Design of a Novel Multiplier and Accumulator using Modified Booth Algorithm with Parallel Self-Time Adder

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 04 Issue: 10 | Oct -2017 www.irjet.net p-ISSN: 2395-0072
© 2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 884
Design of a Novel Multiplier and Accumulator using Modified Booth
Algorithm with Parallel Self-Time Adder
Sareddy. Sindhuja Reddy1, CH. Bhanu Prakash2
1Student,Master of Technology, Dept of Electronics and Communication Engineering Malla Reddy Engineering
college(Autonomous), Telangana, India.
2Associate Professor, Dept of Electronics and Communication Engineering Malla Reddy Engineering
college(Autonomous), Telangana, India.
-------------------------------------------------------------------------***-------------------------------------------------------------------------
Abstract- Multiplier is one of the most desirable
components in DSP processors, Fast Fourier Transform
Units and Arithmetic Logic Units. In this paper novel method
for multiplier and accumulator(MAC) is proposed based on
PASTA. Modified booth algorithm produces less delay in
comparison with a regular multiplication process, and it also
moderates the number of partial products. The major
purpose of designing is to reduce the circuit complexity,
power consumption and no loss of information. We also
proposed a CSA design from the conventional system
(modified booth algorithm) which exhibits high
performance regarding computation, power consumption,
and area. Area, delay and power complexities of the
resulting design is reported. The proposed MAC design with
PASTA shows better performance compare to the
conventional method and has advantages of reduced area
overhead and critical path delay. The results are simulated
and synthesized using Xilinx ISE simulator.
Keywords: Multiplier and accumulator (MAC), modified
booth algorithm (MBA), Carry Save adder(CSA).
I.INTRODUCTION
The propel improvement in the field of microelectronic
makes it proficiently to utilize input energy to scramble the
information and to exchange the information speed. In vast
numbers of these abilities are produced given low power
utilization keeping in mind the end goal to meet the well-
liked applications. The multiplier is an extremely
fundamental number juggling sensible unit and is utilized
mostly in circuits. Convolution, sifting and inward items are
the input procedures of computerized flag preparing which
utilize the MAC application. Discrete wavelet transform or
discrete cosine transform is the broadly used DSP methods
which are not linear functions in nature. This is because they
are principally done by repetitive application of addition and
multiplication which determine the execution performance
and speed of the entire calculation. The modified booth’s
algorithm (MBA) is usually used for high-speed
multiplication.
Power dispersal is perceived as a essential
parameter in present-day VLSI configuration field. To fulfill
MOORE'S law and to deliver customer hardware products
with more reinforcement and less weight, low power VLSI
configuration is vital.
Quick multipliers are fundamental parts of computerized
flag preparing frameworks. The speed of duplicate operation
is of extraordinary significance in advanced flag handling
also in the universally useful processors today, mainly since
the media preparing took off. In the past duplication was for
the most part executed using an arrangement of expansion,
subtraction, and move operations. Growth can be considered
as a progression of rehashed increases. The number to be
included is the multiplicand, the quantity of times that it is
incorporated in the multiplier, and the outcome is the result.
Each progression of expansion creates a halfway item. In
many PCs, the operand typically contains a similar number
of bits. At the point when the operands are translated as
whole numbers, the item is, for the most part, double the
length of operands with a specific end goal to save the data
content. This rehashed expansion strategy that is
recommended by the number juggling definition is
moderate that it is quite often supplanted by a calculation
that makes utilization of positional portrayal. It is
conceivable to deteriorate multipliers into two sections. The
initial segment is committed to the age of fractional items,
and the second one gathers and includes them.
The fundamental increase guideline is two overlap, i.e.,
assessment of fractional items and gathering of the moved
incomplete items. It is performed by the progressive
augmentations of the segments of the moved halfway item
framework. The 'multiplier' is effectively moved and
entryways the proper piece of the 'multiplicand.' The
deferred, gated occurrence of the multiplicand should all be
in a similar segment of the moved fractional item network.

They are then added to frame the item bit for the specific
shape. Duplication is along these lines a multi- operand
operation. To stretch out the duplication to both marked and
unsigned numbers, an advantageous number framework
would be the portrayal of numbers in two’s complement
form
The MAC (Multiplier and Accumulator Unit) is
utilized for picture handling and computerized flag
preparing (DSP) in a DSP processor.
II. LITERATURE SURVEY
In an attempt to improve the speed of signal
processing VLSI systems, a new architecture for high- speed
Multiply Accumulate Units is proposed. The structural
design is based on Binary trees constructed using 4-2
compressor circuits. Increasing the speed of operation is
achieved by taking advantage of the available free input
lines of the 4-2 compressors, which result from the
parallelogram shape of the generated partial products, and
using the bits of the accumulated value to fill in these gaps.
This outcome in merging the accumulation operation within
the multiplication process. An 8-bit Multiplier Accumulator
prototype circuit using the proposed architecture is
prototyped in 0.35 -micron double metal CMOS technology
and simulated using hospice. Simulation results at 3.3 V
show that the proposed design has a delay of 4.26 ns with a
16.8 delay savings. At 150 MHz operating frequency, the
power consumption is 324 milli Watts with a 23.04% power
saving compared to other architectures not using the
merging technique. Ayman A. Fayed, The Center for
Advanced Computer Studies, the University of Louisiana at
Lafayette, 70504 4330, USA Magdy A. Bayoumi, The Center
for Advanced Computer Studies, University of Louisiana.
Adders are the main parts of processing circuits and
play a vital role in all mathematical operations like
subtraction, multiplication, division, etc. Carry Look ahead
Adder (CLA) is one of the fastest adder structures that is
widely used in the processing circuits. In this article, a new
structure for adder is proposed. The proposed structure has
extremely smaller on-chip area and delay and also it has
lower power consumption. Using the proposed structure, a
64-bit adder is designed, and results are presented. The
circuit is designed in TSMC 0.18μm CMOS technology with
1.8v power supply and simulated with HSPICE. Karami H.
Fatemeh, Isfahan University of Technology, Isfahan 84156-
83111, IranAli K. Horestani School of Electrical and
Electronic Engineering, The University of Adelaide, Adelaide,
SA 5005, Australia.
III.CONVENTIONAL SYSTEM
Modified Booth Algorithm:
It is a dominant algorithm for signed-number
multiplication, which treats both positive and negative
numbers uniformly. Multiplication consists of three steps 1)
the initial pace to create the halfway items; 2) the second
means to include the formed incomplete objects until the
point when the last two columns remain; 3) the third means
to process the last augmentation comes about by including
the last two lines.
The number of partial products are significantly
reduced in the initial step. We used the changed Booth
encoding (MBE) conspire. It is known as the most proficient
Booth encoding and interpreting plan. To multiply X by Y
utilizing the adjusted Booth calculation begins from
gathering Y by three bits and encoding into one of {-2, - 1, 0,
1, 2}. Table I demonstrates the guidelines to create the
encoded motions by MBE plan and Fig. 1 (a) explains the
relating rationale chart. The Booth decoder creates the
fractional items utilizing the encoded motions as appeared
in Fig. 1(b).
Table 1: TRUTH TABLE OF MBE SCHEME
The new MBE recorder [2] is designed according to
the following analysis. Table (1) displays reality table of the
new encoding plan
Figure. 1. The Encoder and Decoder for the new
MBE scheme. (a) Simple encoder (b) Decoder.

Figure. 2 Generated partial products and sign
extension scheme
Fig. 2 demonstrates the created fractional items and sign
expansion plan of the 8-bit adjusted Booth multiplier. The
fractional items produced by the adjusted Booth calculation
are included parallel utilizing the Wallace tree until the
point that the last two lines remain. The final augmentation
comes about are created by including the last two lines.
CSA DESIGN:
Carry Save adder.
In Carry Save Adder (CSA); three bits are added
parallel at a time. In this scheme, the carry is not propagated
through the stages. Instead, carry is stored in the present
phase, and updated as addend value in the next stage. Hence,
the delay due to the carry is reduced in this scheme. The
architecture of CSA is shown in Fig3.
Figure.3 (a)Carry Save Adder (CSA)
Figure. 3(b): Schematic of CSA
It is also same as full Adder. From the two inputs,
we first produce two temporary Outputs named as Sum and
Carry. To obtain sum bit, we first perform bitwise XOR, and
for the Carry bit, we execute bitwise AND for the two input
numbers. And then finally add them by shifting Carry bit left
by one place to Sum bit up to produce the final answer.
Figure 4. Working of Carry Save Adder
IV.PROPOSED SYSTEM
DESIGN OF PASTA
In this section, the architecture and theory behind
PASTA is presented. The adder first accepts two input
operands to perform half
Figure 5. General block diagram of PASTA.

additions for each bit. Subsequently, it iterates using earlier
generated carry and sums to perform half-additions
repeatedly until all carry bits are consumed and settled at
zero level. A. Architecture of PASTA The general architecture
of the adder is shown in Fig. 7. The selection input for two-
input multiplexers corresponds to the Req handshake signal
and will be a single 0 to 1 transition denoted by SEL. It will
initially select the actual operands during SEL = 0 and will
switch to feedback/carry paths for subsequent iterations
using SEL = 1. The feedback path from the HAs enables the
multiple iterations to continue until the completion when all
carry signals will assume zero values.
B. State Diagrams
In Fig. 8, two state outlines are drawn for the
underlying stage and the iterative period of the proposed
engineering. Each state is spoken to by (Ci+1 Si) match
where Ci+1, Si speak to complete and aggregate esteems,
individually, from the ith bit viper piece. Amid the
underlying stage, the circuit just fills in as a combinational
HA working in central mode. It is evident that because of the
utilization of HAs rather than FAs, state (11) can't show up.
Figure 6. State diagrams for PASTA. (a) Initial phase. (b)
Iterative phase
A mid the iterative stage (SEL = 1), the input way
through multiplexer piece is initiated. The convey advances
(Ci) are permitted the same number of times as expected to
finish the recursion. From the meaning of principal mode
circuits, the present outline can't be considered as a central
mode circuit as the input– yields will experience a few
advances previously creating the last yield. It isn't a Muller
circuit working outside the key mode either as inside, a few
changes will occur, as appeared in the state outline. This is
comparable to cyclic successive circuits where door delays
are used to isolate singular states.
.
Figure.7: Design of MAC using PASTA
V.SYNTHESIS AND SIMULATION RESULTS
For designing the multiplier and accumulator
parallel self time adder was used. In this section, first, we
will see the synthesis and simulation of the MAC using
modified booth algorithm with parallel self time adder. They
are designed on Xilinx ISE 14.7 with Verilog HDL. The RTL
schematics and simulation results of the proposed design
are shown below.
Figure 8: RTL schematic of proposed MAC unit using
modified booth algorithm with PASTA

Figure 9: Internal RTL schematic of proposed MAC unit
using modified booth algorithm with PASTA
Figure 10: Technology schematic of proposed MAC unit
using modified booth algorithm with PASTA
Figure 11: Simulated outputs of proposed MAC unit using
Figure 12: Design Summary of proposed MAC unit using
Comparison of the CSA and PASTA results are shown in
table III. They vary in area and delay which shows that
there is a decrease in them and the performance is more
better in Pasta.
Table III Results
AREA
SLICES LUTS FFS
DELAY
(ns)
CSA 95 170 32 4.063
ns
PASTA 42 83 16 3.667
ns
VI. CONCLUS ION
In this paper, another MAC essential arrangement is
proposed. The proposed procedure for PASTA has less
combinational route delay exactly when appeared
differently about the existing system. As pasta has utilized
the range was lessened, and the deferral is likewise
diminished that shows elite the proposed plan of MAC was
executed and mixed through Xilinx ISE gadget. The proposed
arrangement can be used efficiently where we require a
quick of operations, for instance, DSP. The results are
compared with CSA and PASTA. The results of pasta are
better in performance.
REFERENCES
[1] Seo, Young-Ho, and Dong-Wook Kim. "A new VLSI
architecture of parallel multiplier–accumulator based on
Radix-2 modified Booth algorithm." Very Large Scale

Integration (VLSI) Systems, IEEE Transactions on 18.2
(2010): 201-208
[2] Raghava Garipelly, P.Madhu Kiran, A. Santhosh Kumar,
”A Review on Reversible Logic Gates and their
Implementation,” International Journal of Emerging
Technology and Advanced Engineering (ISSN 2250-2459,
ISO 9001:2008 Certified Journal, Volume 3, Issue 3, March
2013).
[3] Yeh, Wen-Chang, and Chein-Wei Jen. "High-speed Booth
encoded parallel multiplier design." Computers, IEEE
Transactions on 49.7 (2000): 692- 701.
[4] Elguibaly, Fayez. "A fast parallel multiplier accumulator
using the modified Booth algorithm." Circuits and Systems
II: Analog and Digital Signal Processing, IEEE Transactions
on 47.9 (2000): 902- 908.
[5] Kim, Soojin, and Kyeongsoon Cho. "Design of high-speed
modified booth multipliers operating at GHz ranges." World
Academy of Science, Engineering and Technology 61 (2010):
1-4.
[6] Fayed, A.A.; Bayoumi, M.A., "A merged multiplier-
accumulator for high speed signal processing applications,“
In Acoustics, Speech, and Signal Processing (ICASSP), 2002
IEEE International Conference on, vol.3, no., pp.III-3212- III-
3215, 13-17 May 2002.
[7] Fatemeh Karami.H and Ali K. Horestani, “New Structure
for Adder with Improved Speed, Area and Power,” 2nd IEEE
International Conference on Networked Embedded Systems
for Enterprise Applications Perth, Australia, December 2011.

Design of a Novel Multiplier and Accumulator using Modified Booth Algorithm with Parallel Self-Time Adder

Recommended

More Related Content

What's hot (19)

Similar to Design of a Novel Multiplier and Accumulator using Modified Booth Algorithm with Parallel Self-Time Adder (20)

More from IRJET Journal (20)

Recently uploaded (20)

Design of a Novel Multiplier and Accumulator using Modified Booth Algorithm with Parallel Self-Time Adder