitter and receiver inductor is 3 cm in this case, and they are aligned to each other. For every low period of receiver clock, the output data of the receiver become high since pmoss in the Fig. 13. SA. (a) Circuit. (b) Output determination table. (c) Timing diagram.

7 2372 IEEE JOURNAL OF SOLID-STATE CIRCUITS, VOL. 45, NO. 11, NOVEMBER 2010 Fig. 14. Chip microphotograph. Fig. 15. Transceiver operation at 3-cm communication distance. Top: transmitted clock. Middle: transmitted data. Bottom: recovered data. SA are turned on by the receiver clock. It affects the recovered data to have a 50% duty for data 1 while it always remains low for data 0 as shown in Fig. 15. The BER measurement results are plotted in Fig. 16. The BER at 3-cm communication distance is two times lower than that of [4]. With the proposed AGC scheme, BER characteristics are improved for communication over 5 cm. In the HS mode, 7 cm is measured as the maximum communication distance when, which is 57% longer than in [4] and increased by 16% over the simulation result of [2]. Also, the proposed DDM scheme increases the misalignment tolerance. As shown in Fig. 16(b), the transceiver itself can compensate for 1.5 cm of misalignment, and the gain amplifier in HS mode can make it up additional 0.5 cm of misalignment. Compared with 1 cm of misalignment tolerance in [4], the amount of tolerance is increased twice. The enhanced tolerance can support the convenient usage for users. Table I summarize the performance of the proposed inductive coupling transceiver chip. All blocks operate with 1-V supply, and the power consumption of the complete transceiver is 65 W (23.74 W for the transmitter, and W for the receiver in LP mode) when the data rate is 50 Mbps. The total energy consumption is and pj/b for the transmitter and the receiver, respectively. The performance comparison with the previous works is summarized in Table II. With the help of three proposed schemes, TTC, AGC, and DDM, the proposed inductive coupling transceiver chip consumes the lowest energy in both the transmitter and the receiver with the longest communication distance. Since the proposed transceiver chip operates as either the transmitter or the receiver and does not operates as both at the same time, Fig. 16. BER measurement result. (a) Communication distance. (b) Inductor channel misalignment. TABLE I PERFORMANCE SUMMARY the effective maximum energy consumption is pj/b in LP mode of the receiver. VII. CONCLUSION An inductive coupling transceiver for high-speed data transaction in mobile device applications is proposed and implemented in m CMOS process. The TTC scheme of the transmitter and the AGC scheme of the receiver make the transmitter and the receiver consume only and pj/b of energy, respectively, under a 1-V supply voltage. Moreover, adopting an inductor channel made of the planar inductors on the flexible substrate with around 200-MHz self-resonance frequency improves the data rate up to 50 Mbps. The DDM of nodal voltages across the receiver inductor increases the receiver sensitivity. Finally, by employing the AGC scheme,

8 LEE et al.: LOW-ENERGY INDUCTIVE COUPLING TRANSCEIVER WITH CM-RANGE 50-MBPS DATA COMMUNICATION IN MOBILE DEVICE APPLICATIONS 2373 TABLE II PERFORMANCE COMPARISON the receiver can increase the communication distance and misalignment tolerance in the HS mode up to 7 and 2 cm, respectively, without increasing power consumption in LP mode. The transceiver consumes only 65 W in total with 50-Mbps data rate. REFERENCES [1] Development of 1 Gbit/s infrared communication technology for mobile devices, J. Inst. Electron., Inf. Commun. Eng., vol. 91, no. 5, pp , May 2008, News Analysis. [2] S. Lee, J. Yoo, and H.-J. Yoo, A 200 Mbps 0.02 nj/b dual-mode inductive coupling transceiver for cm-range multimedia interconnection, IEEE Trans. Circuits Syst. I, Reg. Papers, vol. 56, no. 5, pp , May [3] J. Yoo, S. Lee, and H.-J. Yoo, A 1.12 pj/b resonance compensated inductive transceiver with a fault-tolerant network switch for multilayer wearable body area network applications, IEEE J. Solid-State Circuits, vol. 44, no. 11, pp , Nov [4] S. Lee, J. Yoo, H. Kim, and H.-J. Yoo, A dynamic real-time capacitor compensated inductive coupling transceiver for wearable body sensor network, in Symp. VLSI Circuits Dig. Tech. Papers, Jun. 2009, pp [5] D. Guermandi, S. Gambini, and J. Rabaey, A 1 V 250 kpps 90 nm CMOS pulse based transceiver for cm-range wireless communication, in Proc. IEEE Eur. Solid-State Circuits Conf., Sep. 2007, pp [6] N. Miura, D. Mizoguchi, M. Inoue, K. Niitsu, Y. Nakagawa, M. Tago, M. Fukaishi, T. Sakurai, and T. Kuroda, A 1 Tb/s 3W inductive-coupling transceiver for 3D-stacked inter-chip clock and data link, IEEE J. Solid-State Circuits, vol. 42, no. 1, pp , Jan [7] A. Rakotonirainy, Trends and future of mobile computing, in Proc. 10th Int. Workshop Database Expert Syst. Appl., Sep. 1999, pp [8] J. Yoo, Wearable body sensor network SoC for continuous health monitoring, Ph.D. dissertation, Dept. of Elect. Eng., KAIST, Daejeon, Korea, [9] H. Ishikuro, N. Miura, and T. Kuroda, Wideband inductive-coupling interface for high-performance portable system, in Proc. IEEE Custom Integr. Circuits Conf., 2007, pp [10] S. S. Mohan, The design, modeling, and optimization of on-chip inductor and transformer circuits, Ph.D. dissertation, Dept. of Electron. Eng., Stanford Univ., Stanford, CA, [11] N. Miura, H. Ishikuro, K. Niitsu, T. Sakurai, and T. Kuroda, A 0.14 pj/b inductive-coupling transceiver with digitally-controlled precise pulse shaping, IEEE J. Solid-State Circuits, vol. 43, no. 1, pp , Jan [12] G. D. Vita and G. Iannaccone, A sub-1 V, 10 ppm/ C, nanopower voltage reference generator, IEEE J. Solid State Circuits, vol. 42, no. 7, pp , Jul [13] S. Lee, J. Yoo, K. Song, and H.-J. Yoo, A 1.3 pj/b inductive coupling transceiver with adaptive gain control for cm-range 50 Mbps data communication, in Proc. IEEE Asian Solid-State Circuits Conf., 2009, pp Seulki Lee (S 07) received the B.S. and M.S. degrees from the Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, in 2007 and 2009, respectively, in electrical engineering, where she is currently working toward the Ph.D. degree in electrical engineering. Her current research interests include the inductive coupling transceiver design and near-field communication for wearable computing applications. Kiseok Song (S 09) received the B.S. degree in electrical engineering and computer science from the Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, in 2009, where he is currently working toward the M.S. degree in electrical engineering. His current research interests include wirelessly powered stimulators and body channel communication. Jerald Yoo (S 05 M 10) received the B.S., M.S., and Ph.D. degrees from the Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, in 2002, 2007, and 2010, respectively. In May 2010, he joined the faculty of Microsystems Engineering, Masdar Institute, Abu Dhabi, United Arab Emirates, where he is an Assistant Professor. He is currently also with the Microsystems Technology Laboratory, Massachusetts Institute of Technology, Cambridge, as a Visiting Scholar. As a Chief Researcher with the Semiconductor System Laboratory, KAIST, he developed low-energy body-area network (BAN) transceivers and wearable body sensor network using planar-fashionable circuit board (P-FCB) for continuous health monitoring systems. He is an author of a book chapter in Biomedical CMOS ICs (Springer, 2010). His research focuses on low-energy circuit technology for wearable bio signal sensors, wirelesspower transmission, SoC design to system realization for wearable healthcare applications, and energy-efficient biomedical circuit techniques. Dr. Yoo was a corecipient of the Asian Solid-State Circuits Conference Outstanding Design Award in 2005.

9 2374 IEEE JOURNAL OF SOLID-STATE CIRCUITS, VOL. 45, NO. 11, NOVEMBER 2010 Hoi-Jun Yoo (M 95 SM 04 F 08) received the B.S. degree in electronics from Seoul National University, Seoul, Korea, in 1983, and the M.S. and Ph.D. degrees in electrical engineering from the Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, in 1985 and 1988, respectively. His Ph.D. work concerned the fabrication process for GaAs vertical optoelectronic integrated circuits. From 1988 to 1990, he was with Bell Communications Research, Red Bank, NJ, where he invented the 2-D phased-locked VCSEL array, the front-surface-emitting laser, and the high-speed lateral HBT. In 1991, he became a Manager of the DRAM Design Group, Hyundai Electronics, and designed a family of fast-1m DRAMs and 256M synchronous DRAMs. In 1998, he joined the faculty of the Department of Electrical Engineering at KAIST and now is a Full Professor. From 2001 to 2005, he was the director of System Integration and IP Authoring Research Center (SIPAC), funded by the Korean government to promote worldwide IP authoring and its SoC application. From 2003 to 2005, he was the full-time Advisor to the Minister of the Korea Ministry of Information and Communication and National Project Manager for SoC and Computer. In 2007, he founded System Design Innovation & Application Research Center (SDIA), KAIST, to research and develops SoCs for intelligent robots, wearable computers, and biosystems. He is the author of the books DRAM Design (Hongleung, 1996, in Korean) and High Performance DRAM (Sigma, 1999, in Korean) and chapters of Networks on Chips (Morgan Kaufmann, 2006). He is a member of the executive committee of ISSCC, Symposium on VLSI, and A-SSCC. He is the TPC chair of the A-SSCC His current interests are high-speed and low-power networks on chips, 3-D graphics, body area networks, biomedical devices and circuits, and memory circuits and systems. Dr. Yoo was the recipient of the Electronic Industrial Association of Korea Award for his contribution to DRAM technology the 1994, the Hynix Development Award in 1995, the Korea Semiconductor Industry Association Award in 2002, Best Research of KAIST Award in 2007, Design Award of 2001 ASP- DAC, and Outstanding Design Awards 2005, 2006, 2007 A-SSCC.