Digital Signal Processors

Inderdeep Kaur Aulakh

Asst. Prof. (IT),UIET
PU, CHD

10/14/20 1
 INTRODUCTION

• Digital Signal Processors (DSPs) take real-

world signals like voice, audio, video,
temperature, pressure, or position that have been
digitized and then mathematically manipulate
them.
• A DSP is designed for performing mathematical
functions like "add", "subtract", "multiply" and
"divide" very quickly
• Signals need to be processed so that the
information that they contain can be displayed,
analyzed, or converted to another type of signal
that may be of use.
• In the real-world, analog products detect signals
such as sound, light, temperature or pressure and
manipulate them.
• Converters such as an Analog-to-Digital converter
then take the real-world signal and turn it into the
digital format of 1's and 0's.
• From here, the DSP takes over by capturing the
digitized information and processing it.
• It then feeds the digitized information back for
use in the real world.
• It does this in one of two ways, either digitally
or in an analog format by going through a
Digital-to-Analog converter.
• All of this occurs at very high speeds.
• The diagram below shows how a DSP is used in an MP3 audio player.
• During the recording phase, analog audio is input through a receiver or other source.
• This analog signal is then converted to a digital signal by an ADC & passed to the DSP.
• The DSP performs the MP3 encoding and saves the file to memory.
• During the playback phase, the file is taken from memory, decoded by the DSP and then
converted back to an analog signal
 WHAT’S INSIDE DSP???
• Program Memory: Stores the programs the DSP will use
to process data
• Data Memory: Stores the information to be processed
• Compute Engine: Performs the math processing,
accessing the program from the Program Memory and the
data from the Data Memory
• Input/Output: Serves a range of functions to connect to
the outside world
 ARCHITECTURE

• DSP algorithms generally spend most of their

execution time in loops.
• This means that the same set of program instructions
will continually pass from program memory to the
CPU.
• The Super Harvard architecture is used to overcome
this situation by including an instruction cache in
the CPU.
• This is a small memory that contains about 32 of the
most recent program instructions.
•
 WHY USE A DSP?
To get an idea of the type of calculations a DSP does and
get an idea of how an analog circuit compares with a DSP
system, one could compare the two systems in terms of a
filter function. The familiar analog filter uses resistors,
capacitors, inductors, amplifiers. It can be cheap and easy
to assemble, but difficult to calibrate, modify, and
maintain­a difficulty that increases exponentially with
filter order. For many purposes, one can more easily
design, modify, and depend on filters using a DSP because
the filter function on the DSP is software-based, flexible,
and repeatable.
 LETS TALK DSP
• Algorithms: A defined procedure for solving a
problem or performing an operation. Algorithms are
stored in a DSP through a series of instructions.
• Analog: Analog refers to electrical signals that vary
continuously over time. Analog signals must be
converted into digital form by an analog to digital
converter in order for a DSP to process them.
• Bits: The smallest unit that a digital word can be
broken down to. The number of bits reflects the word
width. The greater the number of bits in a word, the
larger the number that can be represented by that
word.
• Cycle Time: Length of time to complete an instruction.
• Digital: Digital describes electronic technology that generates,
stores, and processes data in terms of two states: positive and
non-positive. Positive is represnted by the number 1 and non-
positive by the number 0. Data transmitted or stored with digital
technology is expressed as a string of 0's and 1's. Each of these
state digits is referred to as a bit.
• DMA (Direct Memory Accessing): The DSP's I/O processor
supports DMA of data between DSP memory and external
memory, host or peripherals through the external, host, serial,
SPI, and UART ports. Each DMA operation transfers an entire
block of data.
• Fixed Point: A method used by the DSP to manipulate and
represent numbers best suited not to grow larger than the biggest
number that can be held in a single internal register. The size of
the register is determined by the number of bits it can contain.
E.g., 32-bit numbers can be larger than 16-bit numbers.
• Floating Point: A system where a number is represented with a
mantissa and an exponent. E.g., a x 2b where "a" is the mantissa
and b is the exponent. This method allows the DSP to manipulate
very small numbers or very large numbers. Floating Point
processors represent numbers in a standard format established by
the IEEE. It is a scientific notation represented by 32-bits.
• Harvard Architecture: DSPs use memory architectures that
have separate buses for program and data storage. The two buses
let the DSP get a data word and an instruction simultaneously.
• I/O: Input/Output. Interfaces/devices used for transferring data
and control information between the DSP and peripherals.
• I/O Processor: The DSP has a distributed DMA architecture with
DMA controllers for each DMA capable peripheral. Also, most
ports have some direct (non-DMA) access to internal memory
and I/O memory. The term I/O processor refers globally to the
DMA controllers, DMA channel arbitration, and peripheral-to-
bus connections
• MIPS: Millions of Instructions Per Second where each
instruction can perform multiple operations.
• MACS: Multiply-and-Accumulate Per Second
• MOPS: Million of Operations Per Second
• MFLOPS: Million of Floating Point Operations Per
Second
• Von Neumann Architecture: This is the architecture
used by most (non-DSP) microprocessors. This
architecture uses a single address and data bus for
memory access.
• Word: A string of bits representing a data value. Word
lengths in DSPs are typically 16-bit, 24-bit, 32-bit, and
48-bit.
Modern Digital Signal Processor
TI TMS320C6000 Family, Simplified Architecture

Program RAM
Data RAM
or Cache
Addr
Internal Buses
DMA
Data
Regs (A0-A15) .D1 .D2 Serial Port

Regs (B0-B15)
External .M1 .M2 Host Port
Memory
-Sync .L1 .L2 Boot Load
-Async
.S1 .S2
Timers
Control Regs
Pwr Down
CPU
 Modern DSP: TI TMS320C6000
Architecture
• Very long instruction word (VLIW) size of 256 bits
– Eight 32-bit functional units with single cycle throughput
– One instruction cycle per clock cycle
• Data word size is 32 bits
– 16 (32 on C6400) 32-bit registers in each of 2 data paths
– 40 bits can be stored in adjacent even/odd registers
• Two parallel data paths
– Data unit - 32-bit address calculations (modulo, linear)
– Multiplier unit - 16 bit  16 bit with 32-bit result
– Logical unit - 40-bit (saturation) arithmetic & compares
– Shifter unit - 32-bit integer ALU and 40-bit shifter
 APPLICATIONS
■ Radar and sonar
■ Communications systems
■ Process control
■ Image processing
■ Audio applications
 DSP Applications Examples
Embedded system cost & input/output rates
– Low-cost, low-throughput: sound cards, 2G cell
phones, MP3 players, car audio, guitar effects
– Medium-cost, medium-throughput: printers,
disk drives, PDAs, 3G cell phones, ADSL
modems, digital cameras, video conferencing
– High-cost, high-throughput: high-end printers,
audio mixing boards, wireless basestations,
high-end video conferencing, 3-D sonar,
3-D medical reconstruction from 2-D X-rays
THANK YOU