Journal of Integrated Circuits and Systems, vol. 19, n.

02, 2024 1

A High Gain, High CMRR, Low Noise Bio-Potential amplifier based on Switched
Capacitor Feedback Amplifier
Bhaskara Rao Kasipogula1 , and Gurumurthy Komanapalli2
1
Research Scholar,School of Electronics Engineering, VIT-AP University, India.
2
Assistant Professor, School of Electronics Engineering, VIT-AP University, India.
e-mail: gurumurthy.k@vitap.ac.in

Abstract— This paper describes a biopotential amplifier millivolts (mv), which allows them to be sensitive to overall
based on an operational transconductance amplifier (OTA) de- neural activity on a wide scale. However, it is not as sensitive
signed specifically for bio-potential recordings to achieve high to the firing of individual neurons.
gain and low noise. The amplifier uses a differential architec- Intracellular recording refers to the technique of measur-
ture with chopper modulators and pseudo-resistors to achieve ing electrical activity within a cell. Methods such as patch
a Common Mode Rejection Ratio (CMRR) of over 100 dB. clamping enable the direct quantification of electrical activ-
The design has several key features, including precise and high ity in individual neurons. These techniques provide a high
steady gain, large output swing, good load-driving capabilities, level of sensitivity and are capable of detecting even little
high input impedance (Zin), and a small footprint. The paper alterations in membrane potential, measuring in the range
compares the proposed circuit with other bio-potential ampli- of mV or even µV . Extracellular electrodes can simultane-
fiers in the literature, focusing on key parameters such as gain, ously detect the activity of several neurons. The sensitivity of
input referred noise (IRN), CMRR, and Zin. The outcomes the electrode design and placement determines the variation,
show that the suggested amplifier has a gain of 57.7 dB, an however, the usual signals range from µV to mV. There is
IRN of 0.839 µVrms, a CMRR of 108.13 dB, and consumes background noise that contains all possible frequency ranges
8.854 µW at 0.8 V. The proposed bio-potential amplifier was and has greater amplitudes than the actual signal of inter-
implemented using a 180 nm CMOS technology node and was est. Since the neuron’s detection sensitivity has increased,
verified through simulations using the Cadence Virtuoso Spec- the weak periodic signal particles should be concentrated
tre Simulator. The findings indicate that the proposed design in a single state determined by the starting conditions. The
is highly effective for recording bio-potentials in various appli- stochastic resonance (SR) model has the ability to preserve
cations. and boost the amplitude of weak signals [6, 7].
Index Terms— Enhanced OTA; Low noise chopper modula-
In measuring the neural signals in various portable bio-
tor; Pseudo-Resistor; High gain; High CMRR; Input referred
medical monitoring devices, the impact of interfering signals
noise(IRN).
such as electrode offset, flicker noise (1/f), and powerline in-
terference (PLI) is high. Therefore, to provide good perfor-
I. I NTRODUCTION mance, the amplifier gain and CMRR should be increased,
A signal is a physical phenomenon that carries informa- and the power utilization of the device must be minimized.
tion. Information about a biological system under study may Minimizing power consumption is indeed crucial, especially
be obtained through the use of bio-electrical signals [1]. in applications like bio-potential amplifier systems where en-
The bio-electric signals generated by the human body will ergy efficiency is paramount. A milliwatt power consump-
be in the range of a few microvolts to millivolts. The classi- tion level could potentially be acceptable, depending on the
fication of neural signals based on frequency amplitude and specific requirements and constraints of the system.
measuring methods is shown in Table I. The capture of bio- While a milliwatt power consumption level might pose
signals has become a critical component in advanced med- some challenges, it’s not necessarily a compromise if the
ical applications and wearable monitoring systems such as amplifier is designed with efficiency and signal integrity in
Electroencephalograph (EEG), ElectroCardiogram (ECG), mind. Close attention to design parameters and trade-offs
Electromyogram (EMG) signals, functional electrical stim- can ensure that the bio-potential amplifier system meets its
ulation (FES), etc., [2, 3]. The task of measuring neural sig- performance requirements while minimizing power utiliza-
nals such as ECG, EEG, and EMG in portable biomedical tion.
monitoring devices against a variety of interfering signals is In this analysis, we performed a focused review by selec-
enormous. The goal is to maintain performance while mini- tively examining a limited number of key prior studies.
mizing power consumption. [4, 5]. Carolina et al. introduced a preamplifier as the initial stage
The ability to identify neuron signals is contingent upon of an Analog Front End (AFE) intended for biological signal
several aspects, such as the measurement technique, the cal- acquisition [8]. The AFE utilizes depletion threshold varia-
iber of equipment, and the distinct attributes of the neurons tions to achieve an impressive IRN of 1.3 µVrms , along with
under investigation. Below are many frequently used tech- a bandwidth of 20 Hz to 11 kHz and a high CMRR exceeding
niques along with their normal levels of sensitivity. The var- 70 dB.
ious neural signals and recording methods have been able Prateek et al. presented a circuit comprising two sections:
to detect signals that are measured in microvolts (µV ) and the core circuit featuring a differential amplifier, and the sup-

Digital Object Identifier 10.29292/jics.v19i2.813

2 Bhaskara et al.: High Gain and CMRR Bio-Potential Amplifier Design Using SCFA
Table I. A summary of various neural signals and recording methods.
Frequency Dynamic
Classification Acquisition Comments
Range (Hz) Range (V)
Electroencephalogram(EEG) Surface Electrodes 0.5-100 2-100 µ Multi-Channel (6-32) Scalp Potential
Electrocardiogram (ECG) Surface Electrodes 0.05–100 1–10 m Cardiac Action Potential to the surface
Electro-oculogram (EOG) Surface electrodes dc–100 10 µ–5 m Steady-corneal-retinal potential
Electroneurogram (ENG) Needle electrode 100 –1 k 5 µ–10 m Potential of a nerve bundle
Action potential (AP) Micro electrodes 100 –2 k 10 µ–100 m Invasive measurement of cell membrane potential
Electroretinogram (ERG) Micro electrode 0.2–200 0.5 µ–1 m Evoked flash potential
Evoked potentials (EP) Surface electrodes 0.2 – 5 k 0.1–20 µ Response of brain potential to stimulus
Local Field Potential (LFP) Micro electrodes 0.2–200 <5m Extracellular space in brain tissue
Intramuscular and
Electromyography (EMG) 0.01 to 10 k 0-10 m Neuromuscular activities
surface electrodes

ply control circuit, which governs the circuit’s switching be- −the refs [14, 15, 10, 12, 16, 13, 9, 8, 18, 21, 23, 24] shows
havior [9]. By implementing double-stacked MOS at the that high in IRN.
pull-up and pull-down portions, there is a notable decrease This paper presents a method for developing a bio-
in the dissipation of leakage power. This circuit’s power dis- potential amplifier that is optimal for bio-potential record-
sipation is only 1.65 µW while offering a CMRR of 65 dB ings, as compared to other papers in this field. The rec-
and a gain of 12 dB. ommended amplifier meets essential requirements for bio-
Erwin et al. introduced a versatile, low-noise chopper am- potential recordings by incorporating key features such as
plifier [10]. The amplifier is composed of a single-stage cur- high differential gain, maximum CMRR to mitigate PLI, and
rent reuse amplifier with a pseudo-resistor, common-mode low IRN. Achieving these criteria is facilitated by employ-
feedback, a DC servo loop utilizing an active RC integrator, ing a CMOS-based switched capacitor feedback amplifier
an efficient ripple reduction technique, and a chopping spike (SCFA) OTA with a differential configuration, input chopper
filter. The total IRN in the bandwidth is as low as 2.9 µVrms , modulators, and pseudo resistors. This paper also includes a
while the mid-band gain is approximately 40 dB. With the comprehensive evaluation of the recommended amplifier in
positive feedback loop, the input impedance is 17 M Ω, and relation to existing amplifiers, proving the improved perfor-
the chopping frequency is set at 20 kHz. mance of the new design.
Deng et al. proposed a DC servo loop (DSL) that utilizes The work reported in this paper is organized into four sec-
an active Gm-C integrator to eliminate electrode-dc offset tions. The implementation and design of switched capac-
(EDO) [11]. This DSL has a low power consumption of only itor feedback amplifier described in section II. Section A.
3.24 µW per channel under a 1.8 V supply voltage, a supply presents the proposed circuit architecture; section B. repre-
current of 1.8 µA, and a gain of 40 dB. sents the gain enhancement with current mirror OTA topol-
Jie Zhang et al. proposed a capacitive feedback ampli- ogy; section C. presents the pseudo-resistor (Rp); and section
fier that utilizes a current-reused OTA for ECG recordings D. presents the low-noise input chopper modulator. Section
[12]. This amplifier power consumes 160 nA from a 2 V III. comprises various simulation results performed using ca-
supply yet manages to deliver an IRN of 2.05 µVrms . The dence virtuoso. Finally, the conclusion is given in section IV.
CMRR and PSRR were measured to exceed 65 dB and 70
II. I MPLEMENTATION AND D ESIGN OF S WITCHED
dB, respectively, demonstrating the amplifier’s high level of
C APACITOR F EEDBACK A MPLIFIER
rejection for common-mode and power supply noise.
Rajasekhar et al. introduced the split differential pair tech- A. Circuit Architecture
nique, which enhances the OTA gain without requiring addi- Figure 1 depicts the proposed overall system architecture
tional bias current [13]. The proposed circuit achieves an of the bio-potential amplifier. It consists of a low noise in-
impressive 65 dB CMRR along with a 3-dB bandwidth of 17 put chopper modulator, a capacitive feedback OTA ampli-
kHz. The circuit only draws a static current of 0.6 µA and fier with Cin and Cf as feedback capacitors, and pseudo-
operates from a modest 1 V supply. resistors (RP ), which offer significantly higher resistance.
To assess the effectiveness of the proposed biopotential Amplification circuits are commonly employed to en-
amplifier, a comparison was made between its performance hance the magnitude of electrode output signals prior to sub-
parameters given in the simulation section. A full review sequent processing or analysis. Amplifiers have the ability
of the bio-potential amplifiers of prior works can be found in to enhance low-strength signals to levels that are appropri-
[14, 15, 10, 12, 11, 16, 17, 13, 9, 8, 18, 19, 20, 21, 22, 23, 24], ate for measurement and recording devices, resulting in im-
it is observed that: proved accuracy of the signal and signal-to-noise ratio. An
−the gain of amplifier in [14, 15, 10, 12, 11, 16, 17, 13, 9, input modulator involves the modulation of the input signal
8, 18, 19, 21, 23, 24] is less than 50 dB. using a high-frequency square wave. The implementation
−the CMRR shown in [14, 15, 10, 12, 16, 17, 13, 9, 8, 18, of a chopper modulator can enhance the bio-potential ampli-
20, 21, 22, 23, 24], is relatively low, which is less than 100 fier’s efficiency by mitigating the impact of low-frequency
dB. noise, such as 1/f noise or flicker noise. To reduce these
−the input impedance is less than 1 GΩ in Refs [12, 11, noises, the chopping frequency (fchop ) of 30 kHz chosen is
19, 20]. higher than the corner frequency (fc ) [25]. Additionally, it
Journal of Integrated Circuits and Systems, vol. 19, n. 02, 2024 3
The amplifier closed loop transfer function [13] can be ex-
pressed as:

 
Vout W LCox 1
= 1+ AM 1 ,
Vin Cin 1 + LG(s)
 " #
W LCox Cin 1 sRP Cf
= 1+ .
Cin Cf (1 + sCLef f ) (1 + sRP Cf )
βGm
(3)
From (3) Cin /Cf ratio is a measure that can be used to
calculate mid-band gain (AM ). The lower cutoff frequency
is 1/Rp Cf , and the higher cutoff frequency is βGm /CLef f .
The electrode resistance (Rs ) has been omitted in the above
analysis since it is very minor compared to the feedback
Fig. 1: The proposed architecture of the Switched Capacitor Feedback Am- component (Rp ) and provides a high-frequency 1/Rs Cin ,
plifier using Chopper Modulator. which is usually more than the amplifier bandwidth. The Rp
and OTA contribute noise to amplifiers, which is substan-
can assist in alleviating offset and drift issues that may arise tially lower due to the extremely low bandwidth [14].
in low-level amplification tasks. The design of the bio-potential amplifier must take into
In order to reject the input DC offset voltage, an AC cou- account the trade-offs between noise and power use. Device
pling capacitor (Cin ) is required. Subsequently, the chop- size is carefully chosen to maximize amplifier factors such as
per modulator’s modulated signal is transmitted through Cin input-referred noise, low-frequency gain, input impedance,
to eliminate any residual DC component before being intro- and bandwidth in order to achieve the desired channel inver-
duced to a high-gain amplifier for further amplification and sion level.
processing. B. Gain Enhancement with current Mirror OTA Topol-
As depicted in Figure 1, the Miller integrator (Cf ) of the ogy
amplifier ‘A’ output voltage appears at VOU T . The output
The OTA design seen in Figure 2 is performed with the
sample will be integrated and the corresponding feedback
current mirrors. The input transistors M1 and M2 are drawn
voltage will appear at terminal Vin . As a result, node Vin
with identical aspect ratios, and their transconductances are
tracks the dc level seen by ‘A’ at Vip , which effectively sup-
denoted as gm1 and gm2 , respectively. Similarly, the circuit
presses the drift, dc offset, and other low-frequency input
has nMOS current mirror transistors M3, M4 whose aspect
voltages [13].
ratio is (W/L)3 and transconductance gm3 , and M5, M6 are
Any amplifier with negative capacitive feedback must also
having the aspect ratio is (W/L)5 , transconductance gm5 .
include DC feedback to maintain the optimum operating
The transistors M7 and M8 in the pMOS current mirror have
point. Connecting a high-value resistor that constructs a sub-
a (W/L)7 aspect ratio, and their transconductance is denoted
threshold region-based pseudo resistor (Rp ) is required in
by gm7 .
parallel with Cf . The RC feedback network determines the
When a small AC input voltage is applied to the OTA, the
lower cut-off frequency, and the unity gain bandwidth deter-
differential amplifier drain current gets the value of Gm vin .
mines the higher cut-off frequency. To maintain symmetry
This AC differential drain current is cascoded to current mir-
and optimize offset, the OTA non-inverting terminal’s capac-
ror MOSFETs M3 through M5 and MOSFETs M4 through
itance (Cf ) is grounded [12].
M6, ensuring that the drain current of M5 is cascoded to cur-
The loop gain [13] can be expressed as:
rent mirror MOSFETs M7 through M8.
βGm Ro The output voltage of the OTA is
LG(s) = . (1)
1 + sRo CLef f
vout = Gm vin × Ro . (4)
Cf
Here the feedback factor of an amplifier β is Cf +Cp +Cin , Where, R0 =(R looking into drain of M6)∥ (R looking into
Gm is the transconductance of OTA, Ro is the output resis- drain of M8),
tance of OTA, and CLef f is effective load capacitance, Cp is
input parasitic capacitance, and load capacitance (CL ).The
feedback network influences the effective capacitance, in- R0 = [ro6 (1 + gm6 ro4 )]∥[ro8 (1 + gm8 r07 ]. (5)
cluding a fraction of the capacitance that is attributed to The output resistance Ro ∼ = (r06 ∥r08 ), multiplied by a
its presence, and which consists of the apparent capacitance corresponding factor that is less than unity. The output re-
portion based on the feedback system. sistance and effective transconductance (Gm ) are two of the
The effective load capacitance is expressed as: most important factors in determining the OTA gain. The
OTA gain may be increased by either enhancing Gm or Ro .
Cf (Cp + Cin ) For a given device size, increasing transconductance requires
CLeff = CL + . (2) increasing the bias current, making this technique inefficient
Cf + Cp + Cin
4 Bhaskara et al.: High Gain and CMRR Bio-Potential Amplifier Design Using SCFA
while earlier RP were biased in the subthreshold inversion
point [17].

In the RP , the M P5 and M P6 transistors are biased in the

saturation level when the voltage difference between node
A, and node M is low and the average of node A and B volt-
ages is equal to the node M. In situations where the voltage
drop over the resistance path RP is minimal, the average of
the input gate voltages of both M P5 -M P8 transistors of the
differential pair serves as their common source voltage. In
other words, the differential pairs perform the function of an
averaging circuit, and the voltages of their common source
are determined by their gate voltages [30].

If there is a significant voltage difference between node

A and node M and the voltage at node M is greater than
Fig. 2 Schematic of OTA used in proposed amplifier [26]. that at node A, transistors M P5 and M P6 will switch on
and off, respectively. This results in node A acting as the
source for transistor M P1 , which will have a negative VSG .
in terms of power consumption. The bias current and cas- As well, if the voltage at node M is lower than the voltage
code bias voltages were produced using conventional circuits at node A, transistors M P6 and M P5 will switch on and
[27]. To enhance their Gm , the input pair transistors function off, respectively. Consequently, it represents the source of
with weak inversion [28, 29]. the transistor M P1 , and this transistor once again exhibits a
The output transconductances of transistors M6 and M8 negative VSG . As a consequence of this, the bias circuit of
determine the output current. The CMRR value is known this RP consists of two-level shifters, which are responsible
to be highly dependent on the signal frequency; thus, it is for producing a negative VSG for M P1 and M P2 transistors
important to consider the function of differential signal gain [31].
and common signal gain.
The differential gain of the amplifier Adm is To estimate the resistance value of the RP , a DC voltage is
fixed on one of its terminals, and the other terminal is swept.
Adm = gm1,2 (r06 ∥r08 ) . (6) These enhancements are realized because the suggested RP
The common mode gain of the amplifier Acm is is biased in the cut-off area with a nearly constant negative
VSG for M P1 and M P2 , and all of them are of the same de-
sign. By varying the input voltage, the RP equivalent resis-
Acm = 2 (gm3,5 gm4,6 + gm4,6 gm7,8 + gm7,8 gm3,5 ) tance is also approximately symmetric. Suppose the voltage
(1 + 2gm1,2 Rss ) drop across the RP terminals is raised. In that case, one of
× . (7) the differential pairs in the biasing circuit will be switched
gm1,2
off, acting as a simple level shifter and reducing the pseudo
The CMRR of the amplifier is resistor’s linearity. Even under these difficult situations, the
recommended pseudo resistor’s linearity is still significantly
2
gm1,2 (r06 ∥r08 ) superior to that of the traditional RP .
CM RR =
2 (gm3,5 gm4,6 + gm4,6 gm7,8 + gm7,8 gm3,5 )
1
× . (8)
(1 + 2gm1,2 Rss )
Here Rss is the resistance of the current source, and the
OTA gain of transconductance is Gm = gm1,2 .
The amplifier CMRR has been improving by increasing
transconductances gm1 , gm2 . For this reason, the best way to
maximize the transconductance of the input transistors is to
work in weak inversion. Moreover the transconductance of
transistors gm3−7 << gm1,2 , which can minimize the noise
contributions of transistors M3-M8.
C. Pseudo-Resistors
The pseudo resistor (RP ) shown in Figure 3, transistors
M P1 and M P2 work as the resistor, and are used in series
to increase the amount of resistance. The M P3 -M P8 tran-
sistors are operated as the biasing circuit. The RP attained
for low cutoff frequencies, which offer a significantly higher
resistance despite taking up less space than passive resistors, Fig. 3 Schematic of Pseudo-Resistor (RP ) [17].
Journal of Integrated Circuits and Systems, vol. 19, n. 02, 2024 5
D. Low Noise Input Chopper Modulator
The chopper modulator operation and principle are de-
picted in Figure 4, which works by mixing an input sig-
nal x(t) with low-frequency. The periodic representation of
the signal m(t) is multiplied with the input signal of the in-
put chopper modulator to get the odd harmonics of fchop′ s .
The signal is amplified by the flipped transistors with the
transfer function of A(f), followed by the demodulation with
m(t). This returns the modified spectrum to its original posi-
tion, leaving duplicates at fchop′ s odd harmonics. To prevent
aliasing, the signal bandwidth must be less than fchop /2 [32].

Fig. 5 Schematic of a Input Chopper Modulator [33].

Flicker noise, sometimes referred to as 1/f noise or pink

noise, is a form of noise signal characterized by a PSD that
exhibits an inverse relationship with frequency [35]. In al-
ternative terms, it can be observed that as the frequency of
a signal increases, the amplitude of the accompanying noise
decreases. The term pink noise is derived from its resem-
blance to the color pink in the visible spectrum, as both ex-
hibit comparable power distributions [6].
The power spectral density (PSD) at the output of the
amplifier circuit is evidently influenced by the PSD of the
Fig. 4 Operation principle of Chopper Modulator[32]. source, which undergoes autozeroing. The PSD of the input-
referred noise of an amplifier typically consists of two com-
The schematic of an input chopper modulator, as depicted ponents: white noise and 1/f noise.
in Figure 5 is implemented with nMOS switches. The signals The input-referred noise of a CMA quantifies the magni-
m(t) and m(t) periodically flipped and applied to the tran- tude of noise present at the amplifier’s input. It is a com-
sistor switches MC1, MC2, MC3, and MC4, respectively. bination of the input-referred voltage noise of the chopper
During the switching process of nMOS transistors, a specific modulator and the input-referred voltage noise of the ampli-
magnitude of electrical current is observed to traverse from fier itself.
the source to the drain of the MOS transistor. However, it Accordingly, it is possible to determine the amplifier’s
is important to note that a small discrepancy is consistently baseband total input referred noise as [33]:
present in the amplifier’s input capacitance (Cin ), despite the
fact that the charge injection of transistors MC1 and MC2  
is equivalent and appears to be in common mode at the in- fC,1/f
Sin,total (f ) ∼
= So 0.8525 +1 ,
put. In the same way, the mismatch between MC3 and MC4 fchop
causes an equal amount of charge to be added to the amplifier fC fC
input capacitance (Cin ) [33]. for ≫ 1 and ≤ 0.5. (9)
fchop fchop
The chopper modulation approach can enhance the
CMRR of the differential amplifier and remove the flicker Here, So represents the PSD of white noise (thermal
noise of the CMOS transistors [34]. The advantage of this noise), while fc denotes the corner frequency. The cor-
technique is that any interference that is present in the input ner frequency (fC,1/f ) is defined as the frequency at which
signal, such as noise or electrical interference, is also mod- the PSD of 1/f noise becomes equal to the PSD of white
ulated by the square wave. This means that when the sig- noise. The So is not affected in the  chopping operation
fC,1/f
nal is demodulated, the interference is removed along with (fc ≫ fchop ) and So 0.8525 fchop is PSD of the input-
the modulation, leaving only the desired signal. Increas- referred 1/f noise of an amplifier.
ing the chopper frequency can help improve the CMRR by In conclusion, if fchop is significantly greater than fc,1/f ,
reducing the effects of low-frequency interference. Conse- the chopper modulation approach may efficiently reduce the
quently, it may be filtered off in the same manner as the flicker (1/f) noise transistor with no impact on the thermal
flicker (1/f) noise and the offset voltage. An intrinsic lim- noise. By minimizing the input-referred noise, the sensitivity
itation of chopper-modulated amplifiers (CMA) is their DC and accuracy of the CMA can be improved, enabling high-
coupling. Therefore, the presence of biopotential electrodes precision amplification of low-level signals.
further amplifies the differential direct current (DC) elec- Overall, a chopper modulator using CMOS transistors is
trode offset voltage. a simple and effective way to modulate a DC voltage signal.
6 Bhaskara et al.: High Gain and CMRR Bio-Potential Amplifier Design Using SCFA

PSRR(dB)
CMRR(dB)
Bandwidth (Hz)
Gain (dB)
Power (µW )
VDD(V)
Topology
Technology
Work
The high-frequency switching between the two phases of the

Frequency
Chopping
Consumption (A)
Current
Noise (µVrms )
Input Referred
Impedance (Ω)
Input
switching cycle helps to eliminate noise and distortion, re-
sulting in a high-quality output signal.

III. S IMULATION R ESULTS

2.26l µ

3.19

77.6
76
0.3–4.4 k
39.75
4.07
1.8
CCIA
0.18 µm
[14]
3.27

54.99
66.55
5.69–5.45 k
40.02
1.53
1.8
DRFC-OTA
0.18 µm
[15]
20 kHz

3.7 µ

2.9

80
94
50 m-10 k
40
4.4
1.2
OTA
0.18 µm
[10]
160 n

2.05

20 M
>70
>65
0.2-200
39.8

2
CT-CMFB
0.13 µm
[12]
20 kHz

1.8 µ

0.65-2.14

440 M

>100
200–5k
40
3.24/ch
1.8
Chop-Amp
0.18 µm
[11]
Table II. The proposed circuit compare to various recording methods
0.12 µ

1.5

75
78
0-120
42.3

1.2
LNA
45 nm
[16]
Fig. 6 Layout of the proposed circuit.

25 kHz

2.02 µ
2.86 (AP)
0.63 (LFP)
1.8 G

0.1–5.1 k
26.04
5 /ch
1.8
CCIA
0.18 µm
[17]
The Bio-potential amplifier presented in Section II. im-
plemented using OTA with a current mirror at CMOS 180
nm technology node and performance is evaluated at a sup-
0.6 µ

7.81

65
17 k
30

1
pair
Split-Diff.
65 nm
[13]
ply voltage of 0.8 V. Figure 6 shows a layout representation
of the proposed architecture. The design area of the simu-
lated capacitive feedback amplifier is 0.012 mm2 , and MIM
20 kHz

49
65

12
1.6
1
CCIA
45 nm
[9]
capacitors are used to realize the on-chip nature of all the
capacitors. In literature survey of the suggested biopotential
20 kHz

29.8 µ

1.3

67
86
20-11k
39.3
35.8
1.2
AFE
0.13 µm
[8]
amplifier’s performance of electrical characteristics is given
in Table II.
The designed amplifier chosen with Cin = 3 µF, Cf = 1
2.14 µ

2.1

4.6 G

40
2.14
1
CCIA
0.18 µm
[18]
pF, CL = 3 pF and fchop = 30 kHz to stabilize the chopping
signal. A sinusoidal signal with an amplitude of 100 mV,
a frequency of 100 Hz, and a DC offset of 50 mV to ac-
0.6

150 M

65-105
0.5-1 k
30-45
0.507
1.2
AFE
0.18 µm
[19]
count for the electrode offset voltage is used to simulate the
circuit working. The achieved closed-loop differential gain,
10 kHz

18 µ

220 n

1.2 G
88
84
300
57
4.5
0.75/1
ERG CCIA
0.18 µm
[20]

common-mode gain, and CMRR are 57.7 dB, -50.43 dB, and
6.6 µ

1.4

2.6 G
85
93
0.5–250
40
11.88
1.8
CRIA
0.18 µm
[21]
0.97 (LFP)
0.44 (AP)

250-5 k (AP)
0.1-250 (LFP)
50.3 - 63.4
1.54 - 1.94
1.8
AFE
0.18 µm
[22]
54.7
59.8

38.2
4.71
1.5
CCFA
0.18 µm
[23]
1.8 (LFP)
5.3 (AP)
1.6 G

200-5 k (AP)
0.1-200 (LFP)

1.2
CCIA
40 nm
[24]
30 kHz

0.3882 n

0.83956

1.545 G
>90
108.13
0.1-2 k
57.7
8.854
0.8
SCFA
0.18 µm
This work

Fig. 7: Simulated CMRR, Differential Gain, and Common Mode Gain of

the proposed amplifier.
Journal of Integrated Circuits and Systems, vol. 19, n. 02, 2024 7

Fig. 8 Input Noise Analysis of proposed circuit. Fig. 11 Power Spectral Density of transient response.

Fig. 9 Output Noise Analysis of proposed circuit. Fig. 12 Differential Gain Corner Analysis of proposed design.

Fig. 10 Simulated power dissipation of overall amplifier structure. Fig. 13 CMRR Corner Analysis of overall design.
8 Bhaskara et al.: High Gain and CMRR Bio-Potential Amplifier Design Using SCFA
the noise is adjusted in order to determine the optimal values.
Figures 12,13 demonstrate the results of the corner anal-
ysis performed to assess the circuit’s performance in chal-
lenging gain, CMRR, and temperature conditions. The sim-
ulation results indicate that the SS, FF, and standard corners
provide high performance suitable for bio-potential signals.
The performance of the amplifier involves various corners
of voltage, and temperature across various operating condi-
tions. The system evaluated in the standard corner that tem-
perature and power supply variations had very little impact
on the performance of the circuit, as depicted in Figure 14.
The switched capacitor chopper amplifier’s performance un-
der 180 nm CMOS technology was evaluated at 30 kHz us-
ing 200 Monte Carlo simulations that considered amplifier
gain performance, as depicted in Figure 15. The simulation
Fig. 14: Differential gain with different temperature and supply for the am-
mean value of the differential gain is 56.79 dB, and the stan-
plifier. dard deviation is 0.0919 dB, respectively.

IV. C ONCLUSION
This paper describes a switching capacitive-feedback am-
plifier with low noise and high CMRR that satisfies the se-
vere criteria of bio-potential amplifier recordings. The OTA
uses a current mirror-based differential pair design to at-
tenuate the common-mode signal and eventually boost the
CMRR. The amplifier was created utilizing CMOS technol-
ogy with a 180 nm feature size. About 200 monte carlo sim-
ulations were carried out in order to simulate the CMRR, dif-
ferential mode gain, and common mode gain. The modified
SC amplifier with a chopper modulator had a mean CMRR
value of 108.13 dB and a gain value of 57.7 dB. Finally, it
has been demonstrated that the CMRR may be enhanced by
applying capacitive load to other SCFA amplifiers if no load-
ing influences the differential gain. This modified amplifier
is an excellent alternative where the input signal is signifi-
Fig. 15 Monto-Carlo analysis of differential gain output. cantly weaker than the common mode signal, such as those
found in bio-potential amplifiers.
108.13 dB respectively as shown in Figure 7. When the gain ACKNOWLEDGEMENT
drops by 3 dB, the range of frequencies is from 0.1 to 2 kHz
and bandwidth is limited. We sincerely thank VIT-AP University for providing re-
sources and facilities for the smooth functioning of the re-
The amplifier’s measured IRN spectrum, which is spec-
search work.
ified by the ratio of the output noise spectrum to the am-
plifier’s mid-band gain. The total IRN of the amplifier is C ONFLICT OF INTEREST
0.83956 µV rms, and the input impedance is 1.545 GΩ. The
proposed circuit is designed for analyzing biological signals, The authors declare no Conflict of interest.
which are extremely small in amplitude. As a result, simu- DATA AVAILABILITY STATEMENT
lation work has been conducted to investigate the effects of
noise. The Figures 8, 9 presented illustrate the optimized The data supporting the conclusions of this study may be
noise performance of the proposed circuit. The obtained from the corresponding author upon a reasonable
√ input noise
and output noise were measured at 5.3 µV / Hz and 1.02 request.
√
nV / Hz , respectively. Furthermore, the average power dis-
R EFERENCES
sipation is relatively high, limited to 8.854 µW as shown in
Figure 10. [1] R. M. Rangayyan, Biomedical signal analysis. John Wiley & Sons,
2015.
The spectrum obtained from the spectrum analyzer is uti-
lized for analysis in the subsequent procedures. As depicted [2] A. Palumbo, P. Vizza, B. Calabrese, and N. Ielpo, “Biopotential sig-
in Figure 11 the frequency is represented on the X-axis, and nal monitoring systems in rehabilitation: A review,” Sensors, vol. 21,
the power spectrum amplitude is represented on the Y-axis. no. 21, p. 7172, 2021.
When the highest point of the spectrum aligns with the fre- [3] J. Rajeswari and M. Jagannath, “Advances in biomedical signal and
quency of 30 kHz, it can be inferred that the amplitude of the image processing–a systematic review,” Informatics in Medicine Un-
input noise is at its optimal level. Therefore, the amplitude of locked, vol. 8, pp. 13–19, 2017.
Journal of Integrated Circuits and Systems, vol. 19, n. 02, 2024 9
[4] K. A. Ng and Y. P. Xu, “A low-power, high cmrr neural amplifier [20] L. Fang and P. Gui, “A low-noise low-power chopper instrumentation
system employing cmos inverter-based otas with cmfb through supply amplifier with robust technique for mitigating chopping ripples,” IEEE
rails,” IEEE Journal of Solid-State Circuits, vol. 51, no. 3, pp. 724– Journal of Solid-State Circuits, vol. 57, no. 6, pp. 1800–1811, 2022.
737, 2016.
[21] W. Cao, Y. Liu, S. Liu, L. Wang, R. Ma, and Z. Zhu, “A 2.6 gω ,
[5] Y.-Y. Qian, Z.-G. Wang, Y.-K. Liu, and Z.-J. Zhou, “A high gain and 1.4 µvrms current-reuse instrumentation amplifier for wearable elec-
high cmrr instrumentation amplifier for biomedical applications,” in trocardiogram monitoring,” Microelectronics Journal, vol. 107, p.
2019 IEEE 4th International Conference on Integrated Circuits and 104940, 2021.
Microsystems (ICICM). IEEE, 2019, pp. 61–64.
[22] N. Tasneem and I. Mahbub, “A low-power reconfigurable readout cir-
[6] K. Aghababaiyan, “Improving performance of neurons by adding cuit with large dc offset reduction for neural signal recording applica-
colour noise,” IET nanobiotechnology, vol. 14, no. 5, pp. 433–439, tions,” in 2020 IEEE 63rd International Midwest Symposium on Cir-
2020. cuits and Systems (MWSCAS). IEEE, 2020, pp. 521–524.

[7] B. Ando and S. Graziani, “Adding noise to improve measurement,” [23] M. Martincorena-Arraiza, A. Carlos, A. Carlosena, and A. Lopez-
IEEE Instrumentation & Measurement Magazine, vol. 4, no. 1, pp. Martin, “Ac coupled amplifier with a resistance multiplier technique
24–31, 2001. for ultra-low frequency operation,” AEU-International Journal of
Electronics and Communications, vol. 149, p. 154176, 2022.
[8] C. Cabrera, R. Caballero, M. C. Costa-Rauschert, C. Rossi-Aicardi,
and J. Oreggioni, “Low-voltage low-noise high-cmrr biopotential in- [24] H. Chandrakumar and D. Marković, “An 80-mvpp linear-input range,
tegrated preamplifier,” IEEE Transactions on Circuits and Systems I: 1.6- G Ω input impedance, low-power chopper amplifier for closed-
Regular Papers, vol. 68, no. 8, pp. 3232–3241, 2020. loop neural recording that is tolerant to 650-mvpp common-mode in-
terference,” IEEE Journal of Solid-State Circuits, vol. 52, no. 11, pp.
[9] P. Jain and A. M. Joshi, “Low leakage and high cmrr cmos differential 2811–2828, 2017.
amplifier for biomedical application,” Analog Integrated Circuits and
Signal Processing, vol. 93, no. 1, pp. 71–85, 2017. [25] D. A. Johns and K. Martin, “Analog integrated circuit design john
wiley & sons,” New York, USA, 1997.
[10] E. Shad, M. Molinas, and T. Ytterdal, “A low-power and low-noise
multi-purpose chopper amplifier with high cmrr and psrr,” in 2020 [26] R. R. Harrison and C. Charles, “A low-power low-noise cmos am-
42nd Annual International Conference of the IEEE Engineering in plifier for neural recording applications,” IEEE Journal of solid-state
Medicine & Biology Society (EMBC). IEEE, 2020, pp. 3998–4001. circuits, vol. 38, no. 6, pp. 958–965, 2003.

[11] D. Luo, M. Zhang, and Z. Wang, “A low-noise chopper amplifier de- [27] D. A. Johns and K. Martin, Analog integrated circuit design. John
signed for multi-channel neural signal acquisition,” IEEE Journal of Wiley & Sons, 2008.
Solid-State Circuits, vol. 54, no. 8, pp. 2255–2265, 2019.
[28] E. Sackinger and W. Guggenbuhl, “A high-swing, high-impedance
mos cascode circuit,” IEEE Journal of Solid-state circuits, vol. 25,
[12] J. Zhang, H. Zhang, Q. Sun, and R. Zhang, “A low-noise, low-power
no. 1, pp. 289–298, 1990.
amplifier with current-reused ota for ecg recordings,” IEEE transac-
tions on biomedical circuits and systems, vol. 12, no. 3, pp. 700–708, [29] R. Saha, B. Bhowmick, and S. Baishya, “Temperature effect on
2018. rf/analog and linearity parameters in dmg finfet,” Applied Physics A,
vol. 124, no. 9, pp. 1–10, 2018.
[13] R. Nagulapalli, K. Hayatleh, S. Barker, S. Zourob, and N. Yassine,
“An ota gain enhancement technique for low power biomedical appli- [30] R. Nagulapalli, K. Hayatleh, S. Barker, P. Georgiou, and F. Lidgey,
cations,” Analog Integrated Circuits and Signal Processing, vol. 95, “A high value, linear and tunable cmos pseudo-resistor for biomedical
pp. 387–394, 2018. applications,” Journal of Circuits, Systems and Computers, vol. 28,
no. 06, p. 1950096, 2019.
[14] R. Sanjay, V. Senthil Rajan, and B. Venkataramani, “A low-power low-
noise and high swing biopotential amplifier in 0.18 µm cmos,” Analog [31] H. Rezaee-Dehsorkh, N. Ravanshad, R. Lotfi, and K. Mafinezhad,
Integrated Circuits and Signal Processing, vol. 96, no. 3, pp. 565–576, “A linear tunable amplifier for implantable neural recording applica-
2018. tions,” in 2011 IEEE 54th International Midwest Symposium on Cir-
cuits and Systems (MWSCAS). IEEE, 2011, pp. 1–4.
[15] S. R. Venishetty, S. Kumaravel, and S. A. Durai, “A power efficient
low-noise source degenerated bio-potential amplifier,” Analog Inte- [32] A. Adesara and A. Naik, “A 5.5 µw 42 nv/ hz chopper stabilized
grated Circuits and Signal Processing, vol. 103, no. 2, pp. 291–301, amplifier for biomedical application with input impedance enhance-
2020. ment,” International Journal of Electronics and Telecommunications,
pp. 523–530, 2021.
[16] P. V. Lakshmi, S. Musala, and A. Srinivasulu, “A high gain low noise
amplifier for implantable cardio technologies,” Microsystem Tech- [33] R. F. Yazicioglu, C. Van Hoof, and R. Puers, Biopotential readout cir-
nologies, vol. 28, no. 4, pp. 1047–1057, 2022. cuits for portable acquisition systems. Springer Science & Business
Media, 2008.
[17] F. Ansari and M. Yavari, “A fully-differential chopper capacitively-
coupled amplifier with high input impedance for closed-loop neural [34] C. C. Enz, E. A. Vittoz, and F. Krummenacher, “A cmos chopper am-
recording,” Circuits, Systems, and Signal Processing, vol. 41, no. 7, plifier,” IEEE Journal of Solid-State Circuits, vol. 22, no. 3, pp. 335–
pp. 3679–3705, 2022. 342, 1987.

[18] X. T. Pham, D. N. Duong, N. T. Nguyen, N. Van Truong, and J.-W. [35] D.-H. Kyon, W.-H. Lee, M.-S. Kim, and M.-J. Bae, “Hi-pass pink
Lee, “A 4.5 gω -input impedance chopper amplifier with embedded noise: Its acoustic features and standard volume,” International Jour-
dc-servo and ripple reduction loops for impedance boosting to sub- nal of Multimedia and Ubiquitous Engineering, vol. 8, no. 6, pp. 229–
hz,” IEEE Transactions on Circuits and Systems II: Express Briefs, 236, 2013.
vol. 68, no. 1, pp. 116–120, 2020.

[19] Y. Zeng, Z. Li, W. Chen, W. Zhou, Y. Bao, Y. Chen, and Y. Li, “A

672 nw, 670 nvrms ecg acquisition afe with noise tolerant heartbeat
detector,” IEEE Open Journal of Circuits and Systems, 2023.