See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/272864857

Modified Booth Multiplier using Wallace Structure and Efficient Carry Select
Adder

Article  in  International Journal of Computer Applications · April 2013

DOI: 10.5120/11643-7130

CITATIONS READS

5 2,377

2 authors:

Neeta Sharma Kothari Ravi Sindal

Indian Statistical Institute, India, Bangalore Devi Ahilya Vishwavidyalaya, Indore (DAVV)
4 PUBLICATIONS   10 CITATIONS    25 PUBLICATIONS   95 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Analysis of Handover of UE in EUTRAN View project

Performance Analysis of Link Adaptation in EUTRAN View project

All content following this page was uploaded by Ravi Sindal on 23 June 2017.

The user has requested enhancement of the downloaded file.

 International Journal of Computer Applications (0975 – 8887)
Volume 68– No.13, April 2013

Modified Booth Multiplier using Wallace Structure and

Efficient Carry Select Adder
Neeta Sharma Ravi Sindal, PhD.
ME (DI) Associate Professor
Electronics & Instrumentation Department Electronics & Instrumentation Department
IET DAVV, Indore IET DAVV, Indore

ABSTRACT Adders arranged in a Wallace tree structure. In the last stage,

The multiplier forms the core of systems such as FIR filters, the two-row outputs of the tree are added using any high-
speed adder .Ripple Carry Adder (RCAs) have the most
Digital Signal Processors and Microprocessors etc. This paper
compact design among all types of adders. Carry Look Ahead
presents a model of two different 16X16 bit multipliers. First Adders (CLAs) are the fastest adders, but they are worst from
is Radix-4 Multiplier with SQRT CSLA and Second one is the area point of view. Carry select adders have been
Radix -4 multiplier with Modified SQRT CSLA. Modified considered as a compromise solution between RCAs & CLAs
Booth Algorithm is used for Partial Products Generation. because they offer a good tradeoff between the compact area
Wallace Tree Structure is used to accumulate partial Products. of RCAs and the short delays of CLAs. CSLAs designs can be
SQRT Carry Select Adder is used for addition of last two further modified to get better performance. CSLA with non
uniform block size & CSLA with non uniform block size &
rows. SQRT Carry Select Adder uses non uniform block size
Binary to Excess-1 Converter (BEC) will achieve lower delay
& Modified SQRT Carry Select Adder uses non-uniform as compared to regular CSLA.
block size with Binary to excess one Converter .Both the The paper is organized as follows: section 2 describes Partial
adders proved to be fast as compared to regular carry select Product Generation, section 3 discusses Carry save adders
adder. An efficient VHDL code has been written & used in Wallace tree structures, section 4 discusses Final
successfully synthesized & Simulated using Xilinx ISE Adder that is Efficient Carry Select Adders, section 5 shows
simulation results and comparison and section 6 shows the
12.1.Simulation results shows that the delay of both the
Conclusion, section 7 gives the references.
multipliers is reduced & the number of logic levels is also
reduced with slight increase in number of slices & LUTS as
compared to multiplier that uses regular carry select adder.

Keywords
VHDL, Booth Encoder, Carry Save Adder, Wallace tree,
SQRT Carry Select Adder, BEC.

1. INTRODUCTION
Many application systems based on DSP require extremely
fast processing of a huge amount of digital data. The
multiplier is an essential element of the digital signal
processing such as filtering and convolution. The multiplier is
also an important element in microprocessor. The demand of
fast processors is increasing for high-speed data processing
[6]. Since the multiplier requires the longest delay among the
basic operational blocks in digital system, the critical path is
determined more by the multiplier [4].
Any multiplier can be divided into three stages: Partial
products generation stage, partial products addition stage, and
the final addition stage. The speed of multiplication can be
increased by reducing the number of partial products. Many
high-performance algorithms and architectures have been
proposed to accelerate multiplication. Various multiplication
algorithms such as Booth, Modified Booth, Braun, and Fig.1.Block Diagram of Wallace Booth Multiplier
Baugh-Wooley have been proposed [9]. The modified Booth
algorithm reduces the number of partial products to be 2. BOOTH ENCODER
generated and is known as the fastest multiplication
Booths algorithm involves encoding of multiplier bits and
algorithm. Wallace Tree Carry Save Adder structures have
partial product generation. Different modified booths
been used to sum the partial products in reduced time. Our
algorithms have been proposed according to how many
goal is to reduce computation time by using Booth's algorithm
number of bits are used to encode multiplier. Modified booth
for multiplication and to reduce chip area by using Carry Save
algorithm reduces the number of partial products. Radix 2,

39
 International Journal of Computer Applications (0975 – 8887)
Volume 68– No.13, April 2013

radix 4, radix 8, radix 16, radix 32 are the different modified

booth algorithms. NxN bit multiplication involves N partial
products but modified Radix 2r produces N/r partial .In the
proposed modified Radix -4 Booth Algorithm, multiplier has
been divided in groups of 3 bits and each group of 3 bits have
been considered according to modified Booth Algorithm for
generation of partial products 0, ±1A, ±2A. In first group, first
bit is taken zero and other bits are least significant two bit of
multiplier operand. In second group, first bit is most
significant bit of first group and other bits are next two bit of
multiplier operand. In third group, first bit is most significant
bit of second group and other bits are next two bit of
multiplier operand. This process is carried on. Partial product
is generated using multiplicand operand A. For n bit
multiplier there is n/2 or [n/2 + 1] groups and partial products
.for 16X16 bit multiplication 8 partial products are generated.
So it reduces the number of partial products in comparison to
Booth algorithm (radix-2) improves the computational
efficiency of multiplier, reduce the calculation delay. Fig.3. Wallace Tree Structure

3.WALLACE TREE STRUCTURE 4.FINAL ADDER

The Wallace tree method [8] is used in high speed designs in The CSLA is used in many computational systems to
order to produce two rows of partial products that can be moderate the problem of carry propagation delay by
independently generating multiple carries and then select a
added in the last stage. Half adder, full adder, unit adder and
carry to generate the sum. However, the CSLA is not area
carry save adders are employed in Wallace tree structure to efficient because it uses multiple pairs of Ripple Carry Adders
reduce partial products. Wallace tree accelerate the (RCA) to generate partial sum and carry by considering carry
accumulation of the partial products. The speed, area and input and then the final sum and carry are selected by the
power consumption of the multipliers will be in direct multiplexers (mux). Regular carry select adder is formed by
proportion to the efficiency of the compressor. The structure using uniform block composed of full adders for both the
is made of carry save adders. The carry save adders reduce the carry inputs viz cin=0 & cin=1and mux to select correct sum
number of partial products and sum rows [7]. Compressors are output. A 16-bit carry select adder with a uniform block size
arithmetic components, similar in principle to parallel adders. has the delay of four full adder delays and three MUX delays.
A 16-bit carry select adder shown in fig 4 with variable block
The 4:2 compressor[11] shown in Fig.2 has 4 input bits and
size has the delay of two full adder delays, and four mux
produces 2 sum output bits (out0 and out1 ), it also has a delays[3].Therefore we use carry select adder with variable
carry-in ( cin ) and a carry-out ( cout ) bit (thus, the total block size. 32 bit SQRT CSLA[5] as final adder enhance
number of input/output bits are 5 and 3). speed of multiplier. SQRT CSLA is further modified to get
better performance. It is similar to regular 32-bit SQRT
CSLA. In Modified SQRT CSLA one RCA fed with constant
X3 X2 X1 X0 1 carry in of Basic block is replaced by Binary to one
Converter (BEC)[2]. The BEC is designed by using NOT,
XOR and AND gates. Four Bit BEC is shown in Fig.5.
C0 C1
Carry save
add.
C S
Fig.2. Carry save Adder, 4:2 Compressor

The Carry save adder structure for eight partial products is

shown in Fig.3. 16 bit carry save adder blocks is used. The
first row adds partial products as P4, P3, P2 and P1and the
partial products P8, P7, P6 and P5. The second row adds the
sum & carries outputs from the first row.

Fig.4. 16 Bit SQRT Carry Select Adder

40
 International Journal of Computer Applications (0975 – 8887)
Volume 68– No.13, April 2013

Fig.5. 4 Bit Binary to excess one converter (BEC)

Fig.8 Internal view of multiplier using SQRT CSLA

Fig.6. 16 Bit Modified SQRT Carry Select Adder

5. SIMULATION RESULTS
The simulation results of the Modified Booth Multiplier with
SQRT Carry Select Adder are shown in fig. 7, 8 and 9. &
simulation results of the Modified Booth Multiplier with
modified SQRT Carry Select Adder are shown in Fig. 10, 11
and 12.Table 1 shows the comparison of Modified Booth
Multiplier using SQRT CSLA & using modified SQRT CSLA
with MBM using regular CSLA in terms of area , speed & Fig.9. Simulation of multiplier using SQRT CSLA.
levels of logic.

Fig.7.RTL View of multiplier using SQRT CSLA. Fig.10.RTL View of multiplier using Modified SQRT
CSLA.

41
 International Journal of Computer Applications (0975 – 8887)
Volume 68– No.13, April 2013

6.CONCLUSION
In this paper, the design & implementation of two high speed
multipliers is proposed. The first multiplier makes use of
radix-4 booth algorithm with 4:2 compressors & SQRT
CSLA. The second Multiplier uses Radix-4 booth algorithm
with 4:2 compressor & modified SQRT CSLA.
Both the Multipliers show reduction in delay and levels of
logic with slight increase in area. The first multiplier shows
more reduction in delay. The second multiplier shows more
reduction in levels of logic.
We can extend this work by employing Radix-8 Booth
algorithm for partial products generation. Expected outcome
is less number of partial products, reduced area & reduced
delay.

7.REFERENCES
[1] Kulvir Singh , Dilip Kumar “Modified Booth Multiplier
with Carry Select Adder using 3-stage Pipelining
Technique” International Journal of Computer
Applications (0975 – 8887) Volume 44– No14, April
2012
[2] A. Andamuthu, S. Rithanyaa.” Design Of 128 Bit Low
Power and Area Efficient Carry Select Adder”
International Journal of Advanced Research in
Engineering (IJARE) Vol. 1, Issue 1,2012 Page 31-34
[3] Aparna P R, Nisha Thomas” Design and Implementation
Fig. 11.Internal view of multiplier using SQRT CSLA
of a High Performance Multiplier using HDL”
[4] Dong-Wook Kim, Young-Ho Seo, “A New VLSI
Architecture of Parallel Multiplier-Accumulator based on
Radix-2 Modified Booth Algorithm”, Very Large Scale
Integration (VLSI) Systems, IEEE Transactions, vol.18,
pp.: 201-208, 04 Feb. 2010
[5] Kim, Y. and Kim, L.-S., (May2001), “64-bit carry-select
adder with reduced area, “Electron Lett., vol. 37, no. 10,
pp. 614–615.
[6] V. Oklobdzija, “High-Speed VLSI Arithmetic Units:
Adders and Multipliers in Design of High-Performance
Microprocessor Circuits”, Book Chapter, Book edited by
A Chandrakasan, IEEE Press, 2000.
[7] J. Fadavi-Ardekani, “M × N booth encoded multiplier
generator using optimized Wallace trees”, IEEE
Transaction on Very Large Scale Integration (VLSI)
System, vol. 1, pp. 120–125, 1993.
[8] C. S. Wallace, “A Suggestion for a Fast Multiplier”,
Electronic Computers, IEEE Transactions, vol.13,
Fig.12. Simulation of multiplier using Modified SQRT Page(s): 14-17, Feb. 1964
CSLA. [9] Neil H.E Weste, David Harris, Ayan Banerjee, “CMOS
VLSI DESIGN”,PEARSON Education, New Delhi,2004
Table1.Comparison of Multipliers
[10] J. Bhasker, “A VHDL PRIMER” PEARSON Education,
Logic Utilization Multiplier Multiplier Multiplier New Delhi,2006
using CSA using using [11] Behrooz Parhami, "Computer Arithmetic", Oxford Press,
SQRTCSA Modified 2000, pp131
SQRTCSA
Number of 669 678 676
Slices
Number of 1272 1287 1281
LUTS
Levels of Logic 26 23 22
Combinational 34.023ns 32.672ns 33.852ns
Path Delay

42

View publication stats