Design of subthreshold dml logic gates with power gating techniques

IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
__________________________________________________________________________________________
Volume: 03 Issue: 04 | Apr-2014, Available @ http://www.ijret.org 174
DESIGN OF SUBTHRESHOLD DML LOGIC GATES WITH POWER
GATING TECHNIQUES
Lakshmisree P V1
, Raghu M C2
1
M.Tech (VLSI Design), Department of ECE, NCERC, University of Calicut, India
2
Assistant Professor, Department of ECE, NCERC, University of Calicut, India
Abstract
Sub-threshold circuit design is one of the promising methods for low power to ultra-low power applications. Circuits which operate in
the sub-threshold region use a supply voltage that is close to or less than the threshold voltages of the transistors, so that there is a
significant reduction in both dynamic and static power consumption. The low-power dual mode logic (DML) family is a logic family
designed to operate in the sub-threshold region. The proposed logic family can be switched between static and dynamic modes of
operation according to system requirements. The ability of DML circuits to operate in both the static and dynamic modes gives the
opportunity to create efficient logic circuits which balance power consumption and operating frequency (speed of the circuit)
requirements. In the static mode of operation, the dual mode logic gates has very low-power dissipation with moderate performance,
and in the dynamic mode of operation they have higher performance, at the price of increased power dissipation
Keywords: Dual mode logic (DML), Subthreshold
-----------------------------------------------------------------------***--------------------------------------------------------------------
1. INTRODUCTION
With technological advancements and the mobile applications
expansion, power consumption has become a primary focus of
attention in VLSI digital design. Hence Digital sub-threshold
circuit design is one of the main focus areas for low power to
ultra-low power applications. The supply voltage that is applied
to the circuits operating in the sub-threshold region is very
close to or less than the threshold voltages of the transistors, so
it allows a significant reduction of both dynamic and static
power.
One of the most common logic family used for sub-threshold
operation currently is the Complementary Metal Oxide
Semiconductor (CMOS) logic family. Ultra low Voltage (ULV)
operation originally introduced in 1972 was originally used for
low throughput applications like wrist watches, sensors and
biomedical devices. It gives low to moderate performance and
maintains low- power dissipation.
The dual mode logic NAND, NOR and INVERTER is designed
to operate in the sub threshold region. The proposed logic gates
can be operated in two modes: static CMOS-like mode and
dynamic CMOS-like mode as shown in Fig. 1. In the static
mode of operation, the DML gates have very low power
dissipation with moderate performance. When the DML gate is
in the dynamic functional mode, they have much higher
performance, at the cost of increased power dissipation. This
particular feature of the Dual Mode Logic provides the option
to control system performance on-the-fly and hence support
applications in which a flexible workload is required.
Fig-1: Static and Dynamic Modes of Operation
The rest of this paper is constructed as follows. Section II
provides an overview of the basic DML logic gate architecture
and its method of operation. Section III describes the designing
of the Nand, Nor, Not DML gates. Section IV describes the
Dual Mode Logic with Power Gating techniques. Section V
concludes this paper.
2. BASIC DUAL MODE LOGIC ARCHITECTURE
The basic DML logic gate designed to operate in either static
mode of operation or dynamic mode of operation consists:
A static gate having one or more logic inputs, a single logic
output and a switching element that is associated with the static
gate as shown in Fig. 2. The switching element comprises of an

__________________________________________________________________________________________
input that is connected to a constant voltage, and another input
for providing a signal used for mode selection, an output that is
connected to a logic output of the static gate. The switching
element can be configured to operate in either of the two modes
by:
i) disconnecting the static gate output from both the input that is
connected to a constant voltage, and the other input for
providing a signal used for mode selection, when the mode
selection signal applies a constant voltage to the input used for
providing mode selection signal, thereby selecting static mode
of operation
ii) Connecting the static gate output to both the input that is
connected to a constant voltage, and the other input for
providing a signal used for mode selection, when the mode
selection signal applies a dynamic clock signal to the input used
for providing mode selection signal, thereby to select dynamic
mode operation.
Switching the Dual mode logic gates between the two
functional modes, static and dynamic, is performed by applying
either a constant voltage or a dynamic clock signal at the mode
selection input of the switching element.
Fig-2: Basic DML Logic
When static functional mode is selected an appropriate constant
voltage (high or low as required by static gate topology) is
applied to the mode selection input of the switching element.
The applied constant voltage causes the switching element to
disconnect the static gate output from the constant voltage
applied, thus enabling static mode of operation. During
dynamic mode of operation, the switching element input a
dynamic clock signal which has pre-charge and evaluate
phases, which will periodically connect the static gate output to
the constant voltage level, thus enabling dynamic mode of
operation.
In the DML circuit designed, the static gate is a CMOS gate and
switching element is implemented by either a PMOS or an
NMOS transistor. In order to operate the Dual mode logic gate
in the dynamic functional mode, the clock (Clk) signal applied
to the switching element has two distinct phases: the precharge
and evaluation phases. During the precharge phase, the output
of the DML gate is charged to high/low, based on the topology
of the DML gate. During the following evaluation phase, the
DML gate‟s output is evaluated according to the values at the
DML gate inputs. The DML topologies, marked Type A and
Type B, are illustrated in Fig. 3. Type A topology has an added
p-MOS transistor that precharges the output to a logical “1”
during the precharge phase. Type B DML topology has an
added n-MOS that precharges the gate output to a logical “0”
during the precharge phase.
Fig-3: Basic DML Type A topology and Type B topology.
Switching the DML gates to operate in CMOS-like static mode
of operation is done by either by fixing the global Clock (Clk)
signal to logic high for Type A topology or by fixing the global
Clk signal to logic low for Type B topology.
3. DML NAND, NOR, INVERTER GATES DESIGN
Conventional Nand logic gate design is done using Tanner S-
Edit EDA tool and its power and performance are found. Also
Dual Mode Logic Nand gate Type A and Type B topologies
designed and their power consumption and performance were
analyzed for static and dynamic mode of operations. The truth
table of the 2-input NAND gate is given in Table 1.
Table-1: Truth Table of 2-input NAND
A B Q
0 0 1
0 1 1
1 0 1
1 1 0
To design a DML gate, the methodology that is to be followed
is to place the precharge transistor in parallel to the stacked
transistors of the basic CMOS gate. Then, the evaluation of the
logic will be performed with the parallel transistors which will
make the evaluation process faster.

__________________________________________________________________________________________
Fig-4: Schematic of type A Static nand
In the DML Type-A Static NAND topology, the switching
element is a pMOS transistor connected parallel to the Pull-up
network as shown in Fig. 4. The input to the switching element
is a constant high voltage to make it OFF. The only difference
when designing DML Type-A Dynamic NAND topology, is
that the input to the switching element is a clock signal having
pre-charge and evaluate phase for dynamic mode of operation.
In the DML Type-B Static NAND topology, the switching
element is an NMOS transistor connected parallel to the Pull-
down network. The input to the switching element is a constant
low voltage to make it OFF. The only difference when
designing DML Type-B Dynamic NAND topology shown in
Fig. 5 is that the input to the switching element is a clock signal
having pre-charge and evaluate phase for dynamic mode of
operation.
Fig-5: Schematic of type B Dynamic nand
Conventional NOR logic gate design is done using Tanner S-
Edit EDA tool and its power and performance are found. Also
Dual Mode Logic Nand gate Type A and Type B topologies
designed and their power consumption and performance were
analyzed for static and dynamic mode of operations. The truth
table of the 2-input NOR gate is given in Table 2.
Table-2: Truth Table of 2-input NOR
A B Q
0 0 1
0 1 0
1 0 0
1 1 0
In the DML Type-A Static NOR topology, the switching
element is a PMOS transistor connected parallel to the Pull-up
network which is a series connection of 2 PMOS transistors as
shown in Fig. 6. The input to the switching element is a
constant high voltage to make it OFF. The only difference when
designing DML Type-A Dynamic NOR topology, is that the
input to the switching element is a clock signal having pre-
charge and evaluate phase for dynamic mode of operation.
Fig-6: Schematic of type A Static nor
In the DML Type-B Static NOR topology, the switching
down network which is a parallel connection of 2 NMOS
transistors. The input to the switching element is a constant low
voltage to make it OFF. The only difference when designing
DML Type-B Dynamic NOR topology is that the input to the
switching element is a clock signal having pre-charge and
evaluate phase for dynamic mode of operation as shown in Fig.
7.

__________________________________________________________________________________________
Fig-7: Schematic of type B Dynamic nor
Conventional inverter gate design is done using Tanner S-Edit
EDA tool and its power and performance are found. Also Dual
Mode Logic Nand gate Type A and Type B topologies designed
and their power consumption and performance were analyzed
for static and dynamic mode of operations. The truth table of
the NOT gate is given in Table 3.
Table-3: Truth Table of NOT
A Q
0 1
1 0
In the DML Type-A Static NOT topology, the switching
element is a PMOS transistor connected parallel to the Pull-up
network as shown in Fig. 8. The input to the switching element
is a constant high voltage to make it OFF. The only difference
when designing DML Type-A Dynamic NOT topology is that
the input to the switching element is a clock signal having pre-
charge and evaluate phase for dynamic mode of operation.
Fig-8: Schematic of type A Static not
In the DML Type-B Static NOT topology, the switching
down network. The input to the switching element is a constant
low voltage to make it OFF. The only difference when
designing DML Type-B Dynamic NOT topology is that the
input to the switching element is a clock signal having pre-
charge and evaluate phase for dynamic mode of operation as
shown in Fig. 9.
Fig-9: Schematic of type B Dynamic not
4. DML WITH POWER GATING TECHNIQUES
Power Gating is an effective implementation that is used in
Low Power Designs. While Clock Gating saves the dynamic
power of a circuit, Power Gating saves the leakage power. As
the technology moves from micron technology to sub-micron
technology, the leakage power dissipation dominates the
dynamic power dissipation.
Fig-10: Power Gating structure
The power gated sleep circuit has two modes of operations:
1. Active mode
2. Sleep mode

__________________________________________________________________________________________
In active mode, the sleep Signal SLEEP of the transistor is held
at logic' 0 'and SLEEP` at logic „1‟ hence both the sleep
transistors remain ON. In this case both transistors offer very
low resistance and virtual ground (VGND) node potential is
pulled down to the ground potential, making the logic
difference between the logic circuitry approximately equal to
the supply voltage.
In sleep mode, the sleep Signal SLEEP of the transistor is held
at logic' 1 'and SLEEP` at logic „0‟ hence both the sleep
transistors are turned OFF and the logic part is dis-connected
from the supply and ground leading to very less power
consumption during sleep mode. Power gating is incorporated
in to the DML architecture as shown in Fig. 11.
Fig-11: DML gate with Power Gating
Different methods of power gating like the sleep method,
sleepy stack and dual sleep approaches added to the existing
Dual mode logic NAND, NOR, NOT gates.
The Sleep method is the basic power gating method. The sleep
transistors isolate the logic networks and the sleep transistor
technique or the sleep method dramatically reduces leakage
power during sleep mode. Fig.12 shows Type A Dynamic Dual
Mode Logic Nand with sleep power gating technique.
Fig-12: Type A dynamic NAND with Sleep Power Gating
Technique
The sleepy stack approach merges the sleep and stack
approaches. The sleepy stack technique splits the existing
transistors into two half Size transistors like the stack approach.
The activity of the sleep transistors in the sleepy stack method
is same as the activity of the sleep transistors in the sleep
method. The sleep transistors are turned on during the active
mode and they are turned off during the sleep mode. Fig. 13
shows Conventional NOT gate with Sleepy Stack Power Gating
Technique.
Fig-13: Conventional NOT gate with Sleepy Stack Power
Gating Technique
The Dual sleep approach has the advantage of using the two
extra pull- up and two extra pull-down transistors in sleep mode
either in OFF state or in ON state. In normal mode when S=1
the pull down NMOS transistor is in ON state and in the pull-up
network the PMOS sleep transistor is in ON state since S‟=0.
During sleep mode state S is forced to 0 and hence the pull-
down NMOS transistor is in OFF state and PMOS transistor is
in ON state and in the pull-up network, PMOS sleep transistor
is OFF while NMOS sleep transistor is ON. So in sleep mode
state a PMOS is in series with an NMOS both in pull-up
network and pull-down network which reduces the power
dissipation. Fig.14 shows Type B Static NOR with Dual Sleep
Power Gating Technique
Fig-14: Type B Static NOR with Dual Sleep Power Gating
Technique

__________________________________________________________________________________________
The comparison of the NAND, NOR, NOT Dual mode logic
gates in terms of power consumption is shown in Table 4.
Table- 4: Comparison between NAND, NOR, NOT DML gates
NAND NOR NOT
CONVENTIONAL 6.24E-09W 5.11E-09W 3.35E-09W
TYPE A STATIC 5.32E-09W 4.91E-09W 2.37E-09W
TYPE A
DYNAMIC
9.44E-09W 5.54E-09W 1.15E-08W
TYPE B STATIC 6.07E-09W 5.08E-09W 3.27E-09W
TYPE B
DYNAMIC
7.02E-09W 5.15E-09W 3.26E-09W
The NAND Dual mode logic gate is compared in terms of
power consumption for different power gating techniques like
sleep method, sleepy stack method and dual sleep method in
Table 5.
Table-5: Comparison between NAND DML gates with
different power gating techniques
NAND
SLEEP
POWER
GATING
NAND
SLEEPY
STACK
POWER
GATING
NAND
DUAL
SLEEP
POWER
GATING
CONVENTIONAL 1.71E-09W 2.06E-
09W
1.01E-08W
TYPE A STATIC 1.36E-09W 7.42E-
10W
7.39E-09W
TYPE A
DYNAMIC
2.73E-09W 3.29E-
09W
1.34E-08W
TYPE B STATIC 1.55E-09W 8.32E-
10W
9.87E-09W
TYPE B
DYNAMIC
2.13E-09W 8.09E-
09W
1.14E-08W
The NOR Dual mode logic gate is compared in terms of power
consumption for different power gating techniques in Table 6.
Also The NOT Dual mode logic gate is compared in terms of
power consumption for different power gating techniques in
Table 7.
Table-6: Comparison between NOR DML gates with different
power gating techniques
NOR
SLEEP
POWER
GATING
NOR
SLEEPY
STACK
POWER
GATING
NOR
DUAL
SLEEP
POWER
GATING
TYPE A
DYNAMIC
7.72E-09W 2.45E-08W 5.47E-09W
TYPE B
DYNAMIC
3.04E-09W 3.17E-09W 2.91E-09W
Table- 7: Comparison between NOT DML gates with different
power gating techniques
NOT
SLEEP
POWER
GATING
NOT
SLEEPY
STACK
POWER
GATING
NOT
DUAL
SLEEP
POWER
GATING
TYPE A
DYNAMIC
5.57E-09W 1.93E-09W 1.75E-09W
TYPE B
DYNAMIC
1.63E-09W 2.86E-10W 1.96E-09W
5. CONCLUSIONS
The result obtained leads to the conclusion that while operating
in the dynamic mode, sub threshold DML gates achieve an
improvement in speed compared to a standard CMOS, while
dissipating more power and in the static mode, a reduction of
power dissipation is achieved, at the expense of a decrement in
performance. The different methods of power gating applied to
the DML logic have reduced the power dissipation further.
REFERENCES
[1] Asaf Kaizerman, Sagi Fisher, and Alexander Fish,
“Subthreshold Dual Mode Logic,” IEEE
TRANSACTIONS ON VERY LARGE SCALE
INTEGRATION (VLSI) SYSTEMS, VOL. 21, NO. 5,
MAY 2013
[2] M. Alioto, “Ultralow power VLSI circuit design
demystified and explained: A tutorial,” IEEE Trans.
Circuits Syst. I, vol. 59, no. 1, pp.3–29, Jan. 2012.
[3] A.P.Chandrakasan, S.Sheng and R.W.Brodersen,
“Low-power CMOS digital, “Solid-State Circuits,
IEEE Journal of, vol.27, pp.473-484, 2002.
[4] D.Bol, R. Ambroise, D. Flandre, and J. D. Legat,
“Analysis and minimization of practical energy in 45
nm subthreshold logic circuits,” in Proc. IEEE Int.
Conf. Comput. Design, Oct. 2008, pp. 294–300.
[5] N,Verma, J. Kwong and A.P. Chandrakasan,
"Nanometer MOSFET variation in minimum energy
subthreshold circuits," Electron Devices, IEEE
Transactions on, vol. 55, pp. 163-174, 2008.
[6] H.Soeleman and K. Roy, "Ultra-low power digital
subthreshold logic circuits," in Proceedings of the

__________________________________________________________________________________________
1999 international symposium on Low power
electronics and design, pp. 94-96, 1999.
[7] J. Kao, S. Narendra and A. Chandrakasan,
"Subthreshold leakage modeling and reduction
techniques," in Proceedings of the 2002 IEEE/ACM
international conference on Computer-aided design,
pp. 141-148, 2002.
[8] B. H. Calhoun, A. Wang, and A. Chandrakasan,
“Modeling and sizing for minimum energy operation
in subthreshold circuits,” IEEE J. Solid-State Circuits,
vol. 40, no. 9, pp. 1778–1786, Sep. 2005.
[9] D. Markovic, C. C. Wang, L. P. Alarcon, and J. M.
Rabaey, “Ultralowpower design in near-threshold
region,” Proc. IEEE, vol. 98, no. 2, pp.237–252, Feb.
2010.
[10] B. Zhai, S. Hanson, D. Blaauw, and D. Sylvester,
“Analysis and mitigation of variability in subthreshold
design,” in Proc. Int. Symp. Low Power Electron.
Design, Aug. 2005, pp. 20–25.
[11] Chhavi Saxena, Member, IEEE, Manisha Pattanaik,
Student Member IEEE and R.K. Tiwari, “Enhanced
Power Gating Schemes for Low Leakage Low Ground
Bounce Noise in Deep Submicron Circuits”
[12] R. Bhanuprakash, Manisha Pattanaik, S. S. Rajput and
Kaushik Mazumdar, “Analysis and Reduction of
Ground Bounce Noise and Leakage Current During
Mode Transition of Stacking Power Gating Logic
Circuits,” TENCON 2009
[13] http://vlsi-soc.blogspot.in/2012/08/power-gating.html
BIOGRAPHIES
P. V. Lakshmisree, student, is currently
pursuing her M.tech VLSI Design in
department of ECE in Nehru College of
Research and Engineering, Pampady, Trissur.
She has completed B.Tech in Electronics and
Communication Engineering in
SOE,CUSAT,Cochin. Her research areas are VLSI, Low Power
Design.
Mr. M. C. Raghu, Assistant Professor,
Department of ECE, Nehru College of
Research and Engineering, Pampady, Trissur.
He has completed ME VLSI Design in KCG
college of Technology, Chennai under ANNA
University. His research areas are VLSI,
Digital IC Design.

Design of subthreshold dml logic gates with power gating techniques

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Design of subthreshold dml logic gates with power gating techniques (20)

More from eSAT Publishing House (20)

Recently uploaded (20)

Design of subthreshold dml logic gates with power gating techniques