IRJET- Implementation of Low Power 32-Bit Carry-Look Ahead Adder using Adiabatic Logic

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 10 | Oct 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 1387
Implementation of Low Power 32-bit Carry-Look ahead Adder using
Adiabatic Logic
Muttappa Pujari1, Ashwini Desai2, Basavaraj C3
1PG Student, Dept. of Electronics and Communication Engineering.
2Associate Professor Dept. of Electronics and Communication Engineering.
3Assistant professor Dept. of Electronics and Communication Engineering.
123 KLE. Dr. M. S. Sheshgiri College of Engineering and Technology, Belagum-590008
Karnataka, India
------------------------------------------------------------------------***-------------------------------------------------------------------------
Abstract –Adiabatic logic is low power logic, in this
paper an adiabatic logic based 32-bit carry look ahead
adder is designed and implemented on the basis of efficient
charge recovery logic (ECRL). Power dissipation of the
proposed technique is compared with conventional CMOS
circuit. The adiabatic (ECRL) shows less average power
consumption than the conventional CMOS technique. The
adder lies in the critical path of all the arithmetic
operations so it plays a crucial role in determining the
overall system performance. A schematic and simulation of
proposed circuit is implemented in cadence virtuoso 6.1.5
using 45nm technology.
Key Words: ECRL adiabatic logic, Carry-Lookahead
Adder, energy recovery, average power, cadence
virtuoso, 45nm technology.
1. INTRODUCTION
In present scenario power dissipation is one of
the important parameter while designing any portable
devices or embedded devices. The devices are said to be
portable if it is minimized in terms of components like
transistor, register, capacitor etc, which is one of the
prime motto of present generation. According to Moore’s
law the transistors embedded on IC gets doubled for
every 18 months. Most of our customer demand long
battery life for portable devices. If the power dissipation
is larger in any devices, internally it heats the system.
Hence to overcome the above situation it requires heat
sink, which further increases the device size, therefore
use of heat sink is not an appropriate solution for
portable devices [2].
Hence VLSI designers have come up with new
technique called adiabatic logic. As compared to
conventional logic, adiabatic logic circuits are widely
used to reduce power consumption. Depending on the
technique used in adiabatic logic i.e. either partially or
fully, the energy is stored in load capacitor and can be
recaptured back to power supply [5].
1.1 Conventional Logic Switching
In conventional charging power dissipation is
mainly during switching activities. The source of the
pull-up network is given to the power supply VDD where
as the source of pull-down network is given to ground.
During steady state depending on the input signal one of
the transistors in pull-up or pull-down network is ON.
The energy required during discharging is provided by
equations (1) to (4).
Figure -1: Conventional CMOS logic gate
Consider an example of inverter with conventional
charging as shown in figure1 here PMOS transistor turns
ON when the input logic gate is low and hence if forms
the direct path between VDD to output and load capacitor
CL gets charged.
The total charge Q taken from supply voltage can be
given as
 1..............2
DDLVCE 
During charging the energy stored in load capacitance is
given by
 2..........
2
1 2
arg DDLech VCE 
The NMOS transistor turns ON when the supply voltage
is high. Hence it takes direct path from output to ground
and whatever the charge is stored in load capacitance, it

gets discharged through NMOS transistor. The energy
required during discharge of a capacitance is given by [2]
 3...........
2
1 2
arg DDLedisch VCE 
Therefore the total energy required for the conventional
logic circuit is sum of energy charge and discharge.
Therefore
edischechtotal EEE argarg 
22
.
2
1
.
2
1
DDLDDL VCVC 
 4............. 2
DDLtotal VCE 
From the above equation we can say that in conventional
logic we can reduce power either by scaling the supply
voltage or by varying the size of the capacitance.
Figure- 2: schematic of a static CMOS inverter
1.2 Adiabatic Logic Switching
Figure -3: Adiabatic logic energy charging
Figure 3 show the simple RC circuit to demonstrate
adiabatic logic charging. In conventional charging we
used constant voltage source where as in adiabatic
charging varying current source is used to charge the
capacitor. Initially at t=0 the load capacitance doesn’t
contain any charge in it. The voltage across capacitance
as a function of time‘t’ is given by,
     5...............1 TtI
C
tVC 
The current at current source can be given as,
     6.........
T
tV
CtI C
The energy dissipation in adiabatic circuit can be given
as,
   7.............
0
2
dttIRE
T
diss 
Therefore    8..........2
TtRIEdiss 
Use equation [6] in equation [8] we get,
 
 9..........
22
TT
TtVRC
E C
diss


 tTCV
RCE
C
diss 2
Therefore
   10............2
tTCV
RCEdiss 
From equation [10] we can clearly observe that,
If T=2RC then    11.........5.0 2
tCVEdiss 
If T>2RC then    12........5.0 2
tCVEdiss 
From the above equations we can say that by reducing
charge time ‘T’ or reducing resistance ‘R’ we can easily
reduce the power dissipation.
Basic rules to be followed for designig adibatic logic
circuit:
 Replace the PMOS and NMOS tranistor of pull-up
and pull-down network by T-gates.
 Use expanded pull-up and pull-down network to
drive the true and complimentry output.
 Two networks in the changed circuit, is utilized
by both chargeing and dischargeing load
capacitor.
2. EFFICIENT CHARGE RECOVERY LOGIC (ECRL)
Moor and Jeong proposed theory of Efficient Charge
Recovery Logic (ECRL). The schematic logic block of
ECRL is shown in figure 4. The structure of ECRL is
similar to that of CVSL (Cascade Voltage Switching
Logic). It consist of two cross-coupled PMOS transistor
M1 and M2 and two N-Functional blocks. The N-Function
block consists of pull-down networks. To recover the
energy back to power supply, an AC power supply (CLK)
is used. Here two separate outputs, out and out-bar are
generated so that clock can always drive capacitance
with constant load. Here output flowing due to cross
coupled connection of PMOS transistor is opted [5].

Figure- 4: basic schematic block diagram of ECRL
Some times ECRL faces problems for non-adiabatic loss
on pre-charge and recover phase. This problem arises
due to threshold voltage of PMOS transistor, i.e. when
clock reaches threshold voltage ‘Vtp’ of PMOS transistor,
it turns OFF. Hence the path between recovery and clock
gets disconnected, which results in incomplete recovery.
The loss incurred in ECRL is given by
 13..........2
2
tpL VCECRL 
From equation (13) we can say that non-adiabatic
energy loss is highly dependent on CL (Load
Capacitance) and independent of frequency.
Figure- 5: schematic diagram ECRL inverter
Let us consider ECRL inverter shown in figure 5. Here we
observe that, initially ‘in’ is high and ‘in-bar’ is at low
potential, when CLK rises from ground to VDD output
‘out’ remains in low state where as output ‘out-bar’
follows clock. When clock reaches VDD output, out and
‘out-bar’ holds values VDD and zero. Further it is used as
input for next stage, when clock falls from VDD to zero the
energy present in ‘out-bar’ is transformed to CLK. Hence
the charge gets recovered back.
3. CARRY-LOOKAHEAD ADDER
The principle behind adders is to perform addition
between given bits; there are many numbers of adders to
perform addition. Some of them are ripple carry adder,
carry look ahead adder, carry skip adder, parallel prefix
adder and so on. Ripple carry adder is very simple and
cost effective adder where the propagation delay is high,
because each time the MSB bit has to wait for
computation for carry from LSB bit. To overcome the
above problem related to propagation delay, carry look
ahead adder is implemented [2].
Carry look ahead adder consists of one bit full adder as
shown in figure7. A full adder takes two binary numbers
in addition to carry bit. The output is the sum and
another carry out. The carry propagation and generation
is used in full adder. The output propagation of carry is
evaluated by XORing two input Ai and Bi and output of
carry generation is taken by ANDing.
Figure -6: block diagram 32-bit carry look ahead adder
Figure -7: carry propagation and carry generation
The output expression for sum and carry bit can be given
as
 14...................BiAiSi 
  )15.......(..........1 BiAiBiAiCinCi 
The intermediate outputs, propagate and generate Pi and
Gi can be given as
 16............BiAiPi 
 17.............BiAiGi 
Substitute equation (16) in (14)
 18.................CinPiSi 
Substitute equation (17) in (15)
 19.............1 GiPiCinCi 

4. DESIGN AND SIMULLATION
The implementation of this project work is done using
cadence virtuoso 6.1.5 and simulation is done using ADE-
L Spectre (Analog Design Environment-L). Here all the
implementations are performed using 45nm technology.
Adiabatic logic ECRL (Efficient Charge Recovery Logic)
are designed and compared with conventional CMOS
logic. Inverter, basic gates AND, OR, EX-OR and 32 bit
CLA adder are implemented using adiabatic ECRL logic
in 45nm technology and the functionality is verified by
doing necessary simulations as shown in figures 8 to 21.
4.1 Implementation of Inverter Using CMOS
Figure -8: Schematic of CMOS inverter
Figure -9: Transient analysis of CMOS inverter
4.2 Implementation of Inverter Using ECRL
Adiabatic logic
Figure -10: Schematic of ECRL inverter
Figure -11: Transient analysis of ECRL inverter
4.3 Implementation of AND gate Using ECRL
Adiabatic logic
Figure -12:Schematic of ECRL AND gate
Figure -13: Transient analysis of ECRL AND agte
4.4 Implementation of OR gate Using ECRL
Adiabatic logic
Figure -14:Schematic of ECRL OR gate

Figure -15:Transient analysis of ECRL OR agte
4.5 Implementation of EX-OR gate Using ECRL
Adiabatic logic
Figure -16: Schematic of ECRL EX- OR gate
Figure- 17: Transient analysis of ECRL EX-OR gate
4.6 Implementation of 16-Bit CLA Using ECRL
Adiabatic Logic
Figure -18: Schematic of ECRL 16-bit CLA.
Figure- 19: Transient analysis of ECRL 16-bit CLA
4.7 Implementation of 32-Bit CLA Using ECRL
Adiabatic Logic
Figure -20: Schematic of ECRL 32-bit CLA.

Figure -21: Transient analysis of ECRL 32-bit CLA
5. POWER DISSIPATION ANALYSIS
5.1Power Dissipation Analysis for Basic Gates
The table 1 demonstrates the average power consumed
by logic gates namely Inverter, AND, OR and EX-OR
under conventional and adiabatic logic (ECRL). From the
above observations we can say that adiabatic logic
consumes less power than that of conventional logic.
Table-1: comparison between ECRL and CMOS
Technology ECRL(nw) CMOS(µw)
Inverter 53.83 3.073
AND 71.06 4.306
OR 63.74 5.854
EX-OR 159.5 6.694

5.2 Power Dissipation Analysis for Adders
The table 2 shows the average power consumption
calculated for ECRL and static CMOS for 4bit, 8bit, and
16bit and 32bit carry look ahead adder. From the
computation we say that CLA using adiabatic ECRL have
low power dissipation than static CMOS logic.
Table-2: comparison between ECRL and CMOS Adders
Technology 4bit
(µw)
8-bit
(µw)
16-bit
(µw)
32-bit
(µw)
ECRL 1.966 3.23 10.56 36.76
CMOS 35.2 55.34 100.6 230.03
6. CONCLUSION
The implementation of 32-bit carry look ahead adder is
done for both conventional CMOS logic and adiabatic
ECRL logic. From all the above observations made and
from simulated results, we can conclude that the
adiabatic adders consume less power than that of
conventional adders. Adiabatic logic circuits consume
less power when compared to conventional CMOS logic
circuits and also there is less wastage of power, hence
there is more demand for adiabatic logic circuits.
REFERENCES
[1] AainaNandal and Dinesh Kumar “A Study on
Adiabatic Logic Circuits for Low Power Applications”
Special Issue – 2017.
[2] Ashwini A. Pote and Prof.Ashwini Desai “Design and
Implementation of Low Power 16-bit Carry-
lookahead Adder Using Adiabatic Logic”.
[3] YibinYe“Ultra Low Energy Computing Using
Adiabatic Switching Principle” Purdue University,
Purdue e-Pubs ECE Technical Reports, 3-1-1995.
[4] VibhuBindal “Adiabatic Logic Circuit Design” ISSN
2348 – 7968
[5] PoonamKadam “Comparative Analysis of Adiabatic
Logic Techniques”
[6] Durgesh Patel, Dr. S. R. P. Sinha and Meenakshi Shree
“Adiabatic Logic Circuits for Low Power VLSI
Applications” Index Copernicus Value (2013): 6.14 |
Impact Factor (2015): 6.391
[7] Aneesha John and Charishma “Design of Low Power
Energy Efficient CMOS Circuits with Adiabatic Logic”
Vol. 5, Special Issue 9, May 2016.
[8] Simranjeet Singh Puaar “POWER EVALUATION OF
ADIABATIC LOGIC CIRCUITS IN 45NM
TECHNOLOGY” IJECET, Volume 5, Issue 12,
December (2014), pp. 230-237.
[9] Amit Singh Gaur and JyotiBudakoti. “Energy efficient
advanced low power CMOS design to reduce power
consumption in deep submicron technologies in
CMOS circuit for VLSI design”, International Journal
of Advanced Research in Computer and
Communication Engineering Vol. 3, Issue 6, June
2014.
[10] Gojman, B,(August 8,2014).Adiabatic Logic.
[11] R.Sivakumar and D.jothhi, “Recent trends in low
power design”, International journal of computer
and electrical engineering”, November2014

IRJET- Implementation of Low Power 32-Bit Carry-Look Ahead Adder using Adiabatic Logic

More Related Content

What's hot (20)

Similar to IRJET- Implementation of Low Power 32-Bit Carry-Look Ahead Adder using Adiabatic Logic (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Implementation of Low Power 32-Bit Carry-Look Ahead Adder using Adiabatic Logic