See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/351742950

An Adaptive Analog Front End for a Flexible PPG Sensor Patch with Self-
Determined Motion Related DC Drift Removal

Conference Paper · May 2021

DOI: 10.1109/ISCAS51556.2021.9401265

CITATIONS READS

6 114

2 authors:

Rajeev Kumar Pandey Paul C.-P. Chao

MediaTek National Chiao Tung University
31 PUBLICATIONS 136 CITATIONS 469 PUBLICATIONS 6,023 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Rajeev Kumar Pandey on 20 May 2023.

The user has requested enhancement of the downloaded file.

 An Adaptive Analog Front End for a Flexible PPG
Sensor Patch with Self-Determined Motion Related
DC Drift Removal
Rajeev Kumar Pandey and Paul C.-P Chao IEEE Sr., Member, IEEE
EECS International Graduate Program, National Yan Ming Chiao Tung University, Hsinchu City, 300, Taiwan
[email protected]
2021 IEEE International Symposium on Circuits and Systems (ISCAS) | 978-1-7281-9201-7/20/$31.00 ©2021 IEEE | DOI: 10.1109/ISCAS51556.2021.9401265

Abstract—This paper presents, a DC drift compensated (due the signal to noise ratio (SNR). In addition, quality of the PPG
to motion <1Hz), high sensing precision and wide dynamic range signal is also a biggest concern since the LEDs and PDs do not
analog front end (AFE) for the organic type PPG sensor. The attached well to the skin, thence, SNR reduce [4][5][6].
proposed AFE utilizes an auto-programmable trans-impedance Furthermore, an adaptive filtering with hybrid (on the cost of
amplifier (TIA) with the DC-current cancellation loop and a additional sensor for reference) or multi-channel PPG
band pass filter to compensate the DC drift either due to motion acquisition are a popular technique for addressing motion
or ambient light change. The prototype of the readout circuit is artifact rejection (MAR) [12] [13]. In this case an external
implemented in an integrated chip with an area of 3mm2 via the adaptive LMS filter based control system is required to
TSMC 180nm process. The experiment result shows that the
process the receive PPG signal to control the DC drift due to
designed readout can sense linearly from 1nA to maximum
motion. Point to be noted that the DC (non-pulsatile
40µA current, with a resultant dynamic range of 90dB and the
component) is an important feature for accurate measurement
measured input-referred noise is 0.256 nA/ √ Hz. The total
of the SpO2 and blood flow volume [18] [19]. Therefore, an
measured power consumption of the readout circuit is 460µW,
while the power consumption of the on-chip OLED driver is 1.8 adaptive analog front-end (AFE) circuit with high dynamic
mW. The obtained PPG signals from the wrist are subsequently range (DR), high sensing precision to low current, high signal
processed with quality checking, and feature extraction. The to noise ratio (SNR) [4] [5] and low DC-drift is required to
measured accuracy and standard error for the heart rate ensure the long-time continuous monitoring of high-quality
estimation are 96%, and -0.12±5 beats/minute, respectively. PPG signal. This study intends to overcome the
aforementioned challenges like ambient light artifact, SNR
Keywords— Photoplethysmography (PPG) Sensor, filter, degradation, mispositioning, motion artifacts (<1Hz) without
flexible patch, OLED, OPD, Motion artifact rejection (MAR) using any additional sensor on the chip. Therefore, this study
is organized as follows: Section II present the characterization
I. INTRODUCTION and specification of the flexible OLED-OPD sensor. Section
According to the report published by WHO, 9 million III presents the design of the PPG Signal acquisition system
people died in 2019 due to heart disease which represents 19% for the heart rate estimation. Section IV presents the
of the global death. It is well known that the people with heart experimental results and finally section V concludes this
disease and respiratory conditions are more prone to the study.
COVID 19. Therefore, the regular measurement of the vital
bio signs like blood oxygen (SpO2), body temperature, Heart II. CHARATERIZATION OF THE FLRXIBLE PATCH SENSOR
rate (HR) and blood pressure (BP) becomes one of the The structure of the designed circular patch is shown in
common practices. To enable individuals interested in Fig.1, which incorporates an organic red (λ=630nm) OLED,
tracking their body’s key health indicators, wearable devices an organic green (λ=530nm) OLED, and an organic
are one of the main choices. Photoplethysmography (PPG) photodetector (OPD). The designed flexible sensor patch can
sensors is well known technology to measure all the above attach well to the skin therefore there is almost no relative
mention bio signs (except temperature). Recently, many motion between the sensor and skin. The circular geometry of
literatures dedicated there effort to design and develop a the patch helps to minimize the misposition issue along with
highly efficient and robust on-chip PPG devices [1]-[16]. An the direction of the blood vessel. Optical simulation has been
adjustable sampling rate and adaptive intensity are one of the performed to optimize the distance and aperture area of the
prominent techniques to achieve low power and large SNR sensor [17]. Thus, the optimized pulsatile AC to non-pulsatile
[1]. Similarly, another study proposed a 1V, 8.1 µA with DC (AC/DC) ratio during the optical simulation and
>92db dynamic range (DR) readout with off chip timing experiment are 8% and 7%, respectively.
control [2]. In reference [3], a low power and reconfigurable 9.4 mm
DR, PPG sensing system is designed by using the variable 1 mm 1 mm
pulse repetition frequency (PRF) [4]. Su et al. 2020 proposed
a direct current to digital conversion readout circuit. In this a 7.4 mm
digital TIA with oversampled SAR ADC is used as a core.
0.5mm

R3=
R1=

4.2 mm

Although as compare to the analog programmable TIA, digital 1 mm 2.2mm 1 mm

6.2 mm

TIA have the similar noise issue plus the ADC noise [11]. It
is well known that the measured pulsatile AC to non-pulsatile
DC ratio of the LED-PD PPG sensor is less than 10%. Thence,
a modest motion or the slight change in ambient light induces 1 mm 1 mm
a large shift change in the DC component (<0.1Hz), in such a
way that the detection of the pulsatile AC becomes difficult. Fig. 1. The optimizing constant AC-to-DC ratio over varying the OLED
intensity (by the means of reflective path distance) [19].
On the other hand , the high frequency noise (>10Hz) reduces

978-1-7281-9201-7/21/$31.00 ©2021 IEEE

Authorized licensed use limited to: IEEE Xplore. Downloaded on May 14,2021 at 00:03:05 UTC from IEEE Xplore. Restrictions apply.
 Analog Front End readout with integrated OLED Driver and Control
Programmable Bandpass TIA
Time interleaved Difference amplifier Low Dropout C1
Band-gap reference
2x2 OLED Driver Tunable Pseudo Regulator Programmable

1K
A Cp1 C2
Resistance gain amplifier
Vm C3
f
OPA2 Cf
REF Inverting Amplifier Bandpass Filter (0.3Hz -10Hz)
R3
R1 C8
R2
PWM Feedback Loop 1 R2 R3 OPA1 Vi+
VDD Cf R1
REF REF CiB REF

Feedback Loop 2
OPD C1 OPA1 OPA1

PGA_Out
TIA OUT
Same MCU as on Right side

OPA1
Rf1 C1

LPF_out
MP1 Rf2 C2 C2 3:8 Decoder
Rf3
MCU REF REF
Rf4 REF

PGA2

PGA0
PGA1
4x1
PD Pseudo Resistance 2nd Order High Pass 2nd Order Low Pass
120s Array
Control
TIA Out
OPA1
Select
Analog
Cs
DEMUX MCU with BLE
REF Switch Control Signal
(Comparator, U/D Counter and
Analog Multiplexer) Digital Control
Programmable
Control Signal
from MCU Current mode Current DAC Secondary Ambient
DAC 8 bit resolution Cancellation

Fig. 2. Analog front-end readout circuit (right side) and OLED driver (left side).

The experimental results show that the OPD current is where GBW is the gain bandwidth product. Thus, the minimal
varied in the range of 100nA to the 11µA (including dark value of the %= is determined by the equation (3). On the other
current). The measured range of the OLED forward drive hand, the programmable gain TIA helps to enhance the
current is 0.01-1.6mA [19]. dynamic range. TIA can boost the output voltage under low
III. DESIGN OF PPG ACQUISITION SYSTEM photocurrent from the OPD. Also, it helps to retain the output
in the linear region and minimize the saturation of the TIA
The simplified circuit diagram of the PPG acquisition under high photocurrent from the OPD. The auto-
system is shown in Fig. 2. The designed circuit incorporates a programming of the TIA is achieved by connecting the TIA to
TIA (100dbΩ) with continuous DC cancellation loop, a the input of the two comparators. The upper and lower bound
second order bandpass filter (0.1-10Hz), analog multiplexer,
of the comparators is set at 1.6V and 0.1V respectively. When
programmable gain amplifier (PGA), digital control, a
programmable DAC-PWM based auto intensity tuned OLED the input to comparator crosses the bounded threshold then the
driver low dropout regulator and band gap reference circuit. UP/Down counter changes the feedback resistance to the
required value within 4 counts.
A. Programmable Transimpedance Amplifier
B. Second order Bandpass Pass Filter and Programmable
The OPD is biased in the negative bias-photoconductive Gain amplfier
mode. The programmable resistive transimpedance amplifier
(TIA) convert the OPD current into an equivalent voltage. As A band pass filter, and a PGA are designed to enhance the
shown in Fig. 1, the feedback loop 1 consist of a feedback MAR, SNR, and DR further. The overall 3-dB cutoff of the
capacitor (0.1uF external capacitor) with four on-chip band pass filter is 0.1Hz-10Hz. A second-order high-pass
resistance (220K, 110K, 55K, 28K). Similarly, the negative filter circuit is used herein to reject the noise < 0.3 Hz (slow
feedback loop 2 inside TIA show’s an inherent high pass DC drift). Low pass filter helps to alleviate the effect of the
filtering with a cutoff of 0.3Hz. The overall response of the high frequency noise >6.8Hz. Tunable pseudo resistances
design TIA is given by have been used to achieved the GΩ resistance [15] [16]. In
addition, a PGA (1x to 8x) is used herein in at the output of
A s ≈R .
the bandpass filter to acquire the dynamic range of the inbuild
, "#
(1)
! ADC of the MCU . MCU controlled additional 8-bit resolution
current-DAC is designed in the feedback to ensure the ambient
According to the equation (1), the pass band of the design light rejection.
TIA is 0.3-10 Hz. The advantage of the designed negative
C. OLED array Driver Circuitry and Control Mechanism
feedback loop 2 is that it acts as a continuous DC
cancellation. Therefore, the design loop automatically A 3-bit control DAC is designed herein to limit the
cancelled out any uncertain shift in the DC either due to maximum current of the OLED driver. The maximum current
motion or the ambient light. Also, the feedback capacitance limit for the design OLED is 1.2mA. The designed OLED
%& ) helps to limit the low pass bandwidth of the TIA to the driver circuit incorporate a 1 bit-current DAC, a control
10Hz. The total input-referred noise power spectral density resistance (Rd), and a 4:1 multiplexer. The intensity of the
(PSD) of the TIA is given by equation OLED is tuned by using the pulse width modulated (PWM)
signal from the MCU. The required modulating voltage (LM )
*'***) = 2-' + 123 + *7******) 9 :
+ 4<= %& + %>8? @ (2) is given by,
( ./ (.8 456 456
LM = 'NO? ∗ QR = LST ∗ UVWX %XYZ7 % ,
where, 7*******
(.8 is the input referred noise and %>8? is the
)
(4)
internal capacitance of the OPD. On the other hand, the where QR = 300 Ω and LST is 1.8V. The tunable range of the
stability of the system is ensured by the equation given below current ('NO? ) for the design OLED driver circuit is 100uA to
D
8 mA.
%& > C EFG (3)
HIJ.K.4 5

Authorized licensed use limited to: IEEE Xplore. Downloaded on May 14,2021 at 00:03:05 UTC from IEEE Xplore. Restrictions apply.
D. Digital control unit, Quality assessment Algorithm and harmonics of the single sided spectrum (Fig 3(c)) is used to
Heart Rate measurement calculate the HR in frequency domain. Therefore, frequency
The process flow of the digital control and quality domain and time domain HR are related by the equation given
assessment unit is shown in Fig. 3, which comprised a moving below
:
zQ = 60 ∗ |zQ = 60 ∗ } ~.
average filtering, peak-through detection, periodicity check,
(7)
amplitude check, power spectrum (PS) check, OLED array 843
control and PGA gain control [18]. To implement this, a 4:1 IV. EXPERIMENT RESULTS
analog multiplexer is designed in the AFE and the 2-bit select
signal is used to turn ON the OLED one-by-one for 3 seconds. The readout circuit is designed and implemented by using
At the same time PPG signal due to each OLED is acquired T18 process. The micrograph and the test environment of the
readout chip is shown in Fig. 4. For the experiment, the
from the single channel readout. The OLED which satisfy the
designed readout is attached to the OLED-OPD sensor patch.
quality check criteria (periodicity, amplitude and PS) will be
Output of the readout circuit at different TIA gain is shown
used for the long-time continuous monitoring of the PPG Fig. 5. The output of the DAC-PWM based intensity control
signal. If it’s failed, it will change the PGA gain/intensity and OLED driver at varying duty cycle is also included in Fig. 5.
redo the procedure. Point to be noted that the periodicity The measured maximum TIA gain is >100dBΩ @220K. The
check is performed on the three-consecutive cycle and the received PPG signal and its FFT is shown in Fig. 6. According
periodicity check criteria is given by to the FFT of the receives PPG signal, the first harmonics
3
0.9 ≤ ≤ 1.1,
reflects the incident wave, second harmonics shows the first
(5)
3 reflected wave, similarly the third harmonics reflects the third
where, `: is the time between first and second peak and `) is
reflected wave. Thus, the entire band is 0.9Hz-4Hz. Therefore,
measured SNR at the 3rd harmonics is 74dB. Further, Fig. 7
the time between second and third peak. After this the average shows the SNR of the design AFE at the varying input current.
of peak voltage and difference of peak-through (VP-P) is The measured DR is 90dB. The minimum current that the
measure. Then the frequency spectrum of the PPG signal as circuit can sense is 1nA. The measured input referred noise is
shown in Fig. 7 is used for the PS (%) check (PS>30%) and it 0.26 nA/√Hz. The total power consumption of the readout
is given by given by circuit is 460uW and the power consumption of the integrated
cde(fghRi j& :kg ldmMj(fn oinj(R pdmMdj(fn
ab =
OLED driver (100% duty cycle, 1mA drive current) is 1.8
(6)
q∑6vw
6vx )|tt3FFu ( |y
mW. The comparison of the output PPG signal from the
where T is the number of the samples.
existing AFE [5] with the designed chip is shown in Fig. 8.
The measurement is taken from both the device at same wrist
Reading the raw data
within 3 seconds
position from the same subject with palm movement and
finger movement. Result shows that the DC drift due to
Moving average filter

Find peak and trough Start

motion and ambient artifact has been reduce with the new
Period
Check
AFE. The experimental result shows that the design readout
Whether
periodic No
Ignore this
data
Input PPG Yes NO
system has the fast DC settling time < 1 second. The overall
Save the Pair
Yes
Convolution of Peak and
Remove
Display specification of the design readout and its comparison with
Filter
Calculate VP-P and
Trough on APP
recent state-of-arts are shown in Table I. The non-invasive
PS (%) Peak
Detection PRT Calculate
PPG sensor integrated with sensor patch is used to sense the
Next 3 seconds Take average and

Acceptable
Move to next Trough
Calculation HRV sent to Display PPG signal from the wrist artery of the 35 subjects in real time.
OLED
NO Detection
PRT to
Comparisons
The Bland -Altman plot of the HR measurement is shown in
HR
Yes
Assign Best quality PPG Signal form
Highest & if the
difference is
Fig. 9. The obtained accuracy and the mean ± 1.96 x standard
Single sided FFT Peak HR
OLED to PPG Channel Position detection
less than 10 deviation error for the HR measurement in real time are 96%,
and -0.012 ± 5 bpm, respectively.
(a) (b) Output PPG

OLED Readout Power Supply

Driver
OLED-OPD Sensor

Heart Rate=60*FHR ≈60*1/PRT

DAC
*Peak

PRT Control OLED-OPD Test Chip

Sensor at
wrist

*Through
Fig. 4. Micrograph of the chip and the test environment.
FHR
(c) TIA @27.5K TIA@220K 25% Duty Cycle
25% Duty Cycle PPG SNR 80 dB
Fig. 3. (a) Time interleave strategy and primary quality detection algorithm. PPG SNR 37dB

(b) Flow diagram of Heart rate estimation algorithm [18] (c) Typical
PPG signal and its FFT spectrum.
TIA @27.5K 75% Duty Cycle
After the quality check is finished, the heart rate estimation TIA @27.5K PPG SNR 90 dB
50% Duty Cycle
will be performed. The algorithm for the heart rate estimation
is shown in 3 (b). The raw PPG signal will be filtered by the
convolution filter [18], then the peak-peak interval/pulse
repetition time (PRT) is measured to estimate the HR in the
time domain. Also, the frequency (FHR) of the first Fig.5. Output of the readout circuit at different combination of the TIA gain
and OLED intensity.

Authorized licensed use limited to: IEEE Xplore. Downloaded on May 14,2021 at 00:03:05 UTC from IEEE Xplore. Restrictions apply.
 TABLE I. SPECIFICATION OF THE DESIGN READOUT CIRCUIT
Parameters This work [3] [2] [11] [12]
Technology 0.18µ 0.18 µ 0.18µ 55n 0.18 µ

Supply voltage 1.3-2 1.2/3.3 1/2.5 1.8/.8 5/1.5

2nd Harmonics (V)
(1st Reflected Wave)
Architecture PTIA+Filter+ LDC LDC TIA+A TIA+
MAR DC BIoZ +
37dB

MAR
Readout For OLED-OPD LED-PD LED- LED- Organi
74dB

(OPD max @ PD+∆∑@ PD c

3rd Harmonics
current 11 Finger finger @finge @wrist
(2nd reflected Wave) µA) @wrist r
SNR of PPG 37db@100nA NA NA 80db@ 30dB
1st Harmonics -80db@6 µA 6 µA
(incident wave)

0.1-5Hz Sensing Range 1nA->100 µA ≈6- NA 1nA- NA

40µA 200 µA
Input Noise 0.26 48.6 NA 0.7µVr 0.57
Fig. 6. (a) Received PPG Signal. (b) its Frequency spectrum (@50%duty cycle nA/(Hz)0.5 ms
and TIA at 27.5k-ohm. Dynamic Range >90dB 89dB 92.7dB 130dB NA
@10
µA
Dc Rejection 10µ 50uA 10µ NA 4.8 µ
loop
Power (W) 460 µ@100% 55.7uw 8µ@10% 72 µ NA
readout duty cycle of duty
(Includes cycle
biasing and (does not
LDO) include
bias and
others)
OLED drive 0.01-8mA NA 0.1/103.2 NA 0.06-
Current min/max (8-bit mA 0.96
resolution) mA
Area (mm2) 3 (readout + 7 4.8 4.5 8.25
OLED driver)
Fig. 7. SNR vs input current AFE.
V. CONCLUSION
The designed single channel AFE shows that the DC drift
due to motion artifact has been significantly reduce. The
achieved sensing precision is 1nA. The measured dynamic
range is 90dB and the measured input referred noise is 0.256
nA/√Hz. Auto programmable TIA with DC cancellation loop
helps to minimize the impact of the motion artifact (<1Hz).
Adaptive AFE with OLED-OPD sensing patch shows that the
(a) impact of the noise, mispositioning and motion artifact is
significantly reduced. The experimental result shows that the
design readout system has the fast DC settling time (<1
second) after the occurrence of the motion artifact. The total
measured power consumption of the readout circuit is 460µW,
while the power consumption of the integrated OLED driver
(100% duty cycle, 1mA drive current) is 1.8 mW. The
obtained accuracy and the mean ± 1.96 x standard deviation
error for the HR measurement in real time are 96%, and -0.012
(b) ± 5 bpm, respectively.
Fig. 8. (a) AFE without MAR (DC drift cancellation) (b) output of the AFE
with DC drift Cancellation during the palm and finger movement. ACKNOWLEDGMENT
Author would like to thank Taiwan semiconductor
research institute (TSRI) and TSMC for providing chip
implementation software resources and chip implementation
shuttle. This is study is supported by Ministry of Science and
Technology, Taiwan grant No. MOST 108-2823-8-009 -002 -
,109-2622-8-009-018 -TE1,109-2221-E-009-163 – and 109-
2622-E-009 -027-. This work was financially supported by the
" Center for Intelligent Drug Systems and Smart Bio-devices
(IDS2B) " from The Featured Areas Research Center Program
within the framework of the Higher Education Sprout Project
by the Ministry of Education (MOE) in Taiwan. It was also
supported in part by Hsinchu Science Park Bureau, MOST
grant No. 108A31B.
Fig. 9. The Standard error plot for the HR measurement.

Authorized licensed use limited to: IEEE Xplore. Downloaded on May 14,2021 at 00:03:05 UTC from IEEE Xplore. Restrictions apply.
 REFERENCES Incremental ADC with 12b SAR Quantizer and Reset-Then-Open
Resistive DAC Achieving 140dB DR and 8ppm INL at 4kS/s," 2020
[1] A. Sharma et al., "A Sub-60- $\mu \text{A}$ Multimodal Smart IEEE International Solid- State Circuits Conference - (ISSCC), San
Biosensing SoC with >80-dB SNR, 35- $\mu \text{A}$ Francisco, CA, USA, 2020, pp. 154-156, doi:
Photoplethysmography Signal Chain," in IEEE Journal of Solid-State 10.1109/ISSCC19947.2020.9062990.
Circuits, vol. 52, no. 4, pp. 1021-1033, April 2017, doi:
10.1109/JSSC.2016.2642205. [11] Shu, Yun-Shiang et al. “26.1 A 4.5mm2 Multimodal Biosensing SoC
for PPG, ECG, BIOZ and GSR Acquisition in Consumer Wearable
[2] Marefat, Fatemeh et al. “A 1-V 8.1uW PPG-Recording Front-End With Devices.” 2020 IEEE International Solid- State Circuits Conference -
> 92-dB DR Using Light-to-Digital Conversion With Signal-Aware (ISSCC) (2020): 400-402.
DC Subtraction and Ambient Light Removal.” IEEE Solid-State
Circuits Letters 3 (2020): 17-20. [12] Y. Lee et al., "Sticker-Type Hybrid Photoplethysmogram Monitoring
System Integrating CMOS IC With Organic Optical Sensors," in IEEE
[3] Lin, Binghui et al. “14.85 µW Analog Front-End for Journal on Emerging and Selected Topics in Circuits and Systems, vol.
Photoplethysmography Acquisition with 142-dBΩ Gain and 64.2- 7, no. 1, pp. 50-59, March 2017, doi:
pArms Noise.” Sensors (Basel, Switzerland) vol. 19,3 512. 26 Jan. 10.1109/JETCAS.2016.2630301.1109/TBCAS.2019.2945114.
2019, doi:10.3390/s19030512
[13] Lee J, Kim M, Park HK, Kim IY, “Motion Artifact Reduction in
[4] J. Lee, D. Jang, S. Park and S. Cho, "A Low-Power Wearable Photoplethysmography Based on Multi-Channel Sensors
Photoplethysmogram-Based Heart Rate Sensor Using Heartbeat with Multiple Wavelengths”, Sensors (Basel). 2020; 20(5):1493.
Locked Loop," in IEEE Transactions on Biomedical Circuits and Published 2020 Mar 9. doi:10.3390/s20051493.
Systems, vol. 12, no. 6, pp. 1220-1229, Dec. 2018, doi:
10.1109/TBCAS.2018.2876671. [14] Orozco, Luis Munguía. “Programmable-Gain Transimpedance
Amplifiers Maximize Dynamic Range in Spectroscopy Systems.”
[5] R. K. Pandey, E. F. Pribadi and P. C. -. Chao, "A New Adaptive (2013).
Readout System For a New OLED OPD Flexible Patch PPG Sensor,"
2019 IEEE SENSORS, Montreal, QC, Canada, 2019, pp. 1-4, doi: [15] E. Guglielmi et al., "High-Value Tunable Pseudo-Resistors Design," in
10.1109/SENSORS43011.2019.8956825. IEEE Journal of Solid-State Circuits, doi:
10.1109/JSSC.2020.2973639.
[6] Y. Khan, D. Han, J. Ting, M. Ahmed, R. Nagisetty and A. C. Arias,
"Organic Multi-Channel Optoelectronic Sensors for Wearable Health [16] M. -. Shiue, K. -. Yao and C. -. A. Gong, "Tunable high resistance
Monitoring," in IEEE Access, vol. 7, pp. 128114-128124, 2019, doi: voltage-controlled pseudo-resistor with wide input voltage swing
10.1109/ACCESS.2019.2939798. capability," in Electronics Letters, vol. 47, no. 6, pp. 377-378, 17
March 2011, doi: 10.1049/el.2010.3286.
[7] Lin, Qiuyang et al. “A 196μW, Reconfigurable Light-to-Digital
Converter with 119dB Dynamic Range, for Wearable PPG/NIRS [17] Pribadi, E.F., Pandey, R.K. & Chao, P.CP. Optimizing a novel PPG
Sensors.” 2019 Symposium on VLSI Circuits (2019): C58-C59. sensor patch via optical simulations towards accurate heart rates.
Microsyst Technol 26, 3409–3420 (2020).
[8] A. K. Y. Wong, K. Pun, Y. Zhang and K. N. Leung, "A Low-Power https://doi.org/10.1007/s00542-020-04895-6
CMOS Front-End for Photoplethysmographic Signal Acquisition With
Robust DC Photocurrent Rejection," in IEEE Transactions on [18] Pandey, R.K., Chao, P.CP. External temperature sensor assisted a new
Biomedical Circuits and Systems, vol. 2, no. 4, pp. 280-288, Dec. 2008, low power photoplethysmography readout system for accurate
doi: 10.1109/TBCAS.2008.2003429. measurement of the bio-signs. Microsyst Technol (2020).
https://doi.org/10.1007/s00542-020-05106-y
[9] Khoman Phang and D. A. Johns, "A CMOS optical preamplifier for
wireless infrared communications," in IEEE Transactions on Circuits [19] Pandey, R.K., Lin, TY. & Chao, P.CP. Design and implementation of
and Systems II: Analog and Digital Signal Processing, vol. 46, no. 7, a photoplethysmography acquisition system with an optimized
pp. 852-859, July 1999, doi: 10.1109/82.775380. artificial neural network for accurate blood pressure measurement.
Microsyst Technol (2021). https://doi.org/10.1007/s00542-020-05109-
[10] S. Wu, Y. Shu, A. Y. Chiou, W. Huang, Z. Chen and H. Hsieh, "9.1 A
Current-Sensing Front-End Realized by A Continuous-Time

View publication stats Authorized licensed use limited to: IEEE Xplore. Downloaded on May 14,2021 at 00:03:05 UTC from IEEE Xplore. Restrictions apply.