0% found this document useful (0 votes)
434 views12 pages

Joshi-VLSI Signal Processing

This document discusses efficient implementation of signal processing algorithms for real-time applications using different approaches like digital signal processors (DSPs), field-programmable gate arrays (FPGAs), and full custom VLSI designs. It outlines typical DSP algorithms, applications, and techniques for optimization at the architectural and implementation levels. Examples of implementing a second-order notch filter using an allpass structure with minimum resources including only two multipliers are also provided.

Uploaded by

sureshgr
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
434 views12 pages

Joshi-VLSI Signal Processing

This document discusses efficient implementation of signal processing algorithms for real-time applications using different approaches like digital signal processors (DSPs), field-programmable gate arrays (FPGAs), and full custom VLSI designs. It outlines typical DSP algorithms, applications, and techniques for optimization at the architectural and implementation levels. Examples of implementing a second-order notch filter using an allpass structure with minimum resources including only two multipliers are also provided.

Uploaded by

sureshgr
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 12

VLSI Signal Processing

Y. V. Joshi SGGS Institute of Engineering and Technology, Nanded


IUCEE Workshop presentation-YVJoshi

Introduction
Efficient implementation of signal processing algorithms required for real time signal processing Many ways
Working at architectural level Implementation level
Using Digital Signal Processors Using FPGAs Using Full custom design
IUCEE Workshop presentation-YVJoshi

Typical DSP algorithms


Simple
Convolution/Correlation Transformations- FFT/DCT/Wavelet Digital filtering Filter banks

Complex
Neuro Computing Computer Vision Artificial organ development

etc
IUCEE Workshop presentation-YVJoshi

Applications
MP3 players Mobiles Embedded systems (automobiles, aircraft, satellite etc) Robots Projection systems Biomedical/Life support systems Communication systems- Software defined radios, etc
IUCEE Workshop presentation-YVJoshi

Architectural level
Techniques of reducing the critical path
Pipelining and Parallel Processing Retiming Unfolding Folding

Dependent on optimization in data flow graphs

IUCEE Workshop presentation-YVJoshi

Implementation levelDS Processors

Selection of DSP based on application demand


Analog Devices DSP TI DSP Motorala DSP

Selection of Tools/Hardware platforms Development of algorithms on DSP


IUCEE Workshop presentation-YVJoshi

Implementation levelFPGA

Selection of FPGA based on application demand


Xilinx Alterra

Selection of Tools/Hardware platforms Development of algorithms for FPGA

IUCEE Workshop presentation-YVJoshi

Implementation levelFull Custom VLSI based

Selection of EDA tool based on application demand


Cadence/Tanner/Laker

Selection of Design methodology Development of layouts (LVS/DRC clean) Chip mfg.

IUCEE Workshop presentation-YVJoshi

Resources
Basic units
Adder Multiplier and A delay

Structures used in Signal processing


Simple Direct form/Transpose form Cascade/Parallel Digital two pair/Lattice Single multiplier per order structure Derived structures (low sensitivity to coefficient quantisation)
IUCEE Workshop presentation-YVJoshi

Example
A second order notch filter 1 2 b0 b1 z b2 z H ( z) a0 a1 z 1 a2 z 2

Using allpass structure


1 H ( z) 1 A2 ( z ) with 2 k 2 k 2 (1 k1 ) z 1 z 2 A2 ( z ) 1 k 2 (1 k1 ) z 1 k 2 z 2
IUCEE Workshop presentation-YVJoshi

Example (Contd)
With values of
1 tan(B / 2) k1 cos( 0 ) and k 2 1 tan(B / 2)

With 0 being the notch frequency and B being the rejection bandwidth implementation with only two multipliers with minimum critical path
IUCEE Workshop presentation-YVJoshi

Thanks
Contact : Y. V. Joshi [email protected]

IUCEE Workshop presentation-YVJoshi

You might also like