POWER REDUCTION IN VLSI SYSTEMS BY USING MULTI-BIT FLIP-FLOPS

Scientific Journal Impact Factor (SJIF): 1.711
International Journal of Modern Trends in Engineering and
Research
www.ijmter.com
@IJMTER-2014, All rights Reserved 51
e-ISSN: 2349-9745
p-ISSN: 2393-8161
POWER REDUCTION IN VLSI SYSTEMS BY USING
MULTI-BIT FLIP-FLOPS
1
Sivakrishna Modiboyena, 2
Kanumuri Koteswararao
1
PG Scholar, VLSI& Embedded System, prakasam engineering college,kandukuru
2
ECE department, prakasam engineering college,kandukuru
Abstract- The utilization of power has turned into a smoldering issue in current VLSI design.
Power consumption can be reduced by substituting some flip-flops with less multi-bit flip-flops.
Multi-bit flip-flops are one of the strategies for reducing the clock power consumption. This project
concentrates on reducing of clock force utilizing multi-bit flip-flops by clock synchronization.
Diminishment of the clock power consumption with two single bit flip-flops are synchronized with
single clock pulse. Uniting single bit flip-flops into one multi-bit flip-flop evades duplicate inverters,
brings down the aggregate clock power utilization which lessens the total area. A mixture table is
fabricated to acquire a multi-bit flip-flop which can store the flip-flops that can be consolidated. This
task concentrates on D flip-flop which builds the loading of the clock signal. QCL adder is utilized as
an application for multi-bit flip-flop. Highest ‘1’ bit finding algorithm is utilized to discover the
highest 1 bit from the yield of QCL adder. This calculation checks the yield of QCL adder in each
one cycle.
Keywords: QCL, SOC, Mbffs, merging, power reduction, clock buffer.
I.INTRODUCTION
Because of the prominence of compact electronic items, low power framework has pulled in
more consideration lately. As technology advances, a system-on-a-chip (SOC) configuration can
contain more parts that prompt a higher power density. This makes power dissipation achieve the
cutoff points of what packaging, cooling or other framework can help. Decreasing the power
consumption can upgrade battery life as well as can evade the overheating issue, which would build
the level of trouble of packaging or cooling consequently, the thought of power consumption in
complex SOCs has turned into a huge test to designers. In addition, in advanced VLSI plans, power
consumed by clocking has taken a significant piece of the entire plan particularly for those designs
using deeply scaled CMOS technologies. In this way, a few strategies have been proposed to
decrease the power consumption of clocking.
For a given plan that the areas of the cells have been firm, the power consumed by clocking
can be decreased further by substituting a few flip-flops with multi-bit flip-flops. At clock tree
synthesis, less number of flip-flops implies reduced number of clock sinks. Therefore, the resulting
clock system uses reduced power consumption and utilizes less routing resource.
Furthermore, smaller flip-flops are substituted by bigger multi-bit flip-flops; gadget varieties
in the relating circuit can be orderly reduced. As the CMOS technology progresses, the driving
capacity of an inverter-based clock buffer increments fundamentally. The ability to drive a clock

International Journal of Modern Trends in Engineering and Research (IJMTER)
Volume 01, Issue 05, [November - 2014] e-ISSN: 2349-9745, p-ISSN: 2393-8161
buffer can be assessed by the quantity of least measured inverters that it can drive on a given rising
or falling time. Due to this sensation, a few flip-flops can impart a common clock buffer to evade
unnecessary waste of power.
Fig. 1 shows the block diagrams of 1- and 2-bit flip-flops. If we replace the two 1-bit flip-
flops as shown in Fig. 1(a) by the 2-bit flip-flop as shown in Fig. 1(b), the total power consumption
can be reduced because the two 1-bit flip-flops can share the same clock buffer.
Figure1. Example of merging two 1-bit flip-flops into one 2-bit flip-flop
(a) Two 1-bit flip-flops (before merging). (b) 2-bit flip-flop (after merging)
In any case, the areas of some flip-flops would be changed after this substitution, and
subsequently the wire lengths of nets connecting pins to a flip-flop are additionally changed. To
abstain from damaging the timing imperatives, we confine that the wire lengths of nets uniting pins
to a flip-flop can't be longer than detailed values after this procedure. On the other hand, to ensure
that another flip-flop can be put inside the desired region, we likewise need to consider the area
capacity of the region.
The power plays a significant part in any design one may need to focus on power reduction
strategies. To diminish the power consumption, a lot of low-power plan procedures have been
presented, for example, clock gating, power gating making multi-supply-voltage plans, dynamic
voltage per frequency scaling, and minimizing clock system. Among these procedures, minimizing
and fusing the clock system is essential in reducing power consumption of a Soc (System on Chip).
By diminishing the power in circuit design it naturally reduces the many-sided quality and wire
length. In this manner, distinctive systems have been proposed [2], [3] to design a reduced power
consumption design.
The power had been expanded for diverse stages are static and dynamic power. In dynamic
power, change in input signal at distinctive rationale level will result in exchanging and short out
force in the configuration. In static force, it doesn't have any impact of level change in information
and yield. The Multi-bit Flip-flop (MBFF) is a successful power reduction procedure. It is utilized to
decrease the quantity of Flip Flop away stage. Sending numerous bits of information with single FF

utilizing single clock pulse is called MBFF. The idea of MBFF is presented in adder application
which is utilized to diminish the quantity of FFs which are not empowered in the circuit outline.
Mbffs have advantage over SBFF as more modest outline zone, controllable clock, less delay on
clock system and effective use of routing resources.
The working of multi-bit flip flop is same as single-bit flip-flop, at whatever point the clock
gets dynamic state flip flop latches all data to yield. For idle state the flip flop holds the information.
The fundamental structure of multi-bit flip failure is given in Fig. 1, it demonstrates that as opposed
to utilizing single bit FF we can supplant into multi bit FF as 2-bit FF, 4-bit FF and 8-bit FF are
produced as a different assignment. At the point when will the obliged bit of capacity FF is required
the specific errand is, no doubt brought in active region and others will be in-active (sleep mode)
region.
In the proposed work it takes after that it is utilized to store the quantity of bits that are
empowering specifically flip-flop utilizing single check and others are in sleep mode. It doesn't
devour power for other flip-flop which is not empowered during the storage stage.
Figure2. Block diagram of MBFF
The multi smaller FF is supplanted by larger MBFF utilizing the less clock source; all the
more over gadget varieties in the relating circuit can be successfully reduced. The FF can be fused
with the assistance of combinational table which will be powerfully empowered built in light of
the number of bit capacity necessity with force thought. The FF going to be united can be utilized
for memory shows. By decreasing the quantity of Ffs, the clock sinks area and clock dynamic power
have been viably diminished.
II.EXISTING METHOD
Ya-Ting Shyu et al [1] had utilized the numerous single bit flip- flop are supplanted by multi
bit flip-flop. Because of this force is expanded and intricacy in configuration. The procedure [4]
used to diminish control in the post-arrangement stage. In this work a chart based method is utilized

within request to decrease the clock power. The flip-lemon is spoken to for every hub in the chart.
The flip-failure relating to the hubs in an m-inner circle can be supplanted by an m-bit flip lemon.
The calculation is utilized to discover m-coteries in a chart are extension and-bound and
backtracking calculation. An alternate calculation is additionally used to discover the most extreme
autonomous gathering of factions is ravenous heuristic calculation. In this work there is a
probability of discovering outlandish mix of flip tumble in a library. Because of this it may prompt
the wastage of time and more number of Ffs is utilized within every hub.
III.PROPOSED METHOD
In the past technique [1] the measure of time is wasted by discovering the impossible
combination of FF furthermore numerous single bit FF is utilized. This may expand the
complicated nature. So as to decrease the power MBFF idea is utilized. it portrays that need to
recognize a legal placement region for every FF. In first stage, the reasonable placement regions of
a FF connected with diverse pins are discovered focused around the timing stipulations
characterized on the pins. At that point, the legal placement region of the FF can be obtained by
overlapped area of these regions.
Nonetheless, these regions are fit as a diamond shape; it is not simple to recognize the
overlapped region. Accordingly, the overlapped zone can be recognized all the more effectively in
the event that it can change the coordinate arrangement of cells to get rectangular regions. In the
second stage, it might want to manufacture a combination table, which characterizes all
combinations of FF keeping in mind the end goal to get another multi-bit Ffs given by the library.
Figure3. Flow-chart of merging flip-flop
The flip-flops can be united with the assistance of the table. After the legal placement regions
of flip-flops are discovered and the combination table is fabricated, we can utilize them to merge

flip-flops. To accelerate our project, we will isolate a chip into a few canisters and consolidation
flip-flops in a neighborhood bin.
However, the flip-flops in diverse bins might be mergeable. In this way, we need to
consolidate a few bins into a bigger bin and repeat this venture until no flip-flop can be fused any
longer. In this area, we would detail each one phase of our technique. In the first subsection, we
demonstrate a basic equation to change the original coordination framework into another one so that
a legal placement region for each one flip-flop can be distinguished all the more effectively. The
second subsection shows the flow of building the combination table. At long last, the substitutions
of flip-flops will be depicted in the last subsection.
A.TRANSFORMATION OF PLACEMENT SPACE
The equations used to transform coordinate system are shown in (1) and (2). Suppose the
location of a point in the original coordinate system is denoted by (x, y). After coordinate
transformation, the new coordinate is denoted by (x‟, y‟). In the original transformed equations,
each value needs to be divided by the square root of 2, which would induce a longer computation
time. Since we only need to know the relative locations of flip-flops, such computation are
ignored in our method. Thus, we use x” and y”, to denote the coordinates of transformed
locations.
B.COMBINATION TABLE
A few flip-flops can be replaced by multi-bit flip-flop. In this proposed methodology, the
combination table is assemble, which is utilized to get achievable flip-flops before substitution. This
makes to use for recognizing the specific flip-flop which will be empowered in active region and
cannot be covered. Utilizing this combination table, the flip-flop can be bit by bit replaced and this
makes lessens the multifaceted nature of the configuration. Since one and only combination of flip-
flop need to be considered in each one time, the clock signal can be successfully decreased.
IV APPLICATION DEVELOPED
The 1-bit, 2-bit, 4-bit and 8-bit Ffs are created as partitioned assignment as demonstrated in
Fig. 3. The two inputs zone and b, is spoken to as input1 and b is spoken to as input2. These two
inputs are included and put away in the FF updating. After that it checks the bits that are accessible
in the area. The chosen Ffs are used when it is empowered and yield is shown. This makes
decreases the power and delay in the design. The low power affects in the expense, size, weight,
execution and unwavering quality.
The multiplier application can likewise be carried out in this proposed work. As opposed to
including the bits, reproducing is possible and it is put away in the specific enabled flip-flop. For
case, accept that a library just helps two sorts of flip-flops whose bit widths are 1 and 4
methods the specific flip-flop will be chosen and it will be empowered in the area and will be
in sleep mode (in-active region).

The D-FF is utilized as a part of this proposed work. It gives synchronous information
exchange and utilized for capacity reason. In any case, a dissimilar latch element, a FF just
duplicates the information from the data pin to the yield once for every clock period and does not
permit various multiple logic values to be passed in a clock cycle. Information is exchanged at
either the rising or the falling clock edge, contingent upon the flip-flop setup. Unlike latch, a FF is
not level-sensitive, yet rather edge-activated. As it were, information gets put away into a FF just at
the dynamic edge of the clock. The 16 bit FF can likewise be produced as indicated in Fig. 4; it
diminishes the power and memory gadgets contrasted with single bit flip lemon. By and large, the
snake libraries comprises AND, XOR as well as dominant part doors. The register banks are
utilized to store the bits when they are enabled.
Figure4. Block diagram for MBFF using adder application
The D Flip-flop is the edge-triggered variation of the transparent latch. On the rising
(typically, albeit negative edge triggering is possible) edge of the clock, the yield is given the

estimation of the D data at that minute. The yield can be just change at the clock edge, and if the data
changes at different times, the yield will be unaffected.
D flip-failures are by a wide margin the most well-known sort of flip-flops and a few gadgets
are made altogether from D flip-flops. They are regularly utilized for shift- registers and input
synchronization.
Objectives
1. Reduce the power consumption.
2. To reduce to the area.
3. To reduce the delay and power of a clock network.
4. To control clock skew because of common clock signal.
The above objectives can be achieved by merging several flip-flops and synchronizing with clock
signals.
Problem Statement
The following problem statement has been identified:
1) Several Flip-flops needs a separate clock signal, hence Power consumption, is high.
2) Since several flip-flops needs a separate clock signal area consumed is also high.
V. BLOCK DIAGRAM AND ITS MODULES
This deals with the block diagram of the proposed method and its modules.
Block Diagram
The block diagram of the Application of Multi-bit flip-flop using QCL Adder as shown in figure
4.Two inputs are given to QCL adder. QCL adder are developed by Majority Logic XOR, AND, OR
gate. The output of QCL adder is fed to highest bit "1‟ finding Algorithm. This Algorithm finds the
number of bits and the combination table is built in order to merge the Flip-flops and it is stored in
the Variable register banks.
Modules
This focuses on three different types of modules which are explained below.
1) Design And Analysis Of Multi-Bit Flip-Flops
2) Design Of Memory Device Using Multi-Bit Flip Flop
3) Design and analysis of the Integration module
Design and Analysis of Multi-Bit Flip-Flops
This module is utilized to decrease the power utilization by substituting some flip flop with less
Multi-Bit flip flops. We are utilizing the Multi-Bit flip flop rather than more single bit flip flop to

expand the clock synchronization. This will diminish the unnecessary force wastage through the
utilization of numerous clock sinks.
Design of Memory Device Using Multi-Bit Flip Flop
This is the application module to be developed. The memory designed by mainly using the multi-
bit flip flops. In this, power consumption of memory devices is reduced compare to the single bit
memory.
Design and Analysis of the Integration Module
We are integrating all the sub modules and output signals are simulated.
VI. RESULTS
For the application module given above section was simulated. For adder, when clock leading
edge input is 0 and the trailing edge input is 1, reset input is 1, input for a is 0000000 and the input
for b is 01111111.The output is 01111111 as shown in the Fig below.
For example if the a input is 0000011 and if the b input is 0000011 the output will be
00000110 and the clock signal is given to the flip-flop that is required to show the output. The
number of one’s is two, so the two bit register was enabled the remaining flip-flops are in
deactivation mode. By this we can reduce the power that is required for the operation in the system
on chip.
COMPARISION TABLE
This table shows the delay and the power consumption by the clock utilization. The delay and
clock power was almost same for all the designed flip-flops.
Table1. Delay and clock power of flip-flops
FF size Delay(ps) Clock Power(w)
1 bit 0.487 0.017
4 bit 0.487 0.018
8 bit 0.487 0.018

Figure5. Shows simulation output of adder
VII. CONCLUSION
This project has proposed a methodology for flip-flop substitution for power reduction in digital
integrated circuit design. The system of flip-flop substitutions is relying upon the combination table,
which records the connections among the flip-flop types. By the rules of substitutions from the
combination table, the incomprehensible combinations of flip-failures won't be viewed as that
reductions execution time. Other than power reduction, the destination of minimizing the aggregate
wire length likewise considered to the expense capacity. The verilog source code had produced for
the application module as indicated in above areas and simulated utilizing the Isim test system. The
single bit and multibit flip-flops source code additionally planned and reproduced and combined
utilizing Xilinx ISE Design suite. This methodology can be appropriate for any circuit comprising of
various flip-flops like counters registers.
VIII. REFERENCES
[1] Ya-Ting Shyu, Jai-Ming Lin, Chun-Po Huang, Cheng-Wu Lin, Ying- Zu Lin, and Soon- Jyh Chang, 2013,
„Effective and efficient approach for power reduction by using Multi-bit Flip-flops in IEEE transactions on VLSI,
vol. 21, no. 4.
[2] H. Kawagachi and T. Sakurai, 1997, „A reduced clock-swing flip-flop (RCSFF) for 63% clock power reduction ,
in VLSI Circuits Dig. Tech. Papers Symp., pp. 97–98.
[3] Y. Cheon, P.-H. Ho, A. B. Kahng, S. Reda, and Q. Wang, 2005, Power-aware placement , in Proc. Design Autom.
Conf., pp. 795–800.
[4] Y.-T. Chang, C.-C. Hsu, P.-H. Lin, Y.-W. Tsai and S.-F. Chen, 2010, Post-placement power optimization

with multi-bit flip- flops , in Proc. IEEE/ACM Comput.-Aided Design Int. Conf., SanJose, CA, pp. 218–223.
[5] P. Gronowski, W. J. Bowhill, R. P. Preston, M. K. Gowan, and R.L. Allmon, “High- performance
microprocessor design,” IEEE J. Solid-State Circuits, vol. 33, no. 5, pp. 676–686, May 1998.
[6] L. Chen, A. Hung, H.-M. Chen, E. Y.-W. Tsai, S.-H. Chen, M.-H. Ku, and C.-C.Chen, “Using multi-bit flip-
flop for clock power saving by Design Compiler,” in Proc. Synopsys User Group (SNUG), 2010.
[7] J.-T. Yan and Z.-W. Chen, “Construction of constrained multi-bit flip-flops for clock power reduction,” in Proc.
ICGCS, pp. 675–678, 2010.
[8] S.-H. Wang, Y.-Y. Liang, T.-Y. Kuo, and W.-K. Mak, “Power-driven flip-flop merging and relocation,” in
Proc. ISPD, pp. 107–114, 2011.
[9] J. M. Rabaey, A. Chandrakasan, and B. Nikolic, 2003, Digital Integrated Circuits: A Design Perspective,
2nd ed. Upper Saddle River, NJ: Prentice-Hall
[10] Y. Kretchmer, 2001, “Using multi-bit register inference to save area and power,” EE Times Asia.

POWER REDUCTION IN VLSI SYSTEMS BY USING MULTI-BIT FLIP-FLOPS

POWER REDUCTION IN VLSI SYSTEMS BY USING MULTI-BIT FLIP-FLOPS

More Related Content

What's hot (20)

Similar to POWER REDUCTION IN VLSI SYSTEMS BY USING MULTI-BIT FLIP-FLOPS (20)

More from Editor IJMTER (20)

Recently uploaded (20)

POWER REDUCTION IN VLSI SYSTEMS BY USING MULTI-BIT FLIP-FLOPS