JOURNAL OF SEMICONDUCTOR TECHNOLOGY AND SCIENCE, VOL.17, NO.6, DECEMBER, 2017 ISSN(Print) 1598-1657 https://doi.org/10.5573/JSTS.2017.17.6.904 ISSN(Online) 2233-4866

Manuscript received Nov. 12, 2017; accepted Nov. 20, 2017 Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST) E-mail : krlee07@kaist.ac.kr

A 206.3 μW Non-contact Compensation IC for Body Channel Communication

Kyoung-Rog Lee, Jaeeun Jang, and Hoi-Jun Yoo

Abstract—A digital differential transmitter based on A first low power non-contact compensation IC for body channel communication (BCC) is proposed. The proposed IC is composed of contact status detection unit (CSDU), non-contact status compensation unit (NCU), and reconfigurable LNA. CSDU detects contact status of the IC with coupling capacitance sensing. NCSU compensates input signal with LC resonance by using inductor bank and capacitor bank. Reconfigurable LNA controls input impedance of the LNA for adaptive signal compensation. As a result, 14 dB signal compensation is achieved with 206.3 μW power consumption in 65 nm CMOS technology. Index Terms—Body channel characteristics, body channel communication (BCC), non-contact signal compensation, low-power signal compensation IC

I. INTRODUCTION

Nowadays, there are many wearable healthcare devices in our life. These devices should communicate with each other for sharing information, and ensure long operation time user’s convenience. To satisfy both requirements in wearable devices, researches for wireless communication scheme for wearable devices are emerging recently. Body channel communication (BCC) is one of the most remarkable communication solution for wearable devices with energy-efficient characteristics. BCC uses human body as a communication channel, so it

has less signal attenuation than other wireless communication schemes because of higher conductivity of human body [1]. Furthermore, BCC can use lower frequency modulation due to channel characteristics of human body, so it consumes much lower power than other wireless communication schemes. These features make BCC as low power, energy-efficient communication solution for wearable devices. With these advantages, there are many previous researches for BCC transceiver and system design [2-9].

In BCC system, low channel attenuation is one of the most important properties for low power consumption and energy-efficient characteristics. Channel attenuation in BCC consists of body channel attenuation and contact attenuation. In previous works, body channel attenuation is analyzed with equivalent circuit modeling [8], and contact attenuation is considered for transceiver design [4]. However, these works only assume well-contact status between electrodes and human body, or turn off the transceiver when electrodes are detached from human body [2]. This assumption can be limitation of BCC system because most wearable system cannot ensure well-contact status of electrodes with human body.

Fig. 1 shows BCC environment with non-contact condition between RX electrode and human body. In

Fig. 1. BCC environment with non-contact status.

JOURNAL OF SEMICONDUCTOR TECHNOLOGY AND SCIENCE, VOL.17, NO.6, DECEMBER, 2017 905

most wearable healthcare system, data from sensor nodes is transmitted to mobile devices to satisfy low power consumption for long operation time and fast data processing for real-time monitoring simultaneously. In the system, sensor nodes are working as a transmitter with well-contact status with human body. However, mobile device cannot maintain well-contact status in most cases because most people take it on their pocket or bag. This non-contact status can cause serious signal attenuation. With non-contact condition between the receiver and human body, signal attenuation between human body and the receiver electrode is about -10~-20 dB with 100 mm distance. This undesirable signal attenuation causes additional power consumption in low noise amplifier (LNA) stage to satisfy sensitivity and linearity requirement simultaneously. Furthermore, the signal degradation can increase data error rate with communication instability.

In this paper, we present first BCC IC design for BCC with non-contact status compensation. The IC is composed of contact status detection unit (CSDU), non-contact status compensation unit (NCU) and reconfigurable LNA. CSDU detects contact status of receiver electrode by measuring coupling capacitance between the receiver and human body with LC impedance resonance, and determines operation mode of NCU. NCU compensates input signal with LC resonance circuit when the system is in non-contact status. The resonance circuit is composed of 3–bit off-chip inductor bank and 6–bit on-chip capacitor bank, and LC configuration controller controls LC configuration with output signal feedback for adaptive signal compensation. Finally, reconfigurable LNA is adopted for adaptive input impedance selection. This paper is extension of [11] with more measurement results and modified circuit implementation.

The rest of this paper is organized as follows. In section II, BCC channel measurement with non-contact status is introduced for circuit specification decision. In section III, overall architecture of the IC is introduced. In section IV, circuit implementation with 3 key building blocks are proposed; CSDU, NSCU and reconfigurable LNA. Then, implementation results are shown in section V, and section VI finally conclude this paper.

II. BCC CHANNEL MEASUREMENT

Fig. 2 shows the block diagram of the proposed digital To determine design specification and verify non-contact signal loss, we measure BCC channel characteristics with non-contact condition. Fig. 2 shows BCC channel measurement environment with non-contact status. A RF generator is adopted as a transmitter for single-tone signal injection, and a portable spectrum analyzer is used as receiver in the measurement. With this configuration, we can make practical BCC environment by separating ground plane of the transmitter and the receiver. Body channel is located in left arm, and the channel length is fixed as 400 mm in the measurement. Distance between the receiver electrode and human body varies within 0~100 mm with 0.1 mm resolution by using vernier calipers. Length of SMA cable in the transmitter and receiver is determined as 5 cm to minimize cable length effect in the measurement [9]. Contact impedance between the receiver and human body is governed by a coupling capacitance [12], so the coupling capacitance value is measured by placing 1.5 μH inductor between the receiver electrode and SMA cable to measure LC resonance frequency.

The measurement result is shown in Fig. 3. Fig. 3(a) shows the coupling capacitance measurement result with 0~100 mm distance between receiver and human body. When the receiver electrode is attached well with human

Fig. 2. BCC channel measurement environment.

906 KYOUNG-ROG LEE et al : A 206.3 Wm NON-CONTACT COMPENSATION IC FOR BODY CHANNEL COMMUNICATION

body (0 mm), the coupling capacitance value is about 100 pF. However, when the receiver electrode is detached from human body, the coupling capacitance value is dropped dramatically to 20 pF, and finally drops to 10 pF at 100 mm distance. This dramatic decrease of coupling capacitance causes drastic signal power degradation in BCC channel environment. Fig. 3(b) shows BCC channel attenuation measurement result with 0~100 mm non-contact distance. As shown in Fig. 3(b), the channel attenuation degrades as distance between the receiver electrode and human body increasing. Finally, -30 dB attenuation is occurred when non-contact distance is 100 mm. The signal degradation can be fatal in BCC environment, because it can cause much power consumption in LNA block to ensure linearity and sensitivity requirement simultaneously.

To compensate the signal attenuation, a new compensation scheme for BCC with non-contact status is proposed. Fig. 4 shows principle of the signal

compensation scheme for non-contact status. As mentioned before, BCC channel environment is mainly governed by coupling capacitance CC between human body and the receiver electrode in non-contact status. Therefore, the signal power can be compensated by placing inductor LC in signal path in series, which makes LC impedance as 0 at resonance frequency f0. LC resonance frequency f0 is expressed in Fig. 4 with LC and CC value. The signal power at LNA input can be easily improved by adjusting f0 to center frequency of communication.

III. OVERALL ARCHITECTURE

Fig. 5 shows the overall architecture of the proposed IC. The proposed IC is composed of 3 key building blocks, which are contact status detection unit (CSDU), non-contact status compensation unit (NCU), and reconfigurable LNA. The input signal from human body passes through the receiver electrode. CSDU measures the coupling capacitance between human body and the receiver electrode by using LC impedance resonance with off-chip LC tank. Then, CSDU generates control signal for NCU to determine operation mode of NCU. When the coupling capacitance is smaller than determined threshold value, which means the system is in non-contact status, compensation path of NCU is activated and signal compensation is occurred. The compensation path is composed of 3-bit series off-chip inductor bank and 6-bit parallel on-chip capacitor bank for adaptive LC impedance resonance. Adaptive LC configuration control is operated by LC configuration

Fig. 3. Non-contact BCC channel measurement results (a) BCC coupling capacitance measurement results, (b) Non-contact BCC channel measurement results.

Fig. 4. Principle of non-contact channel compensation.

JOURNAL OF SEMICONDUCTOR TECHNOLOGY AND SCIENCE, VOL.17, NO.6, DECEMBER, 2017 907

controller with output signal power feedback. Status detection and non-contact signal compensation is occurred with 100 μs period because human motion artifact frequency is under 10 kHz. After the signal compensation is completed, the signal is injected to reconfigurable LNA, which can adjust input impedance of the LNA for adaptive compensation. Output signal of the LNA passes through band pass filter, and go to baseband processer and LC configuration controller for adaptive LC configuration control. The detail circuit implementation about the building blocks will be explained in section IV.

IV. CIRCUIT IMPLEMENTATION

The schematic of CSDU is shown in Fig. 6 [10]. CSDU is composed of 40 MHz oscillator, off-chip LC tank, dividing resistance (RDIV), receiver electrode, envelope detector and comparator. External crystal oscillator generates 40 MHz voltage square wave, and the voltage pulse is divided by Rout and off-chip LC tank impedance. The off-chip LC tank is composed of parallel LC pair LT and CT, which is adjusted to resonate at 40 MHz (220 nH, 70 pF), so the impedance of the tank is almost infinite with 40 MHz square wave. However, the receiver electrode is placed with LC tank with parallel structure, so the coupling capacitance CC by the electrode degrades total LC impedance as (1).

2( )( ) 1

TLC

T T C

sLZ ss L C C

=+ +

(1)

The coupling capacitance CC decreases when the non-contact distance increases, so VRES decreases due to decrease of total LC impedance. An envelope detector averages the waveform of VRES with low-pass filter, and the output voltage VENV is injected to comparator input. The comparator compares VENV with reference voltage (VREF), and generate NSCU control signal. Hysteresis comparator is used for the comparator to prevent fluctuation of CTR signal by fluctuation of VENV. As shown in Fig. 3(a), the coupling capacitance of the receiver electrode drops to 20 pF in non-contact status, so we set threshold value of CSDU as 30 pF to detect contact status of the system. Due to threshold capacitance value, VREF is determined as 350 mV by bandgap reference generator.

Fig. 7 shows simulation result of CSDU. At the beginning of the simulation, the coupling capacitance is 10 pF to assume non-contact status in BCC system. VENV is determined as 450 mV at the condition, which is higher than VREF. Due to the comparator block, CTR signal is in high state, which activates compensation path of NSCU. Contact status transition is occurred in 15 μs,

Fig. 5. Overall block diagram of proposed BCC IC.

Fig. 6. Schematic of contact status detection unit (CSDU).

908 KYOUNG-ROG LEE et al : A 206.3 Wm NON-CONTACT COMPENSATION IC FOR BODY CHANNEL COMMUNICATION

and the coupling capacitance is changed to 100 pF at the moment. VENV decreases due to increase of coupling capacitance, and finally becomes lower than VREF at 45 μs. Total transition time is 30 μs in the simulation. Period of human motion artifact is over 100 μs, so the transition time of CSDU is enough to tracking contact status of the system.

Fig. 8 shows the structure of NCU. NCU is composed of 3-bit off-chip inductor bank (0.22 μH, 1.5 μH, 15 μH), 6-bit on-chip capacitor bank with 0.25 pF unit capacitor and LC configuration controller. Due to hardware cost of off-chip component, coarse signal compensation is operated by log-scale 3-bit inductor bank and fine signal compensation is determined by on-chip 6-bit capacitor bank. As mentioned before, control signal from CSDU (CTR) determines operation mode of NCU. When the system is in well-contact status, the control signal from CSDU activates bypass path of NCU, and signal go through the block without any compensation. On the other hand, when the system is in non-contact status, the compensation path is activated and signal compensation with LC bank is occurred. The inductor is placed in signal path in serial, so it makes LC impedance

resonance with the coupling capacitance. The capacitor bank is placed in parallel with the coupling capacitance, so adaptive impedance control is available with LC bank configuration control. With this LC bank, we can match total impedance of LC pair in the signal path near 0 at communication frequency (40 MHz) due to resonance.

Fig. 9 shows detail selection principle of LC configuration. Fig. 9(a) shows the inductor selection algorithm of NCU. 3 off-chip inductors are placed with switch, and selection algorithm sweeps each inductor by 3-bit one-hot code. The input signal is compensated by selected inductor, and the compensated signal passes through the LNA, band-pass filer and the demodulator, and final output signal power at 40 MHz is saved in the inductor controller. Then, the inductor controller compares each signal power. After the comparison, the controller can select appropriate inductor with maximum output signal power. Fig. 9(b) shows capacitor bank control code selection algorithm. First, the capacitor configuration is selected by the inductor which is selected in previous stage. Each capacitor bank has different unit capacitor size, so effective capacitor switching can be achieved. Furthermore, smaller capacitor bank is reused to larger capacitor bank, so area of the capacitor bank is saved. After the bank selection, capacitor bank control code is selected with SAR algorithm. The control code selected by SAR algorithm is adopted to the capacitor bank, and the controller compares output signal power for each capacitor bank control code. With 3 signal power comparison step, the controller can determine proper control code for the capacitor bank with maximum output signal power at 40 MHz resonance frequency.

Fig. 10(a) shows the schematic of reconfigurable LNA. High input impedance of LNA degrades Q factor of LC resonance, so we choose common gate LNA structure for low input impedance value. Low input impedance improves Q factor of impedance resonance, but it causes higher power consumption of the LNA. So the impedance value should be adjusted with these trade-off condition in BCC environment. The reconfigurable LNA controls current source MOS size and load resistance value simultaneously with 3-bit control code, so we can change input impedance value without change of LNA gain. Fig. 10(b) shows simulation result of the LNA with input impedance and power consumption of the LNA.

Fig. 7. Simulation results of CSDU.

Fig. 8. Schematic of non-contact status compensation unit (NCU).

JOURNAL OF SEMICONDUCTOR TECHNOLOGY AND SCIENCE, VOL.17, NO.6, DECEMBER, 2017 909

With 3-bit control code, input impedance of the LNA varies within 20~125W when the power consumption of the LNA is 0.45~0.06 mW. The condition can be different in various situation, and then control code can be changed with the situation adaptively.

V. SIMULATION RESULTS

Fig. 11 shows layout photograph and summary of proposed IC. The proposed IC is implemented with 65 nm CMOS process with 1.2 V supply voltage. Power

breakdown of the IC is also shown in the figure. Total power consumption of the proposed IC is 206.3 μW. Most power is consumed by LNA block, which is similar as previous BCC transceiver, and additional power consumption for compensation is only 11.3 μW due to CSDU and NCU block.

Fig. 12 shows compensation simulation results of proposed IC. Well-contact status, non-contact status without compensation, and non-contact contact status with proposed compensation scheme is shown in the figure. 1.5 μH inductor is selected for the compensation, and coupling capacitance between the receiver and human body is determined as 15 pF in the simulation. As shown in the figure, 14 dB compensation is occurred with our compensation scheme. As shown in the figure, output signal power with proposed compensation is almost same with well-contact status with non-contact signal compensation.

Fig. 9. Principle of NCU LC configuration selection (a) Inductor selection algorithm, (b) Capacitor selection algorithm.

Fig. 10. Reconfigurable LNA.

Fig. 11. Layout photograph and summary.

Fig. 12. Non-contact compensation simulation result.

910 KYOUNG-ROG LEE et al : A 206.3 Wm NON-CONTACT COMPENSATION IC FOR BODY CHANNEL COMMUNICATION

VI. CONCLUSIONS

In this paper, we propose the first BCC IC with non-contact status compensation. Proposed IC is composed of 3 key building blocks; CSDU with coupling capacitance sensing for system contact status detection, NCU with LC non-contact signal compensation with LC impedance resonance, and reconfigurable LNA with adaptive input impedance control. Thanks to these key building blocks, 14 dB compensation is achieved with 206.3 μW power consumption. Fair comparison with previous work is not available, but only 11.3 μW additional power is consumed for non-contact signal compensation.

ACKNOWLEDGMENTS

This work was supported by Institute for Information and Communications Technology Promotion (IITP) grant funded by the Korea Government (MSIT) (No. 2016-0-00111, Development of Ultra-Low-Power Dual/Wide-Band MICS/BAN Transceiver for Capsule Endoscopy).

REFERENCES

[1] Namjun Cho et al, "A 10.8mW Body Channel Communication/MICS Dual-Band Transceiver for a Unified Body Sensor Network Controller," Solid-State Circuits, IEEE Journal of, Vol.44, No.12, pp.3459-3468, Dec. 2009

[2] Joonsung Bae et al, "A 0.24-nJ/b Wireless Body-Area-Network Transceiver With Scalable Double-FSK Modulation," Solid-State Circuits, IEEE Journal of, Vol.47, No.1, pp.310-322, Jan. 2012

[3] Joonsung Bae et al, "A Low Energy Injection-Locked FSK Transceiver With Frequency-to-Amplitude Conversion for Body Sensor Appli- cations," Solid-State Circuits, IEEE Journal of, Vol.46, No.4, pp.928-937, April 2011

[4] Joonsung Bae et al, "A Low-Energy Crystal-Less Double-FSK Sensor Node Transceiver for Wireless Body-Area Network," Solid-State Circuits, IEEE Journal of, Vol.47, No.11, pp.2678-2692, Nov. 2012

[5] Hyunwoo Cho et al, "A 37.5 μW Body Channel Communication Wake-Up Receiver With Injection-Locking Ring Oscillator for Wireless Body Area Network," Circuits and Systems I: Regular Papers,

IEEE Transactions on, Vol.60, No.5, pp.1200-1208, May 2013

[6] Hyungwoo Lee et al, "A 5.5mW IEEE-802.15.6 wireless body-area-network standard transceiver for multichannel electro-acupuncture application," Solid-State Circuits Conference Digest of Technical Papers (ISSCC), 2013 IEEE International, pp.452-453, 17-21 Feb. 2013

[7] Hyunwoo Cho et al, "A 5.2mW IEEE 802.15.6 HBC standard compatible transceiver with power efficient delay-locked-loop based BPSK demodu- lator," Solid-State Circuits Conference (A-SSCC), 2014 IEEE Asian, pp.297-300, 10-12 Nov. 2014

[8] Joonsung Bae et al, "The Signal Transmission Mechanism on the Surface of Human Body for Body Channel Communication," Microwave Theory and Techniques, IEEE Transactions on, Vol.60, No.3, pp.582-593, March 2012

[9] Hyunwoo Cho et al, "A 79pJ/b 80Mb/s full-duplex transceiver and a 42.5μW 100kb/s super-regenera- tive transceiver for body channel communication," Solid- State Circuits Conference - (ISSCC), 2015 IEEE International, pp.380-381, 22-26 Feb. 2015

[10] Kiseok Song et al, "A 20 µW contact impedance sensor for wireless body-area-network transceiver," Custom Integrated Circuits Conference (CICC), 2011 IEEE, pp.1-4, 19-21 Sept. 2011

[11] Kyoung-Rog Lee et al, " A 635 μW Non-contact Compensation IC for Body Channel Communi- cation," International Symposium on Circuits and Systems (ISCAS), 2016 IEEE, pp.1406-1409, 22-25 May 2016

[12] Neuman M. R. et al; Biopotential Electrodes, In: Medical Instrumentation – Application and Design, 3rd ed., pp. 182-232, John Wiley & Sons, 1998

Kyoung-Rog Lee (S’15) received the B.S. degree in electrical engi- neering from Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, in 2015, and the M.S. degree in electrical engineering in 2017, where he is

currently working toward the Ph.D. degree. His current research interests include low-power bio-medical SoC for wearable healthcare system. He is also interested in body channel communication (BCC) SoC for low-power communication system.

JOURNAL OF SEMICONDUCTOR TECHNOLOGY AND SCIENCE, VOL.17, NO.6, DECEMBER, 2017 911

Jaeeun Jang (S’13) received the B.S. degree in electrical engineering from the Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, in 2014, and the M.S. degree in electrical engineering in 2016, where he is currently

working toward the Ph.D. degree. He has worked on developing a low-power wireless transceiver and low-power sensor front-end. His current research interest includes low power transceiver design for body-area-networks and low-power biomedical system design.

Hoi-Jun Yoo (M’95 – SM’04 – F’08) graduated from the Electronic Department of Seoul National Univer- sity, Seoul, Korea, in 1983 and received the M.S. and Ph.D. degrees in electrical engineering from the Korea Advanced Institute of Science

and Technology (KAIST), Daejeon, in 1985 and 1988, respectively. Since 1998, he has been 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 Korean System Integration and IP Authoring Research Center (SIPAC). From 2003 to 2005, he was the full time Advisor to Minister of 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) at KAIST. Since 2010, he has served the general chair of Korean Institute of Next Generation Computing. His current interests are computer vision SoC, body area networks, biomedical devices and circuits. He is a co-author of DRAM Design (Korea: Hongrung, 1996), High Performance DRAM (Korea: Sigma, 1999), Future Memory: FRAM (Korea: Sigma, 2000), Networks on Chips (Morgan Kaufmann, 2006), Low-Power NoC for High-Performance SoC Design (CRC Press, 2008), Circuits at the Nanoscale (CRC Press, 2009), Embedded Memories for Nano-Scale VLSIs (Springer, 2009), Mobile 3D Graphics SoC from Algorithm to Chip (Wiley, 2010), Bio-Medical CMOS ICs (Springer, 2011), Embedded Systems (Wiley, 2012), and Ultra-Low-Power Short-Range Radios (Springer, 2015).