2. DSP PROCESSORS DATA FORMAT
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
2
Fixed point represent
each no by with the
minimum of 16 bits.
Floating point take 32
bits to store each
value
3. FIXED POINT FORMAT
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
3
There are four common ways that these 2^16 = 65,536
possible bit patterns can represent a number
4. FIXED POINT DATA FORMAT
UNSIGNED INTEGER
o The stored no can take one of any integer value
from 0 to 65536
SIGNED INTEGER
o Use two’s compliment to make the range
o Include negative numbers that is from -32768 to
32767
UNSIGNED FRACTION NOTATION
o The 65536 levels are spread uniformly between 0
and 1
SIGNED FRACTION FORMAT ALLOWS
o Negative no, equally spaced between -1 to 1
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
4
5. FLOATING POINT FORMAT
use a minimum of 32 bits to store each value
2^32 =4,294,967,296
This results in many more bit patterns than for fixed point
A key feature of floating point notation is
that the represented numbers are
not uniformly spaced.
All floating point DSPs can also handle fixed point numbers
e.g.- SHARC processor
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
5
6. FIXED POINT PROCESSORS
Fixed point processor are……….
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
6
Cheaper and less power consuming
Use shorter data word width that will
provide adequate accuracy
Most of fixed point DSP processor use 16 bit
data words
•Sufficient for many applications
Most DSP includes one or more accumulator
registers
•Wider than other registers
•Provide extra guard bits to avoid overflow
7. FIXED VERSUS FLOATING POINT
DSP
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
7
While floating point
devices have
•Better precision,
•higher dynamic
Range
•shorter development
cycle.
Fixed point DSPs are
generally cheaper
8. FIXED POINT VS FLOATING POINT
Fixed Point/Floating Point
FIXED POINT PROCESSOR are :
cheaper
smaller
less power consuming
Harder to program
Watch for errors: truncation, overflow, rounding
Limited dynamic range
Used in 95% of consumer products
FLOATING POINT PROCESSORS
have larger accuracy
are much easier to program
can access larger memory
It is harder to create an efficient program in C on a fixed
point processors than on floating point processors
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
8
9. PERFORMANCE
Provide
HIGH SIGNAL TO NOISE RATIO
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
9
WHAT 32 BIT FLOATING POINT
SYSTEM CAN DO THAT 16 BIT FIXED
POINT CAN’T
???????????????????
10. the signal-to-noise ratio for storing a floating point number is
about 30 million to one
while for a fixed point number it is only
about ten-thousand to one
floating point has roughly 3,000 times less quantization noise than
fixed point
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
10
11. Fixed versus floating point instructions
These are the multiplication instructions used in
the SHARC DSPs.
While only a single command is needed for floating point, many
options are needed for fixed point.
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
11
12. FIXED POINT VS FLOATING POINT
Floating Point Fixed Point
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
12
Applications
•Modems
•Digital Subscriber Line (DSL)
•Wireless Base stations
•Central Office Switches
•Private Branch Exchange (PBX)
•Digital Imaging
•3D Graphics
•Speech Recognition
•Voice over IP
Applications
•Portable Products
•2G, 2.5G and 3G Cell Phones
•Digital Audio Players
•Digital Still Cameras
•Electronic Books
•Voice Recognition
•GPS Receivers
•Headsets
•Biometrics
•Fingerprint Recognition
13. DIGITAL SIGNAL
PROCESSORS COMPANIES
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
13
ANALOG DEVICES
•ADSP-21xx-16 bit, fixed
point
•ADSP-21xxx-32 bit,
floating point and fixed
point
LUCENT
TECHNOLOGIES
•DSP 16xxx-16 bit, fixed
point
•DSP 32xx-32 bit, floating
point
MOTOROLA
•DSP 561xx,16 bit, fixed
point
•DSP 560xx,24 bit, fixed
point
•DSP 96002,32 bit,
floating point
TEXAS INSTRUMENTS
•TMS 320cxx,16 bit fixed
point
•TMS 320cxxx,32 bit,
floating point
14. CONCLUSION
floating point and fixed point usually use 32 bits and
16 bits, respectively, but not always.
Fixed point processor are :
Cheaper , smaller , less power consuming , Harder to
program
Watch for errors: truncation, overflow, rounding
Limited dynamic range
Floating point processors have larger accuracy , are
much easier to program , can access larger memory
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
14
15. REFERENCE
I feachor & Jervis - Digital Signal Processing
The Scientist and Engineer's Guide to Digital Signal Processing
http://www.sundance.com/index.htm
http://www.bdti.com/
5/19/2023
FIXED POINT AND FLOATING POINT PROCESSORS
15