Ijest11-03-03-207 PDF
Ijest11-03-03-207 PDF
GEETHA PRIYA.M
Engineering Faculty, Department of Electronics and Communication Engineering
Amrita Vishwa Vidhyapeetham, Coimbatore, Tamilnadu, India
Abstract:
Under ideal conditions, Reversible logic gates produce zero power dissipation. So these can be used for
low power VLSI design. This paper proposes a new reversible parallel adder/subtractor using 4*4
Reversible DKG gate that can work singly as a reversible full adder and a full subtractor. A serial
adder/subtractor is also designed in this paper using Reversible Universal Shift registers and DKG gate.
This paper provides a threshold to build more complex arithmetic systems using reversible logic.
Keywords: Low power CMOS; Nanotechnology; quantum computing; Reversible logic circuits.
1. Introduction
Conventional Combinational logic circuits dissipate heat for every bit of information that is lost during their
operation. Due to this fact the information once lost cannot be recovered in any way. But the same circuit if it is
constructed using the reversible logic gates will allow the recovery of the information. It has been demonstrated
that circuits and systems constructed using irreversible logic will result in energy dissipation due to information
loss in 1960’s [15]. It is proved that the loss of one bit of information dissipates kT*log2 joules of heat energy,
where k is Boltzmann’s constant and T, the absolute temperature at which computation is performed [15]. Zero
power dissipation in logic circuits is possible only if a circuit is composed of reversible logic gates [2].
Reversible logic has applications in quantum computing, low power CMOS, nanotechnology, optical
computing, and DNA Computing.
A reversible logic gate is an n-input n-output logic device with one-to-one mapping. A gate is considered to be
reversible only if for each unique input there is a unique output assignment. This helps to determine the outputs
from the inputs and also the inputs can be uniquely recovered back from the outputs provided it is having same
number of inputs and outputs. In a reversible gate the outputs are 1’s for exactly half of the inputs. In an n-
output reversible gate the output vectors are permutation of the numbers 0 to 2n-1. The input that is added to an
nxk function to make it reversible is called constant input (CI). The output that is not necessary for further
computations is called garbage output (GO). Quantum cost (QC) refers to the cost of the circuit in terms of the
cost of a primitive gate. It is calculated knowing the number of primitive reversible logic gates (1*1 or 2*2)
required to realize the circuit. [13],[14]. These parameters have to be reduced while designing a reversible
circuit. Some of the major problems with reversible logic synthesis are that fanouts cannot be used, and also
feedback from gate outputs to inputs is not permitted. However fanout in reversible circuits is achieved using
additional gates. A reversible circuit should be designed using minimum number of reversible logic gates.
Reversible circuits have been proposed for different purposes like half adder, full adder, multipliers and
dividers. Several 4x4 reversible gates (e.g. HNG [8], MKG [9], TSG [10], and PFAG [12]) are designed to work
as full adder.
In this paper, a new reversible parallel adder/subtractor is proposed that uses 4*4 Reversible DKG gate which
can work singly as a reversible full adder and a full subtractor. Later a reversible serial adder/subtractor is
proposed.
The binary Full adder/subtractor is capable of handling one bit of each input along with a carry in/borrow in
generated as a carry out/ borrow from addition of previous lower order bit position. If two binary numbers each
consisting of n bits are to be added or subtracted, then n binary full adders/subtractors are to be cascaded. A
Parallel adder/subtractor is an interconnection of full adders/subtractors and inputs are simultaneously applied.
The carry/borrow generated at a stage is propagated to the next stage. Thus, delay is more in such type of
adders/subtractors.
A 4 bit reversible parallel adder/subtractor is implemented using the reversible DKG gate and shown in Fig 2a.
When the control input A=0, the circuit acts as a parallel adder, thus adding two binary numbers of 4 bits each
and produces a 4 bit sum and a carry out, as shown in Fig 2b. If the control input A=1, the circuit acts as a
parallel subtractor, thus subtracting two binary numbers of 4 bits each and produces a 4 bit difference and a
borrow out, as shown in Fig 2c. The same design can be extended to n bits.
A fanout circuit to duplicate the control input A/S consists of an FG [5] and a BVF gate [3] as depicted in the
Fig 3.
A comparison of the full adder/subtractor with the existing circuit [4] is depicted in Table 1. The design of
parallel adder/subtractor is compared with the existing design [4] in Table 2.
Table 1: A comparison of proposed reversible full adder/subtractor with the existing circuit
No of
Garbage Constant
Reversible
outputs inputs
Gates
DKG
1 2 1
Circuit
Existing
Circuit 1 2 3 2
[12]
Existing
Circuit 2 1 2 1
[12]
Table 2: A comparison of proposed reversible parallel adder/subtractor with the existing circuit
Number of
Garbage
Reversible
outputs
Gates
Proposed
4 8
Circuit
Existing
8 12
Circuit 1 [12]
Existing
4 8
Circuit 2 [12]
S1 S0 Operation
0 0 No Change (Qi)
0 1 Right Shift (Qi-1)
1 0 Left Shift (Qi+1)
1 1 Parallel load (Ii)
The basic cell [7] used in the Design of Reversible Universal Shift Register has a clocked DFF with set/reset
input. It makes use of SRK gates and BVF gate to act as a DFF. Asynchronous set/reset implies that the Flip
Flop is set or reset irrespective of the input data and clock. The Flip Flop will set, reset or operate as a normal
DFF depending on inputs X and Y. When the values of X and Y is 00 or 01, the Flip Flop will reset or set
respectively and when XY= 10, it works as a normal DFF.
7. Serial Adder/Subtractor
In the ripple carry adder the delay of carry transfers from one block to the other, results in increase of the total
delay of the circuit. Hence serial adder is one of the adders with a delay less than that of the parallel adder.
The 4-bit numbers to be added X (augend) and Y (addend) are stored in two Shift registers. When S1S0=01, and
the clock is pulsed, SIR is entered into MSB x3 (or y3) as the contents of the register are shifted right one
position. The X register serves as the accumulator, and after 4 shifts, the number X is replaced with the sum of
X and Y. The addend register is connected as a cyclic shift register, so after 4 shifts it is back to its original state
and the number Y is not lost. The serial adder consists of a full adder and a carry FF. At each clock time, one
pair of bits is added. When S1S0=01, the positive edge of clock shifts the sum bit into the accumulator, stores the
carry bit in the carry FF, and causes the addend register to rotate right.
Initially at time t0, when S1S0=11, the accumulator is loaded with augend X and the addend register is loaded
with Y. At this instant the signal E=1. During the next clock pulse, E becomes 0. The carry FF is cleared. Since
the Full Adder is a combinational network, the inputs x0 and y0 along with carry 0 are added after a small
propagation delay to give the sum sum0 and next carry c1. During the clock pulse t1, the generated sum bit is
shifted into the accumulator, and the remaining accumulator digits are shifted right one position. The same shift
pulse stores the generated carry in the carry FF and cycles the addend register right one position. The next pair
of bits x1 and y1 are now at the full adder input, and the adder generates the sum and the carry. The second clock
pulse shifts this sum bit sum1into the accumulator, stores carry c2 in the carry FF and cycles the addend register
right. In this way addition of the other 2 bits is performed and stored in accumulator. After 4 clock pulses, t4, the
sum of X and Y is in the accumulator, and the addend register is back to its original state.
S1 E.k , S 0 k .
The clock signal given to the carry Flip-Flop is clk FF clk .k and that to counter is
clkcounter clk .E
The shift register presented in reference [11] does not have a parallel loading capability. The Universal shift
register used in the proposed design can load or shift the data depending on select lines. The Basic cell used in
USR has the facility of asynchronous set/Reset as well. As the facilities provided in USR are more compared to
that in design [11], number of gates used in the design are more in the proposed design. Number of gates used,
Constant inputs and Garbage outputs of the proposed design are 76, 41 and 88 respectively.
Fig 10: Fan out circuits for clock and select lines
8. Conclusion
This paper proposes a reversible 4 bit parallel adder/subtractor using 4*4 DKG gates that works singly as a Full
adder/subtractor. A new design of Serial adder/Subtractor is proposed using Universal Shift registers and DKG
gate. The same design can work as both adder as well as subtractor. The number of reversible gates used and the
garbage outputs produced in the proposed design are optimal. The reversible adder/subtractor discussed in this
paper can be used to build higher order arithmetic circuits and complex designs of quantum computers. The
control circuit can be further elaborately designed in future.
References
[1] Ahsan Raja Chowdhury ;Hafiz Md. Hasan Babu; Md. Rafiqul Islam,; Syed Mostahed Ali Chowdhury,“Reversible Logic Synthesis for
Minimization of Full Adder Circuit”, Proceedings of the EuroMicro Symposium on Digital System Design(DSD’03), 3-5 September
2003, Belek- Antalya, Turkey,pp-50-54.Journal of Engineering Science and Technology Vol. 2(8), 2010, 3838-3845
[2] Bennett, C. H., “Logical Reversibility of Computation”, IBM J. Research and Development, pp.525-532, November 1973.
[3] Bhagyalakshmi, H.R. ;.Venkatesha, M.K, ‘An improved design of a multiplier using reversible logic gates’, International Journal of
Engineering Science and Technology Vol. 2(8), 2010, 3838-3845
[4] Elsayed, A; Layle, A; Maii T. Emam, Reversible Full AdderlSubtractor, 2010 Xlth International Workshop on Syrnbolic and
Nurnerical Methods, Modeling and Applications to Circuit Design (SM2ACD)
[5] Feynman R., 1985. Quantum mechanical computers, Optics News, 11: 11-20
[6] Krishnaveni,D, Geetha Priya,M, “Design of an Efficient Reversible 8x8 Wallace Tree Multiplier”, Submitted for Review to Circuits
and Systems-I Journal, IEEE
[7] Krishnaveni,D, Geetha Priya,M, “A Novel Design of Reversible Universal Shift Register with Reduced Delay and Quantum Cost”,
Submitted for Review to VLSI Systems Journal, IEEE
[8] Haghparast, M ; Navi, K, "Design of a novel reversible multiplier circuit using HNG gate in nanotechnology", Am. 1. Applied
Sciences, 5 (2008) 282
[9] Haghparast, M ; Navi, K ; Shams, M, "Novel reversible multiplier circuit in nanotechnology", World App!. Sci. 1., 3 (2008) 806.
[10] Himanshu Thapliyal ; Srinivas, M.B, ”Novel Reversible ‘TSG’ Gate and Its Application for Designing Components of Primitive
Reversible/Quantum ALU”
[11] Himanshu Thapliyal ; Srinivas, M.B,” An Extension to DNA Based Fredkin Gate Circuits: Design of Reversible Sequential Circuits
using Fredkin Gates”
[12] Islam, M.S. et a!.,"Low cost quantum realization of reversible multiplier circuit", Information technology journal,8(2009)208
[13] Kerntopf,P; Khan, M.H.A.; Perkowski, M.A, 2004. On universality of general reversible multiple valued logic gates, IEEE proceeding
of the 34th international symposium on multiple valued logic (ISMVL’04), pp: 68‐73.
[14] Kerntopf, P ; Perkowski, M, 2001.Reversible Logic. Invited tutorial, Proc.EURO‐MICRO, Sept 2001, Warsaw, Poland
[15] Landauer, R, “Irreversibility and Heat Generation in the Computing Process”, IBM J. Research and Development, vol. 3, pp. 183-
191,July 1961.