To Develop Efficient layout of CMOS circuit for SRAM and

compare with existing technique to show the proposed

technique is efficient
Prof Ravi H.Talawar Dr. Ramachandra A. C.
Assistant Professor Professor & Research Supervisor
Department Electronics and Communication Engg. (Visvesvaraya Technological University, Belagavi)
Visvesvaraya Technological University, Belagavi Department of Electronics and Communication Engg.
Nitte Meenakshi Institute of Technology, Bangalore

ABSTRACT
Different SRAM cell typologies includes 10T, 9T, 8T, and 7T in performance
improvement and stability that enhances in specific regions as a trade-of. The count of
the transistors have the possibility to be reduces to compensate certain area with the
dynamic CMOS logic utilization that have the possibility in high performance
maintenance. The current research focused on a novel sleepy technique along with
AVS design is proposed for low-power SRAM for reducing power consumption
utilizing multi-threshold CMOS circuit. Therefore, the SRAM us implemented in the
research using sleep transistors along with an additional leakage current feedback-
transistor along with MTCMOS primary structure which is implemented with an
effective results in terms of certain factors such as delay, area, and power.

Keywords: SRAM, Leakage, Sleepy Keeper Transistor, Dynamic Power, CMOS

design, PMOS design.

I. INTRODUCTION
In the previous researches, miniaturization and performance of an associated device
was considered as the main design issues, especially of a VLSI designer. Recently, a
high-performance integrated circuits that are executed in deep sub-micron technology.
There are plenty of previous researches has been carried out in the experimentation
process with the portable devices development and design for emerging applications
that includes space applications, wireless body-sensing networks, medical equipment
implant, etc. Further, the utilization of scale-down in the node of technology,
outcomes to enhance in the dissipation of static power. In order to reduce the leakage-
power dissipation, different power reduction methods are used that includes leakage
feedback, stack keeper along with body-bias, and sleepy keeper [2].
Moreover, the utilization of static RAM in embedded controllers needs minimum
write and read time. Because of the significant enhancement of lower voltage and
power on memory systems recently, SRAM [8] highlight in the certain research
sector. The development of On-chip memory with the array that are tightly packed
with the Static RAM cells that enables increased quality. A 6T-SRAM [10] has been
used for a memory cell and the two types of power dissipation presented in an
electronic circuit such as static power dissipation and switching power. In the time of
performance on active mode during the device is under ON-state, hence the power
dissipation is considered as a major process due to the semiconductor static
components and switching power. With the execution of sleep or standby mode
operation, the devices are found in the OFF state, for this the standby leakage current
is considered as a responsible process for the dissipation of power. With the
utilization of technology measuring the leakage power was commanding and the
dynamic power, therefore is considered as a primary design issue to the VLSI
designers, which was considered as that most of the devices that are portable with the
battery operated [1].

While designing the low-power VLSI circuit, the dissipation of power is considered
as one of the challenging problems that is integrated with threshold voltage.
Therefore, the minimization of threshold voltage maximizes the sub-threshold current
leakage with the leakage power-dissipation enhancement that plays a significant part
in total dissipation of power. Because of the leakage power problem, certain devices
that are executed by battery for a long period in the mode of standby that are drained-
out fast. To reduce the issues on SRAM was developed with dual-control-stacked-
inverter that exploits dual-control-signals [6].

The current research mainly focused on a novel sleepy stack technique with adaptive
voltage scaling -AVS design development for low-power SRAM [9], [10] for
minimizing power consumption utilizing multi-threshold CMOS circuit. The main
contribution of the current research as follows: Section 2 explains the existing
researches on low-power SRAM with the certain factors analyze and work process on
different design metrics. Section 3 defines the proposed method for low power SRAM
for reducing power-consumption using multi-threshold CMOS circuit. Section 5
explains the results and analysis with the power optimization methodologies and
parameters. Finally, Section 6 explains the conclusion of the paper.

II. LITERATURE REVIEW

In [3], the research utilized an SEHF11T (11 transistors) processed on the SRAM cell
with the high RSNM - read-static-noise-margin and write-static-noise-margin -
WSNM. Therefore, the eradicates the write half-select-disturb utilizing cross-point-
data-aware write-word-lines, that have the possibility to mitigate the structure of bit-
interleaving to minimize multiple-bit-upset and maximizes soft-error immunity. This
method has been analyzed and estimated with the PVT effect that defines
temperature, voltage, and process variations on different design performance metrics
when compared to the other cells. Further, the utilization of WSNM enhancement by
eradicating the cross-coupled-inverters pair feedback while write operation that
indicates the power-cutoff-write-assist method. Nevertheless, SEWA10T cell required
more period to execute the read operation because of the decoupled-reading-buffer
employment, developed by three series-connected-transistors.

In [4], KLECTOR - Keeper-in-Leakage-Control-Transistor has been used in the

research to minimize leakage currents presented in SRAM architecture. This research
focused on the SRAM impact on the leakage current in the “standby mode”, that was
processed by the fabric that minimize threshold voltage. Normally, KELECTOR
circuit developed with minimum power consumption by limiting the current flow via
devices of minimum voltage drops and completely dependent on the self-controlled
transistor exhibited at the outcome node. It was denoted that the outcome on the
static-leakage power in the write-operation, which was minimized to 63% & 69%
executed for the operation (read). This design was experimented with the Cadence
EDA, and Virtuoso tool.

In [5], the main focus of the research to include the procedures on current leakage-
power procedures with a super low-voltage and low-power SRAM, that was executed
using each method. By executing the analysis in memory cell, an unmistakable
consideration was captured with the development of memory cells. This research
utilized 16 x 16 SRAM array structure, which was developed utilizing SRAM,
column decoder, address, and sense amplifier design. This research explained the
SRAM cell design was have the capability to diminish the power dissipation leakage.
The outcome obtained from the research with a 47.81% and 53.63% minimized in
power dissipation in the absence of performance destruction executed in the array
structure and memory-cell-level method. Possibly, the significant issues that are
considered in the research are examined with the viable and new circuit execution
method that minimized the power dispersal in the absence of circuit model
compensation execution.

In [7], modified SRAM architecture has been utilized with minimum power
dissipation. This research mainly concentrated on the utilization of 9T-SRAM-Cell to
optimize a conventional SRAM with a single-bit-memory cell and therefore, designed
a stable procedure along with low-power consumption and achieved low PDP -
Power-Delay-Product by differing executing frequency measured in range of MHz.
Further, a 4x4 SRAM array was executed with the comparative analysis amidst 9T-
SRAM cell and 6T-SRAM cell with the minimization of power at 62.83% acquired in
this system with the previous system at a 2GHz operating frequency. This paper
defines a 62.27% power reduction was acquired for the structure of array and the
single-bit 9T PDP - power-delay-product and power dissipation which was lesser,
when compared to the conventional 6T-SRAM. Also, this research implemented with
the utilized method of SRAM to an array with the connecting peripherals. Here, used
a forced sleep method, which was a combination of sleep effect and forced stack
method. While using the sleep transistor method in the existing research, PMOS was
coupled amidst Vdd and pull-up track. Nevertheless, NMOS was coupled amidst
ground and the pull-down track, at the same time in using the forced sleep method,
the connection applied was reversed.

III. PROPOSED METHODOLOGY

The current research focused on a novel sleepy technique along with AVS design is
proposed for low-power SRAM for reducing power consumption utilizing multi-
threshold CMOS circuit. Therefore, the SRAM us implemented in the research using
sleep transistors along with an additional leakage current feedback-transistor along
with MTCMOS primary structure which is implemented with an effective results in
terms of certain factors such as delay, area, and power. Therefore, the power
dissipation of dynamic and static power with the delay formed by sleep transistors are
substantially minimized when the sleep transistors development such as leakage-
current-feedback-transistor addition. As a next step, the power optimization has been
utilized in the research that performs to minimize package cost and to broaden batter
life to minimize SRAM circuit power dissipation (total). Finally, effective layout has
been developed and designed for SRAM circuit and COMS to minimize the area than
existing method.

3.1 Conventional SRAM:

The conventional SRAM used in the research helps to store binary data (1 bit) with
the SRAM [12] cell development, that utilizes a bi-stable-latching-circuitry.
Therefore, the memory cell used in the research defined with the 2 stable states that
denoted as 1 and 0. The 2 inverters circuits used in the research are integrated with the
cross coupled mutually. After this, 2 more NMOS transistors that defined as Cnt_N is
used in the process that control access, especially in the storage cell while executing
both write and read operations. Further, gate terminal control transistors are
interconnected to the operation signal (both read & write) WL - Word Line and six
MOSFET has been used to store cell data in memory with the typical SRAM. When
WL runs high, control-access transistors tend to be turned ON, especially for write
operation. Here, two complementary bit-lines including BL and BLB are used in the
operation with the cell connection with the data transcribed in memory cell. However,
SRAM-cell is inaccessible from bit lines (both), after this the SRAM settled in
standby mode with the unchanged stored value. Fig 1. represents the conventional
SRAM and Fig.2 defines the transient waveform.
 Figure 1 - Conventional SRAM

Figure 2 - Transient Waveform

3.2 Sleepy SRAM:

The sleepy SRAM executed with the circuit level methods that are performed earlier
for minimizing sub-threshold-leakage-power method, which is a sleepy method.
Therefore, the technology on sleepy SRAM utilizes sleep transistors in two places
which has different phases. The pull-up network presented amidst Vdd (ST_P) and
pull-down network presented amidst (ST_N) and Gnd. Here, the size of sleep
transistor is provided, which is relative to either pull-down or pull-up transistor
interconnected to the sleep transistor and the sleepy SRAM represented in Fig 3.

3.2.1 Active Mode

In the active mode state, the sleep signal represented as slp indicates the logic 1 with
complementary sleep signal represented as slpb and voltage level, represented logic 0
with voltage level. Further, the sleep transistors (ST_P, ST_N) represented in active
mode are ON. In the execution, when the power provided to the circuit with the sleep
transistors turned ON with the outcome node on VP defined at VDD with the VG
node presented at ground potential.

3.2.2 Sleep Mode

In the execution of sleep mode with the sleep signal represented as slp denoted with
logic-0 with complementary sleep signal and voltage level slpb that defined the logic-
1 voltage level. The sleep transistor denoted as ST_N and ST_P which are on the
OFF state. Because of this reason, the ground and supply are isolated from the logic-
circuit. Eventually, the static-power is set as virtually 0. Because of the effect on
stack, the off-resistance increased with the minimization of leakage current. The
technique generates state destruction with the floating voltage output, however, the
outcome procured with the floating values after sleep mode. Addition of sleep
transistors enhances the delay as well as area. In case of WL - word line raises high,
the control-access transistors are in the position of ON condition, especially in the
writing operation. Therefore, there are two complementary bit-lines that includes BL
and BLB are interconnected to the cell, and the data obtained are written in the proper
cell memory. In certain stage, SRAM cell is inaccessible from bit lines, while it was
in standby mode. Here, the stored value sustains in unchanged position with the
suppressed leakage current. These functioning are represented in Figure 3.

Figure 3 - Sleepy SRAM

3.3 Stack SRAM:
The stack method has been utilized in the research, which is considered as the another
low-power-reduction strategy. Normally, it segregates the present transistor into 2-
pieces and the transistors which was duplicated enable little reverse bias amidst the
source and gate during the 2 transistors was in the OFF position (turned off)
simultaneously. Because of the sub-threshold present dependency processed on gate
bias with enhanced area and delay, obtained with a significant current minimization.
Probably, certain execution resolves sleep method’s limits with the help of
maintaining state, however it enhances wake-up time. Eventually, the SRAM logic
state circuit is also can be retained. Figure 4 represents the Stack SRAM Inverter.
 Figure 4 - Stack SRAM Inverter

3.3.1 Sleepy Keeper SRAM

This approach holds the sleepy method circuits maintain similar and in-addition with
the P_MOS that defined as ST_P_K and N_MOS that defined as P_MOS transistors,
that are combined in parallel amidst vdd, pull-down and pull-up network with gnd.
Therefore, the effective way to utilize the CMOS transistors and PMOS, that are
interconnected to NMOS and Vdd connection to gnd. At the same time, the NMOS
transistors in the process are widely known for becoming ineffective at shifting Vdd
and the PMOs-transistors are widely known for becoming inefficient at shifting Gnd.
Certain floating voltage output has been surmount with the introduction of sleepy-
keeper method. In case of sleep-mode, the ST_P sleepy PMOS-transistors has to be
turned OFF with the N_MOS transistors (ST_N_K) interconnected amidst oull-up n/w
and VDD that holds specific data that represented as “1”, which is turned “ON”.
Likewise, in order to turn “ON” the P_MOS-transistor with the data holding ‘0’, that
are interconnected amidst the GND and pull-down N/W with the ST_N sleepy
NMOS-transistor to turn in the state “OFF”. Hence, this enables to keep the specific
data utilized in the process are safe, especially in sleep or standby mode. During the
WL drives high with the turned “ON” position of control- access transistors in writing
operation. The 2-complementary bit-lines such as BL and BLB in writing operation
are interconnected to the cell with the written message on data in memory cell. In
certain cases, SRAM cell is unobtainable from bit lines after the SRAM denoted in
standby-mode, in the certain cases, the stored data keeps unchanged. Figure 5 & 6
represents the Sleepy Keeper SRAM and transient waveform of it.
 Figure 5 - Sleepy Keeper SRAM

Figure 6 - Transient Waveform for sleepy Keeper SRAM

3.4 LECTOR SRAM

The main theme of using this method is to enhance the transistor number in the
“OFF” state in the gnd path and source. Therefore, the 2-leakage control transistors
that includes NMOS (LCT_N) and PMOS (LCT_P) interconnected amidst pull-down
and pull-up network. The PMOS gate terminal is interconnected to empty the
SRAM_N pull-down network, which is interconnected to empty the SRAM_P pull-up
network. The PMOS body terminal is interconnected to Vdd and NMOS is
interconnected to gnd. In the process of sleep-mode, the selection of NMOS transistor
or PMOS transistors(any one) should be in the “OFF” state or close to cut-off area.
Normally, this raises the path in resistance amidst gnd and Vdd and this method, also
minimized the leakage power. Hence, the LCT-transistor size is optimized in the
research for effective outcome and the enhancement of dynamic power with the
SRAM static-power is minimized.

Figure 7 - Process of LECTOR SRAM

During the WL drives high with the turned “ON” position of control- access
transistors in writing operation. The 2-complementary bit-lines such as BL and BLB
in writing operation are interconnected to the cell with the written message on data in
memory cell. In certain cases, SRAM cell is unobtainable from bit lines after the
SRAM denoted in standby-mode, in the certain cases, the stored data keeps
unchanged. Figure 7 indicates the process of LECTOR SRAM and Figure 8 & 9
represents the LECTOR SRAM and transient waveform of it.
 Figure 8 - LECTOR SRAM

Figure 9 - Transient Waveform for LECTOR SRAM

3.4.1 Sleepy Keeper LECTOR SRAM

This method are processed with a combination of LECTOR and sleepy keeper
method. During the process of sleep mode with the sleepy transistor represented as
ST_P and ST_N, which are turned OFF. Therefore, the parallel transistors includes
ST_P_K and ST_N_K that helps in holding the data in sleep-mode that reduces the
leakage power and the power minimization done by LCT_N and LCT_P due to the
multiple transistors in sleep mode and “OFF” state that enhances the resistance path
amidst gnd and vdd. During the WL drives high with the turned “ON” position of
control- access transistors in writing operation. The 2-complementary bit-lines such as
BL and BLB in writing operation are interconnected to the cell with the written
message on data in memory cell. In certain cases, SRAM cell is unobtainable from bit
lines after the SRAM denoted in standby-mode, in the certain cases, the stored data
keeps unchanged. Figure 10 & 11 represents the Sleepy Keeper LECTOR SRAM and
transient waveform of it.

Figure 10 - Sleepy Keeper LECTOR SRAM

Figure 11 - Transient Waveform for Sleepy Keeper LECTOR SRAM

3.4.2 Sleepy Stack SRAM

The sleepy stack SRAM [11] structure used in the research defines a novel technique
that minimizes leakage power with the integration of the sleep and stack transistor
structure. Here, the sleep stack method has 2-modes that define as first and second
mode in sleep condition. By using a forced stack method in the research, the previous
transistors used are split into 2-transistors with the benefits of stack structure from the
effect of stack. On the other hand, sleep transistors are placed parallel to a transistor
(stacked) to achieve state retention and low-power leakage by integrating these
techniques.

3.4.2.1 Active Mode

In the Active Mode, the sleep transistors are in “ON” state of the sleepy-stack
transistor. Therefore, this mode have the possibility to allow for circuit delay
minimization and all the presented sleep transistors are in “ON” state, in the mode of
active that outcomes in a rapid switching time. During the process of transistor, which
is parallel to sleep-transistors, the voltage of threshold value may be attained to high.

3.4.2.2 Sleep Mode

In the sleep-mode, the transistors are disabled with the preservation of precise logic
state using the structure of sleepy stack. It is noted that the sleep transistors are being
in a state of “OFF” that results in the effect of stack, that minimizes leakage power
consumption. Utilizing the sleep stack method enhances the speed in different ways
during the sleep transistor on the state of “ON”, and current flows via the circuit.
Hence, it is significant to switch circuit speed and it is enhanced with the minimized
delay.

Figure 12 - SRAM block using Sleepy Stack Technique

3.5 SNM (Static Noise Margin)

The memory cell stability has been acquired in the noise presence is evaluating
utilizing static noise margin. Therefore, it can be featured with the lowest-voltage that
should be enabled to store nodes for a certain state to flip. There is a possibility to
measure SNM by including the voltage transfer features of the 2-inverters that
imparted in the memory cell. Further, the inverters of memory cell’s are developed to
manage stable states with the output nodes includes the data that is stored in the
memory. Further, the node voltages initialized varying due to the noise presented at
the storage nodes that results in a cell stability decline of 55. The utilization of SNM
evaluates the maximum permit level of noise voltage with the capability of these
inverters management with their current state in the noise presence.

Moreover, employed an advanced low-power processors with DVS - dynamic-

voltage-scaling, which is considered as a feasible option, especially for power
reduction. With the help of the DVS, the operation of system is executed in the
highest frequency at the specific supply VDD and VDDnom that defines nominal
supply-voltage. Therefore, it can be operated in low power-mode at the scaled-VDD.
The significant utilization of supply voltage, which is minimized than its value at
nominal stage in the scheduled tasks completion with the minimum energy usage
amount.

In the modern digital VLSI-systems [14] along with the utilization of SRAM cells that
share a substantial portion, that have the similar supply-voltage with the other
processing units. Therefore, since the another digital components that are executed at
their lowest energy-point with the continuation of SRAM that operates with reliability
at specific scaled VDD. The different stages such as M1, M2, M3, M4, M5, M6, and
M7 that defines Conventional Technique, Sleepy SRAM, Stack SRAM, Sleepy
Keeper Technique, LECTOR technique, SK-LCT technique [13], and Sleepy Stack
SRAM.

IV. RESULTS AND DISCUSSION

The current research focused on a novel sleepy technique along with AVS design is
proposed for low-power SRAM for reducing power consumption utilizing multi-
threshold CMOS circuit. In the part of power dissipation, it has been analyzed with
the PDP - Power-Delay-Product - PDP, Leakage-current, Energy-delay-product -
EDP, and PEP - Power-Energy-Product, duty-cycle, and Clock Frequency. The layout
of the developed schematic for existing design has been mentioned in the research
previously. The parameter used in the research such as Area (μm), tpdf (ps), tpdr (ps),
propagation delay tpd (ps), Static power (W), dynamic power NMOS (W), dynamic
power PMOS (W), power (J), energy (J), PDP (J), PEP (J), EDP (J), clock frequency
(Hz), and duty cycle (Q) on power optimization methodologies such as M1 (con), M2
(sleepy), M3 (stack), M4 (SK), M5(LEC), M6(SK_LEC), and M7 (SS-hvt transistor)
represented in Table I. Here, in the current research, figure 13 represents the layout of
the conbf, the conventional technique.

Figure 13 - Layout of the Conventional Techniques

The layout of the LEC, SK, SK_LCT, sleepy layout, sleepy stack, and stack has been
denoted in the simulation process in figure 14 a, b, c, e, f, g, and h.
 (a) (b)

( c) (d)

( e) (f)

Figure 14 - (a) LEC (b) SK ( c) SK_LCT (d) Sleepy layout ( e) Sleepy Stack & (f) Stack
 Table I - Comparison of proposed work with the Existing Researches with the
Power Optimization Methodologies based on parameters

SL. Parameter Power Optimization Methodologies

No
M1 (con) M2(sleepy) M3(stack) M4(SK) M5(LEC) M6(SK_LE M7 (SS-hvt
C) transistor)
1 Area (μm) 3.3885 6.0648 6.3474 7.5187 6.8464 9.8757 8.502975

2 tpdf (ps) 10.882 5.874 7.454 5.536 5.642 4.436 5.833

3 tpdr (ps) 28.306 26.660 28.264 29.796 29.979 49.648 11.228

4 Propagation 19.594 16.267 17.859 17.666 17.810 27.042 8.530

Delay
tpd (ps)
5 Static Power 2.381E-6 76.148E-12 712.94E-12 43.481E-12 1.036E-6 42.503E-12 146.87E-12
(W)
Vin=0V
6 Static Power 1.427E-6 35.13E-12 421.81E-9 26.641E-12 5.633E-9 25.318E-12 9.154E-12
(W)
Vin=1V
7 Dynamic 13.559E-9 30.364E-12 2.417E-9 45.246E-12 459.26E-12 11.990E-12 40.709E-12
Power (W)
NMOS
8 Dynamic 133.29E-9 167.874E-12 29.047E-9 192.56E-12 47.855E-9 111.554E-12 206.854E-12
Power (W)
PMOS
9 Power 1.041E-6 363.603E-9 651.243E-9 363.378E-9 526.87E-9 220.169E-9 336.64E-9
(W)
10 Energy (J) 2.560E-15 601.591E-18 1.513E-15 450.56E-18 1.511E-15 445.682E-18 354.708E-18

11 PDP (J) 20.40E-18 5.914E-18 11.630E-18 6.419E-18 9.383E-18 5.953E-18 2.871E-18

12 PEP (J) 2.665E-21 218.741E-24 985.966E- 163.72E-24 796.17E-24 98.125E-24 119.409E-24

24
13 EDP (J) 50.17E-27 9.786E-27 27.038E-27 7.959E-27 26.914E-27 12.052E-27 3.025E-27

14 Clock 500M 500M 500M 500M 500M 500M 500M

Frequency
(Hz)
15 Duty Cycle 69.73 69.53 69.64 69.48 69.88 41.69 74.98
Q (%)

This research also compared with the parameters in the research such as Area (μm),
tpdf (ps), tpdr (ps), propagation delay tpd (ps), Static power (W), dynamic power
NMOS (W), dynamic power PMOS (W), power (J), energy (J), PDP (J), PEP (J),
EDP (J), clock frequency (Hz), and duty cycle (Q) on power optimization
methodologies such as M1 (con), M2 (sleepy), M3 (stack), M4 (SK), M5(LEC),
M6(SK_LEC), compared with M7 (SS-hvt transistor) represented in Table II with
increment and decrement values. The inc represents the proposed value increased
compared to existing method and dec represents the proposed method value decreases
compared to existing method.

4.2 Comparison table on Parameter with the value increment and value
decrement on M1, M2, M3, M4, M5, and M6 with M7

SL. Parameter M1 vs M2 vs M3 vs M4 vs M5 vs M6 vs
No M7 M7 M7 M7 M7 M7
1 Area (μm) 60.15 inc 28.67 inc 25.35 inc 11.58 inc 19.48 inc 16.14 dec
2 tpdf (ps) 86.56 dec 0.70 dec 27.79 dec 5.09 inc 3.27 inc 23.95 inc
3 tpdr (ps) 60.33 dec 57.88 dec 60.27 dec 62.32 dec 62.55 dec 77.38 dec
4 Propagation dec dec dec
Delay 56.47 47.56 dec 52.24 dec 51.72 52.11 68.46 dec
tpd (ps)
5 Static dec inc inc dec
Power (W) 99.99 48.15 79.40 dec 70.39 99.99 71.06 inc
Vin=0V
6 Static dec dec dec
100.0
Power (W) 100 73.94 dec dec 65.64 99.84 63.84 dec
0
Vin=1V
7 Power dec dec dec
67.66 7.42 dec 48.31 dec 7.94 36.11 52.90 inc
(W)
V. CONCLUSION
The current research focused on on a novel sleepy technique along with AVS design
is proposed for low-power SRAM for reducing power consumption utilizing multi-
threshold CMOS circuit. Here, a high speed and low-power SRAM architecture was
developed utilizing Conventional Technique, Sleepy SRAM, Stack SRAM, Sleepy
Keeper Technique, LECTOR technique, SK-LCT technique, and Sleepy Stack
SRAM.
The modern digital VLSI-systems along with the utilization of SRAM cells that share
a substantial portion, that have the similar supply-voltage with the other processing
units. Therefore, since the another digital components that are executed at their lowest
energy-point with the continuation of SRAM that operates with reliability at specific
scaled VDD.

REFERENCES
[1] Banu, S., & Gupta, S. (2022). Design and Leakage Power Optimization of 6T Static Random Access Memory Cell
Using Cadence Virtuoso. IJEER, 10(2), 341-346.

[2] Kumar, H., & Tomar, V. K. (2021). A review on performance evaluation of different low power SRAM cells in nano-
scale era. Wireless Personal Communications, 117(3), 1959-1984.

[3] Abbasian, E., & Gholipour, M. (2021). Single‐ended half‐select disturb‐free 11T static random access memory cell
for reliable and low power applications. International Journal of Circuit Theory and Applications, 49(4), 970-989.

[4] Tamilarasan, A. K., Edward, D. S., & Sarasam, A. S. T. (2021). KLECTOR: Design of Low Power Static Random-
Access Memory Architecture with reduced Leakage Current.

[5] Gavaskar, K., Narayanan, M. S., Nachammal, M. S., & Vignesh, K. (2021). Design and comparative analysis of
SRAM array using low leakage controlled transistor technique with improved delay. Journal of Ambient Intelligence and
Humanized Computing, 1-10.

[6] Satyaraj, D., & Bhanumathi, V. (2021). Efficient design of dual controlled stacked SRAM cell. Analog Integrated
Circuits and Signal Processing, 107, 369-376.

[7] Kuruvilla, J. (2022). A Stable Low Power Dissipating 9T SRAM For Implementation of 4x4 Memory Array with High
Frequency Analysis.

[8] Sachdeva, A., & Tomar, V. K. (2021). A soft-error resilient low power static random access memory cell. Analog
Integrated Circuits and Signal Processing, 109(1), 187-211.

[9] Chaudhary, D., Muppalla, V., & Mukheerjee, A. (2020, June). Design of low power stacked inverter based sram
cell with improved write ability. In 2020 IEEE Region 10 Symposium (TENSYMP) (pp. 925-928). IEEE.

[10] Agrawal, R., Kumar, A., AlQahtani, S. A., Maashi, M., Khalaf, O. I., & Aldhyani, T. H. (2022). Cache Memory
Design for Single Bit Architecture with Different Sense Amplifiers. Computers, Materials & Continua, 73(2).
[11] Kakkar, R., Goyal, S., Singh, J., Khosla, D., & Singh, S. (2022). IMPLEMENTATION AND MODELING OF LOW
POWER SLEEPY STACK SRAM. Journal of Advanced Sciences, 1(1).
[12] Chaudhary, A., & Rana, A. (2020, June). Ultra Low power SRAM Cell for High Speed Applications using 90nm
CMOS Technology. In 2020 8th International Conference on Reliability, Infocom Technologies and Optimization (Trends and
Future Directions)(ICRITO) (pp. 1107-1109). IEEE.
[13] Wu, M. (2010). On the application of graphics processor to wireless receiver design. Rice University.
[14] Pousia, S., & Murugan, K. (2021, March). VLSI Implementation Of High Speed Low Power Design Using Hybrid
Power Gating Technique. In IOP Conference Series: Materials Science and Engineering (Vol. 1084, No. 1, p. 012058). IOP
Publishing.