2nd Half of UNIT 6 --- DSP Processor

BY
Prof L.S.Kalkonde
Department of Electronics & Telecommunication
Prof Ram Meghe College of Engineering & Technology,Badnera
 Digital Signal Processor---- Definition

A digital signal processor (DSP) is an integrated

circuit designed for high-speed data manipulations,
and is used in
Audio
Communications
image manipulation
Other data-acquisition and
Data-control applications.
 How Digital Signal Processing Works

1. To explain how digital signal processing works, you must

understand the difference between analog and digital signals.
2. Analog signals, which include sound intensity, pressure, light
intensity, etc., are continuously variable.

3. Each of our senses is sensitive to different kinds of analog

signals.

4. Our ears are sensitive to sound, our eyes are sensitive to light,
and so on.

5. Once we receive a signal, our sensory organs convert it to an

electrical signal and send it to our analog computer (the brain).
 How Digital Signal Processing Works

6. Our brains are very powerful parallel computer whose

performance currently is unmatched by any digital computer.

7. Our brains not only analyze the information received, but also
make decisions using this data.

8. Digital signals are those that are transmitted within or

between computers, in which information is represented by
discrete states –
How Analog and Digital Signals Work Together
 How Analog and Digital Signals Work Together

Digital technology such as personal computers (PCs), assist us in

many ways: writing documents, spell checking, and drawing.

Unfortunately, the world is analog, and electronic analog

computers are not as versatile as digital computers.

Therefore, in order to make use of the tremendous processing

power that digital technology offers us, we must do the following:

Convert the analog signals into electrical signals, using a transducer

(such as a microphone, as shown in the diagram).

· Digitize these signals (i.e., convert them from analog to digital using an
analog-to-digital converter (ADC)), as shown in the diagram.
Why Do We Need Digital Signal Processors?
 Why Do We Need Digital Signal Processors?

Add and Subtract

Add and subtract operations are performed quite simply by general-purpose

microprocessors in a single or very few clock cycles. Digital addition is similar to
decimal add. Our example shows adding 1 plus 2. The result is the decimal 3.

Multiply and Divide

The multiply and divide operations are more complex. A digital multiply operation
consists of a series of shift and add operations. example shows a multiplication of 3 &
5. General-purpose microprocessors are quite slow in performing multiply and divide
operations. They will typically sequentially execute a series of shift, add, and subtract
operations from their microcode i.e.
to perform a single multiply operation, it may consume many cycles
to complete
The DSP performs multiplication in a single cycle by implementing all
shift and add operations in parallel.
What’s Inside DSP (Elements of 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
Types of Architecture

Super/ Modified
Harvard Harvard Architecture
Architecture

Von
Neumann
Architecture
Von Neumann Architecture

Memory Address Bus

Instruction
CPU
&
Data Data Bus
 Harvard Architecture

Address
Address Bus
Bus

Program Data
Memory CPU Memory
Data Bus Data Bus
 Which Architecture is Best Suited for DSP?

1. Common general-purpose personal computers use processors designed with the

von Neuman architecture while the Harvard architecture is more commonly used in
specialized microprocessors for real-time and embedded applications.

2. DSPs typically use Harvard architecture, although von Neuman DSPs also exist.

3. Many signal and image processing applications require fast, real-time machines.

4. The drawback to using a true Harvard architecture is that since it uses separate
program and data memories, it needs twice as many address and data pins on the chip
and twice as much external memory. Unfortunately, as the number of pins or chips
increases, so does the price.
 Which Architecture is Best Suited for DSP?

An elegant solution:

A single data and address bus is used externally.

Two (or more) separate buses for program and data are used internally.

Timing (multiplexing) handles the separation of program and data information.

In one clock cycle, the program information flows on the pins, and
In the second cycle, data follows on the same pins.
Program and data information is then routed onto separate internal program and
data buses. Such machines are called modified Harvard architecture processors
because
the internal architecture is Harvard
external architecture is von Neuman.

Also Multiple internal RAM/ROM cells for high-use instructions and data.
 Fixed vs. Floating Point

Characteristic Floating point Fixed point

32-bit 16-bit
Dynamic range much larger smaller
Resolution comparable comparable
Ease of programming comparable comparable
Compiler efficiency much easier more difficult
Power consumption more efficient less efficient
Chip cost comparable comparable
System cost comparable comparable
Design cost less more
Time to market faster slower
 TMS320 Family

16-Bit Fixed Point Devices 32-Bit Floating Point Devices

C5x C6x Advanced VLIW
Voice Processing Processor
C54x Wireless Base
Digital Cellular Phones Stations/Pooled
Modems
TMS32054XX
 Features of TMS32054XX

• 16 bit CPU
• Can execute 40 to 120 Million Instructions Per Second
• 17×17 bit MAC
• 64k × 16 bit physical program memory address space
• 64k × 16 bit external data memory address space
• 64k × 16 bit external IO address space
• Programmable timer & PLL
• DMA interface
• 100/128/144 TQFP & BGA packages
 Functional Units

• 40 bit ALU
• 2- 40 bit accumulators ACCA & ACCB
• Barrel shifter
• 17X17 bit multiplier
• 40 bit adder
• CSSU-Compare, Select & store unit
• Exponent Encoder
• Data Address generation
• Program & address generation unit
 TMS32054XX
•Uses an advanced , Modified Harvard
architecture
•Maximizes processing power by providing

4 pairs
Bus Structure

3 Pairs 1 Pair
Data Memory Program Memory
 ALU
• 40 Bit ALU
• Wide range of Arithmetic & Logic Operation in
single clock cycle.
• After ALU operation destination of result
– Accumulator or
– Memory
 Accumulators
• 40 bit ACCA & ACCB
• To store result for ALU & Multiply/Add.
• Temporary storage for other.
 Barrel Shifter
• The barrel shifter can produce a left shift of 0 to 31
bits and a right shift of 0 to 16 bits on the input data.
• The shift requirements are defined in
– the shift count field of the instruction, the shift count field
(ASM) of status register ST1, or
– In the temporary register T.
 Multiplier/Adder Unit
• The multiplier/adder block consists of several elements:

– a multiplier, an adder, signed/unsigned input

– control logic, fractional control logic,
– zero detector, a rounder , overflow/saturation logic
– and a 16-bit temporary storage register (T).

• The multiplier/adder unit performs 17 x 17-bit 2s-

complement multiplication with a 40–bit addition in a
single instruction cycle
 CSSU- Compare, Select, and Store Unit
• The compare, select, and store unit (CSSU) performs
maximum comparisons between
• the accumulator’s high and low word, allows both
the test/control flag bit (TC) in status register ST0 and
the transition register (TRN) to keep their transition
histories.
 Exponent Encoder
• To implement floating point arithmetic in
fixed point processor require separation
of exponent & mantissa of the floating
point data.
 Data Address Generation Unit
• 2 Auxiliary Register Arithmetic Units
ARAU0 & ARAU1 (Address Generation for
indirect addressing mode i.e. increment,
decrement, indexing, bit reverse ,circular
addressing )
• 8- AR0 to AR7 (To generate 2 data
memory address simult.)
 JTAG-Joint Test Action Group

JTAG, as defined by the IEEE Std.-1149.1 standard.

An integrated method for testing interconnects on printed circuit

boards (PCBs) that are implemented at the integrated circuit (IC) level.

The JTAG test architecture provides a means to test interconnects

between integrated circuits on a board without using physical test
probes.

Potential benefits from JTAG

Shorter test times,
Higher test coverage,
Increased diagnostic capability and
Lower capital equipment cost.
Instruction Pipelining in TMS320C54X Processors

1.Program Pre fetch

PAB is loaded with the address next instruction to be fetched

2. Program Fetch

The op-code is fetched from PB & loaded into Instruction Register

3.Decode
The opcode is decoded to determiine access operation
Instruction Pipelining in TMS320C54X Processors

4.Access
Operand address is loaded on data DAB – Data Address Bus. If 2nd operand
is required , then another address is loaded into CAB

5. Read

The operands are read from the buses DB & CB

6.Execute
Perform the task specified by the instruction
Sr no Parameter DSP Processor GPP Processor

1 Instruction Cycle Single Cycle Multiple instruction cycle for one

( i.e., true instruction cycle) instruction
2 Instruction Execution Parallel execution is possible Always sequential execution is possible

3 Operand fetched from Multiple operands are fetch simultaneously Operands are fetch sequentially
memory
4 Memories Separate program memory and data Normally no such separate memories are
memory present
5 On-chip/off-chip Program memory and data memory are Normally on-chip cache memory is present
memories present on-chip and expandable off-chip. .Main memory is off-chip.
6 Address generation Addresses are generated combinely by Program counter is incremented
DAGs and program sequencer. sequentially to generate addresses.
7 Address/data bus Address and data buses are not multiplexed. Address/data buses can be separate on the
multiplexing They are separate on chip as well as off chip. chip but usually multiplexed off-chip.

8 Computational units Three separate computational units: ALU is the main computational unit.
ALU,MAC and shifter.
9 Suitable for Array processing operations Genral purpose processing
10 Queuing/Pipelining Queuing is implemented through instruction Queuing is performed explicitly by queuing
register and instruction cache register for pipelining of instructions