Authors :
Jalaja S; Pooja M.V
Volume/Issue :
Volume 8 - 2023, Issue 7 - July
Google Scholar :
https://bit.ly/3TmGbDi
Scribd :
https://tinyurl.com/29zf54bn
DOI :
https://doi.org/10.5281/zenodo.8211196
Abstract :
The design of Finite Impulse Response (FIR)
filter performance is analyzed using Reconfigurable
multipliers unit (Dadda, Booth, Wallace, and Shift & Add
multipliers) and retimed SQRT CSLA block. The FIR
filter is frequently used in digital signal processing
technique for a variety of applications including speech
processing, loudspeaker equalization, echo cancellation,
noise cancellation, arithmetic computations, and image
processing. In this paper, the FIR filter takes an input
channel and produces multiple output channels by
multiplying the input samples with corresponding filter
coefficients. The reconfigurable nature of the filter allows
for flexibility in selecting the type of multiplier based on
specific performance requirements or resource
constraints. The specific architecture and
interconnections of the components, such as multipliers,
adders, and output channels, depend on the chosen
multiplier type and the desired property of the filter. The
nature of the Control signals is to switch between
different multiplier types and adjust the filter
accordingly. To optimize the utilization of resources, a
resource sharing principle is employed in the proposed
FIR filter architecture, regardless of the number of
channels and taps. These techniques ensure efficient
resource allocation and utilization. The FIR architecture
is restructured by incorporating adders and different
multipliers in this design. This approach effectively
reduces the area occupied by the adders and multiplier
blocks, resulting in improved area efficiency and delay.
The structure of the FIR filter has the multipliers
arranged in a Multiply-Accumulate (MAC) structure,
where the multiplication and accumulation operations are
performed, and the delay blocks serve as the major
building blocks of the filter. The speed of the multiplier is
one the component of FIR filter performance, as it
determines the critical path in the filter structure. As a
result, the proposed architecture power consumption is
less compared to existing method [18][19][20][21]. The
modified FIR filter coding is implemented using, Verilog
Hardware Description Language (HDL). The simulation
and synthesis processes are carried out which allows for
testing and optimization of the design. The paper
introduces a novel approach low power Reconfigurable
multiplier unit to design Finite Impulse Response
architecture and it shows better efficiency compared to
existing architecture [9].
Keywords :
Reconfigurable Multiplier, FIR Filter, Dadda, Booth, Wallace, and Shift & Add Multipliers, Resource Sharing Principle, Retime SQRT CSLA.
The design of Finite Impulse Response (FIR)
filter performance is analyzed using Reconfigurable
multipliers unit (Dadda, Booth, Wallace, and Shift & Add
multipliers) and retimed SQRT CSLA block. The FIR
filter is frequently used in digital signal processing
technique for a variety of applications including speech
processing, loudspeaker equalization, echo cancellation,
noise cancellation, arithmetic computations, and image
processing. In this paper, the FIR filter takes an input
channel and produces multiple output channels by
multiplying the input samples with corresponding filter
coefficients. The reconfigurable nature of the filter allows
for flexibility in selecting the type of multiplier based on
specific performance requirements or resource
constraints. The specific architecture and
interconnections of the components, such as multipliers,
adders, and output channels, depend on the chosen
multiplier type and the desired property of the filter. The
nature of the Control signals is to switch between
different multiplier types and adjust the filter
accordingly. To optimize the utilization of resources, a
resource sharing principle is employed in the proposed
FIR filter architecture, regardless of the number of
channels and taps. These techniques ensure efficient
resource allocation and utilization. The FIR architecture
is restructured by incorporating adders and different
multipliers in this design. This approach effectively
reduces the area occupied by the adders and multiplier
blocks, resulting in improved area efficiency and delay.
The structure of the FIR filter has the multipliers
arranged in a Multiply-Accumulate (MAC) structure,
where the multiplication and accumulation operations are
performed, and the delay blocks serve as the major
building blocks of the filter. The speed of the multiplier is
one the component of FIR filter performance, as it
determines the critical path in the filter structure. As a
result, the proposed architecture power consumption is
less compared to existing method [18][19][20][21]. The
modified FIR filter coding is implemented using, Verilog
Hardware Description Language (HDL). The simulation
and synthesis processes are carried out which allows for
testing and optimization of the design. The paper
introduces a novel approach low power Reconfigurable
multiplier unit to design Finite Impulse Response
architecture and it shows better efficiency compared to
existing architecture [9].
Keywords :
Reconfigurable Multiplier, FIR Filter, Dadda, Booth, Wallace, and Shift & Add Multipliers, Resource Sharing Principle, Retime SQRT CSLA.