Académique Documents
Professionnel Documents
Culture Documents
Abstract Digital signal processing (DSP) is used in wide Design of low power, high throughput FIR filter is hot topic in
range of applications such as telephone,radio, video, etc. Most of DSP research area. In recent years various technique for low
DSP computations involve the use of multiply accumulate area, low power FIR design have been proposed [1-2].
operations and therefore the design of fast and power efficient
Bhardwaj et al., [3] introduce the new measurement, power
mutiplier is imperative. Moreover, the demand for portable
applications of DSP architectures has dictated the need for low
awareness to indicate the ability of the system energy to scale
power and area designs.Digital Finite Impulse Response (FIR) with changing conditions and quality requirements. Parallel
filter has lot of arithmetic operation modules such as adder and (or block) processing can be applied to digital FIR filters to
multiplier modules,consume much power, energy,and circuit either increase the effective throughput or reduce the power
area.In some applications the FIR circuit must be able to operate consumptions of the original filters. While sequential FIR
at high sample rates while in other applications the FIR filter filter implementation has been given extensive configuration
circuit must be a low power circuit operating at modearte that deals with directly reducing the hardware complexity or
sample rates. power consumptions of parallel FIR filters [4]. Selecting
This paper presents the methods for implementing digital
multiplier with more hardware breadth rather than death
FIR filter that requires optimized area and less power
consumptions. Multiplierless Multiple Constant Multiplications
would not only reduce the delay, but also the total power
(MCM) technique has been used which reduces multiple circuits. consumptions [5]. There is a novel approach for a design
The Digit Serial Adder avoids the unwanted addition and thus method of low power digital base band processing to optimize
minimizes the area too. Various filters are designed using the bit-width of each filter coefficients [7]. Data transitions
MATLAB and implemented using VHDL code. Simulation is power diminution techniques (DPDT) is also used to reduce
performed using ACTIVE -HDL and functional verification is dynamic switching power of FIR filter [8].
carried out using Synopsis Design Compiler and Encounter
compiler. Multiplier plays an important part in todays DSP systems.
Examples of their use occur in implementation of recursive
and transverse filters, Discrete Fourier transform, correlation,
Index Terms Digit serial adder,FIR filter,MCM,low
area,power,shift and and multiplier range measurement and most of these cases it is enough with a
multiplier unit design for specific purpose. Multiplier has a
I. INTRODUCTION large area, long latency and consumes considerable power.
Therefore low power multiplier designs have been an
With the growing demand on battery powered mobile important part in low power VLSI system design. The primary
computing and communication devices, how to achieve low objective is power reduction with small area and delay
power dissipation in order to extend battery life becomes a overhead. By using new coding or architectures, it is even
major concern of IC designer. DSP systems are widely used in possible to achieve both power reduction and area/delay
commuting and communication area. FIR filter is one of the reduction which is strength of high level optimization. In this
basic element in DSP application. Impulse response can either paper, a novel method to design low area power-aware FIR
finite or infinite. The method for designing and implementing filter is proposed. Based on pipelining multipliers and adders
these two filter classes differ considerably. FIR filters are a very high throughput is achieved. For reduced power
digital filter whose response to the unit filter (Unit Sample consumptions and area we are using MCM technique along
Function) is finite in duration. This is in contrast to Infinite with digit serial adder, shift/add multipliers. Various filters are
impulse response (IIR) filters whose response to unit impulse compared for area and power and demonstrated that our
is infinite in duration. approach is most effective for implementation with the
FIR and IIR filters each have advantages and disadvantage, constraints of low cost and low power.
and neither is based in all situations. FIR filter can be
implemented using either recursive or non-recursive
techniques, but usually no recursive technique are used. FIR II. FIR FILTER THEORY
filters are widely used in DSP systems that are characterized
by the extensive sequence of multiplications operations. In
some applications, the FIR filter circuit must be able to Digital filters are typically used to modify or alter the
operate at high sample rates, while in other applications, the attributes of signal in the time or frequency domain. The most
FIR filter circuit must be a low power circuit operating at common digital filter is the linear time invariant (LTI) filter.
moderate sample rates. LTI interacts with its input signal through process called
linear convolution, denoted by Y=f*x where f is the filter
impulse response, x is the input signal and Y is the convolved
Manuscript received September 19, 2014.
Ms.Rashmi Patil B. Eng. Degree in Electronics & Communication output. The linear convolution process is formally defined by:
fromS.S.G.M.C.E.Shegaon, India Y[n]=x[n]*f[n]= x[n]f[n-k]= f[k]x[n-k] (1)
Dr.M.T.Kolte B.Tech, M.E., and Ph.D in Electronics and
Telecommunication. Head of Dept.in M.I.T.C.O..E.,Pune,India.
LTI digital filters are generally classified as FIR or IIR.As the
216 www.erpublication.org
Low Area Power -Aware FIR Filter For DSP
rst
filte rc lk c lk
clk
rst
clk c lk f0 c lk
x o ut(7 :0 ) x (7 :0 ) c (3 1:0 ) d (3 1:0 ) q (3 1:0 ) c(31:0)
rs t U2 e re s ult(31 :0 ) b (3 1:0 ) U3 latch32
III. FIR FILTER IMPLEMENTATION clkdiv3 h0 (7:0 ) h0 (7:0 )
h1 (7:0 ) h1 (7:0 )
h2 (7:0 ) h2 (7:0 )
h3 (7:0 ) h3 (7:0 )
h5 (7:0 )
h4 (7:0 )
h5 (7:0 )
h7 (7:0 )
h6 (7:0 )
h7 (7:0 )
h9 (7:0 ) h9 (7:0 )
h1 1 (7:0 ) h1 1 (7:0 )
h1 2 (7:0 ) h1 2 (7:0 )
h1 4 (7:0 )
h1 3 (7:0 )
h1 4 (7:0 )
h1 6 (7:0 )
h1 5 (7:0 )
h1 6 (7:0 )
filter0
these products are accumulated and give FIR filter con_coeff
output.RTL schematic of filter is shown in Fig.3. Figure 4. RTL schematic of Co-efficient Filter
217 www.erpublication.org
International Journal of Engineering and Technical Research (IJETR)
ISSN: 2321-0869, Volume-2, Issue-9, September 2014
These systems process multiple bits of the input word,
rst='1'
referred to as the digit size in one clock cycle.DSP
xout <="00000000";
architectures have in need of low-power designs [16] causes
S1 fi l tercl k <= '0'; the batteries life. Design of digit serial architectures which
counter := "00000000"; can be pipelined at the bit-level [17]. The advantage of digit
serial architectures is processing in speeds options, less area
and critical path is reduced. Therefore, bit adder of previous
xout <=counter; designed filter is replaced by digit serial adder.RTL schematic
S2
fi l tercl k <= '0';
of Digit Serial Adder Filter is shown in Fig.7.Fig.8 shows the
RTL schematic of digit serial adder.
clk
rst
counter < "11111111" datadum pgen1
U1
portsol3
S4 counter :=counter +1 ;
rst
rst
clk
D DOP
U3 clk lclk
U37
sout(7:0) x(7:0) c(31:0) c(31:0)
U2 eresult(31:0) b(31:0)
sel
h0(7:0) h0(7:0)
S5 counter := "00000000" ; h1(7:0) h1(7:0)
h2(7:0) h2(7:0)
h3(7:0) h3(7:0)
h4(7:0) h4(7:0)
h5(7:0) h5(7:0)
h6(7:0) h6(7:0)
Figure 5. FSM pattern applied to input h7(7:0) h7(7:0)
h8(7:0) h8(7:0)
h9(7:0) h9(7:0)
h13(7:0) h13(7:0)
terms of silicon area [10].So a multiplierless design in MCM
h14(7:0) h14(7:0)
based filter is proposed under multiple constant
h15(7:0) h15(7:0)
multiplications architecture. This significantly reduces the h16(7:0) h16(7:0)
area of filters when compared to those designed previously h17(7:0) h17(7:0)
rst
sel(3:0)
clk clk f0 clk
sel_sumcout sel_sumcout stopwr232
xout(7:0) x(7:0)c(31:0) d(31:0) q(31:0) R(31:0) mux14 A(1:0)
rst
filterdatagen1 U5 latch32 B(1:0)
b(31:0)
U2
clkdiv3
U1 filter2 O1(31:0) a(31:0) y(1:0)
digitadder2
Y(31:0) sel(3:0)
contout321 mux32
Figure 6. RTL schematic of MCM based filter
Figure 8. RTL schematic of digit serial adder
Bit-serial systems process one bit of the input sample in one
clock cycle, for area efficient and ideal for low speed Realization of FIR filter combining the MCM with shift-add
applications [12].On the other hand bit-parallel systems architecture and digit serial adder by using the high-level
process one whole word of the input sample in one clock cycle, optimization technique is proposed in final MCM based Digit
are ideal for high speed applications[13].Both these systems Serial Adder Filter type1.RTL schematic of final proposed
occupy considerable amount of area. To this end, digit serial FIR filter is shown in Fig 9.
systems [14] [15] have become attractive for digital designers.
218 www.erpublication.org
Low Area Power -Aware FIR Filter For DSP
rst
TABLE II.
clk
Comparison Table for Power
clkdiv3
U4
Internal Power
rst
clk
Power
Filter0 3.4263 2.4014 mW 5.8277 mW
f0
mW
Co-efficient 7.7737 5.0117 mW 12.7854 mW
U1 Filter mW
portsol3
U37
D DOP
rst
U3 c lk lc lk
based Filter mW
c (3 1:0 ) c(31:0)
s o ut(7 :0 ) x (7 :0 )
Digit Serial 4.2176 3.0952 mW 7.3128 mW
datadumpgen1 b (3 1:0 ) se l
Adder Filter mW
U2 filterdigitadder2optmul1 MCM 113.2492 35.9731 W 0. 1492223 mW
Y(3 1:0 )
based Digit W
contout321 serial Adder
Filter
Figure 9. RTL schematic of MCM based Digit serial Adder Filter
Due to the combination of both the advanced techniques, area Simulation result for proposed MCM based Digit Serial adder
and power is drastically reduced as given in Table I and Table filter in Active HDL is shown in Fig 10.
II.
Figure 10. Simulation result for proposed MCM based Digit Serial adder
TABLE I.
filter
Comparison Table for Area
Filter0 0 27050.4519 1404.14563 28454.59 In this paper various FIR filter are implemented and their
8 761
speed, power and complexity of the designs are compared. It
Co-efficien 1 29009.7793 5616.58252 34626.36
has been seen that transposed form FIR filter using Digit
t Filter 6 188
Serial Adder and MCM with shift and add technique reduces
MCM 2 15383.1843 5616.58252 20999.76
based Filter 6 688 the complexity and area. Proposed design technique for the
Digit Serial 1 13273.2121 151.114597 13424.32
digit serial architecture in the design of digit-serial operations
Adder 6 676 and FIR filter, yield better performance, with high efficiency.
Filter The proposed activity evaluation method leads to consumed
MCM 1 837.700495 151.114597 988.8150 low power estimation with fast estimation time. Also the
based Digit 92
serial
proposed design is an area efficient multiplier useful in
Adder decreasing area consequently which reduces the cost.
Filter
219 www.erpublication.org
International Journal of Engineering and Technical Research (IJETR)
ISSN: 2321-0869, Volume-2, Issue-9, September 2014
REFERENCES
Dr.M.T.Kolte has completed his B.Tech, M.E., and Ph.D in Electronics and
Telecommunication.He is working as Head of Dept.in
M.I.T.C.O..E.,Pune,India.He has presented and published many papers in
National and International Conference.
220 www.erpublication.org