Download - ADSP Lec 01
-
7/27/2019 ADSP Lec 01
1/52
Dr. Tahir Zaidi
Advanced Digital Signal ProcessingSpring 2012
Lecture 1Introduction
-
7/27/2019 ADSP Lec 01
2/52
DSP is Everywhere
Sound applications Compression, special effects, synthesis,
recognition, echo cancellation,
Cell Phones, MP3, Movies, Text-to-speech,
Communication Modulation, coding, detection, equalization, echo
cancellation,
Cell Phones, dial-up modem, DSL modem,Satellite Receiver,
Automotive ABS, Active Noise Cancellation, Cruise Control,
-
7/27/2019 ADSP Lec 01
3/52
DSP is Everywhere
Medical Magnetic Resonance, Tomography,Electrocardiogram,
Military
Radar, Sonar, Space photographs, remotesensing,
Image and Video Applications DVD, JPEG, Movie special effects, video
conferencing,
Mechanical Motor control, process control, oil and mineral
prospecting,
-
7/27/2019 ADSP Lec 01
4/52
Limitations of Analog Signal Processing
Accuracy limitations due to Component tolerances
Undesired nonlinearities
Limited repeatability due to Tolerances
Changes in environmental conditions
Temperature
Vibration
Sensitivity to electrical noise
-
7/27/2019 ADSP Lec 01
5/52
Limitations of Analog Signal Processing
Limited dynamic range for voltage andcurrents
Inflexibility to changes
Difficulty of implementing certainoperations
Nonlinear operations
Time-varying operations Difficulty of storing information
-
7/27/2019 ADSP Lec 01
6/52
Digital Signal Processing
A/D DSP D/Aanalogsignal
analogsignal
digitalsignal
digitalsignal
Analog input analog output
Digital recording of music
Analog input digital output
Touch tone phone dialing
Digital input analog output
Text to speech
Digital input digital output Compression of a file on computer
-
7/27/2019 ADSP Lec 01
7/52
Pros of Digital Signal Processing
Accuracy can be controlled by choosingword length
Repeatable
Sensitivity to electrical noise is minimal
Dynamic range can be controlled usingfloating point numbers
Flexibility can be achieved with software
implementations Non-linear and time-varying operations
are easier to implement
-
7/27/2019 ADSP Lec 01
8/52
Pros of Digital Signal Processing
Digital storage is cheap Digital information can be encrypted for
security
Price/performance and reduced time-to-
market
-
7/27/2019 ADSP Lec 01
9/52
Cons of Digital Signal Processing
Sampling causes loss of information A/D and D/A requires mixed-signal
hardware
Limited speed of processors
Quantization and round-off errors
-
7/27/2019 ADSP Lec 01
10/52
DSP Introduction
Application of mathematical operationsto digitally represented signals
IN OUT
A/D D/ADSP
-3 -2 -1 0 1 2 3 4
x[0]x[1]
n
-
7/27/2019 ADSP Lec 01
11/52
General IntroductionDiscrete Time Signal
sequence x[n]
- as opposed to continuous-timesignals x(t)
- time = independent variable
-
7/27/2019 ADSP Lec 01
12/52
ExamplesDiscrete in Nature
- stock market indices
NasDaq daily closing value from Aug 1995 to Jan 1996
- population statistics
Birth in Canada from 1995-1996 to 1999-2000
-
7/27/2019 ADSP Lec 01
13/52
Example
Sampled continuous-time (analog) signals
- Speech
-
7/27/2019 ADSP Lec 01
14/52
Digital Images
2-D arrays (matrices) of numbers
-
7/27/2019 ADSP Lec 01
15/52
Typical DSP Applications
Digital
RadiographicImaging
UltrasoundMedicalImaging
SpySatelliteImagingMilitaryAppls
Real Time
Video Cameras& Cell Phones
VideoCommunications
Space
ImagingAppls
OpticalWearableComputers
Web wireless
technology
Data Storage
& Transmission
Car Awake warning system
RealTime DSP
EmbeddedSystems
Speech
Recognition
-
7/27/2019 ADSP Lec 01
16/52
Example: Speech Modeling
Impulse
TrainGenerator
Noise
Generator
Pitch
Period
u(n)
Time-
varying
digital
filter
Vocal Tract
Parameters
s(n)
G
-
7/27/2019 ADSP Lec 01
17/52
An Embedded SystemControl Panel
PROGRAMMABLE
DSP
PROGRAMMABLE
DSP
ASIC
FPGA
MICROCONTROLLER
CODEC
Dual Port Memeory
System Bus
Controller Process
User interface
process
DSP
Assembly
Code
Analog
interface
Real Time
Operatingsystem
Embedded signal
Processing System
Host port
Memory interface
Host port
Memory interface
-
7/27/2019 ADSP Lec 01
18/52
Example Embedded System
Output
Bitstream
49.152
MHz
Sine wave
clock
Xilinx 4062TMS320C6201
68332
SRAM
FLASH
SBSRAM
DDS
A/D
HSP52014
8-bit DAC &LPF
amplifier &squarer
I/Osquare waveoutput
To RF Board
From RF Board
-
7/27/2019 ADSP Lec 01
19/52
SDR Board Design
FPGASPARTAN3
XC3S1500FG676I
- XC3S2000FG676I
VCCINT=1.2V/470mA
VCCAUX=2.5V/100mAVCCO1=3.3V/mA
VCCO2=2.5V/mA
AD9640DUAL ADC
14BIT, 105 MSPS
AVDD=1.8V/310mA
DVDD=1.8V/34mADRVDD=3.3V/35mA
GC5016
Quad Wideband DUC/DDC
VPAD=3.3V/180mAVCORE=1.8V/420mA
DUAL Channel
14 bit ,
125 MSPS (Max)DAC,
DAC2904,
VA=3.3V/64mAVD=3.3V/19.5mA
RS232 Interface DB9
DSPTMS320DM6446
CVDD 1.2V/767mA
DVDD 1.8V/102mADVDD 3.3V/6mA
32
47
IN
AD8352Differential
Amp
AMP
FILTERNETWORK
Not
implemented
IN
POWER
IN
HPI / VLYNQ
interfaceLVCMOS_1.8V
32BIT
JTAG
Title: Tranceiver BoardSize: A Revision: 1.3
Date: 08/04/08 Drawn by: ASK
RSSI
AnalogInterface
8 Channel ADC
MCP3008VD=3.3V/0.5mA
4-Bit
RS232 TRANSCEIVER
MAX3232EID
I-Input
Q-Input
I-Output
Q-Output
167
Clock
GeneratorAD9513
3 outputs
GAIN CONTROL (6-BIT)
PAinterface
6-Bits Output power control
FilterSelection
3-Bit Rx Filter Selection
HMC610
RSSIx2
1-Bit T/R Control
5-Bit Frequency controlSythesizerInterface
T/R Switch
/2
2x MT47H64M16BT-5E
1G DDR SDRAM
64M x 321.8VD/mA?
OSC
EthernetInterface
RJ45
Ethernet PHY
DP83848IIOVDD=3.3V/150mA
AVDD=3.3V/100mA?
20
Digital Power(SMPS)
1.2VD
1.8VD2.5VD
3.3VD
Analog
(LDO Linear PSU)
1.8VA3.3VA
PLATFORMFLASH
XCF08P 3.3VD/20mA
28F256J3, 128Mb16MB Intel Strata flash
3.3V/80mA
JTAGEXP
HEADER16-32 IO
64-LFCSP_VQ
SOIC-16
TQFP-48
PBGA-252
FG-676 (BGA) FSG-48 (BGA)
PBGA-N361
LQFP-48
SPI
IN
IN
IN
16-LFCSP_VQ
SOIC-16
Spartan3
SUPPORTSLVCMOS-1.8
AUDIO SERIAL PORT
ASP HEADER
SSN
Silicon Serial Number
-
7/27/2019 ADSP Lec 01
20/52
Device 0
DataData
Waveform 1
Software Defined RadioAll configurable HW
FPGA
Device 4
Device 1
DSP
General Purpose Processor
Algo4 Proprietary
FECFramer 1 V.35
16QAM
OFDM
-
7/27/2019 ADSP Lec 01
21/52
SDR Platform
Key Features1. DSP core from TI
2. FPGA from Xilinx
3. Dual-channel analog-to-digital
converter
4. Dual-channel digital-to-analogconverter
5. Bandwidth (5 MHz or 20 MHz)
6. RF module operating between 360
MHz and 960 MHz
7. Ethernet remote access capabilities
8. ARM Processor
Design Options
1. Tactical military communications
2. Military communication gateways
3. Handset and man pack systems
4. Vehicular systems
C Obj i
-
7/27/2019 ADSP Lec 01
22/52
Course Objectives
To establish the idea of using computingtechniques to alter the properties of a signalfor desired effects, via understanding of Fundamentals of discrete-time, linear, shift-
invariant signals and systems in Representation and Analysis:sampling, quantization,
Fourier and z-transform;
Implementation:filtering and transform techniques;
System Design:filter & processing algorithm design.
Efficient computational algorithms and theirimplementation.
C O li
-
7/27/2019 ADSP Lec 01
23/52
Course Outline
C O li
-
7/27/2019 ADSP Lec 01
24/52
Course Outline
P i it
-
7/27/2019 ADSP Lec 01
25/52
Prerequisite
A fundamental course in signal andsystem
Liner System analysis and transformanalysis
convolution and filtering
Fourier transforms
Laplace and z transforms
T tb k
-
7/27/2019 ADSP Lec 01
26/52
Textbooks
Oppenheim, Schafer and Buck,Discrete-Time Signal Processing, 2ndedition (Prentice-Hall, 1999)
Mathematics of DSP
Refrences: McClellan, Schafer, & Yoder, DSP First
Ifeachor Jervis Digital Signal Processing-
A Practical Approach, Prentice Hall
-
7/27/2019 ADSP Lec 01
27/52
Historical Perspective
Who is who of DSP
C l d T k
-
7/27/2019 ADSP Lec 01
28/52
Cooley and Tuckey
-
7/27/2019 ADSP Lec 01
29/52
In entors Oppenhiam Schaffer
-
7/27/2019 ADSP Lec 01
30/52
Inventors: Oppenhiam, Schaffer ...
Inventors: Parks & McCllelan
-
7/27/2019 ADSP Lec 01
31/52
Inventors: Parks & McCllelan
Inventors: Gold and Rader
-
7/27/2019 ADSP Lec 01
32/52
Inventors: Gold and Rader
Inventor: J Kaiser
-
7/27/2019 ADSP Lec 01
33/52
Inventor: J. Kaiser
-
7/27/2019 ADSP Lec 01
34/52
Inventor: Haskell
Linear Predictive Coding
-
7/27/2019 ADSP Lec 01
35/52
Original Speech
Analysis:
Voiced/Unvoiced decision
Pitch Period (voiced only)
Signal power (Gain)
G
Pulse Train
Random Noise
Vocal TractModel
V/U
Synthesized Speech
Decoder
Signal Power
Pitch
Period
Encoder
Linear Predictive Coding
Inventor: James G Dunn
-
7/27/2019 ADSP Lec 01
36/52
Inventor: James G. Dunn
DSP Components
-
7/27/2019 ADSP Lec 01
37/52
DSP Components
Mi
-
7/27/2019 ADSP Lec 01
38/52
Microprocessor
Any CPU that is contained on a singlechip
Little chip is the heart of a computer.Often referred to as just the processor
Does all the computations like adding,subtracting, multiplying, and dividing
In PCs, most popular Intel Pentium chip
In Macs, the PowerPC chip (Motorola, IBM,and Apple)
-
7/27/2019 ADSP Lec 01
39/52
Digital Signal Processor A DSP is a general purpose processor with
features specifically designed to make Signalprocessing applications fast and efficient
-
7/27/2019 ADSP Lec 01
40/52
DSP, RISC, CISC Processor
A processor is frequently categorizedbased on the width of its busses(4,8,16,32,64)
Clock Rate (i.e. at what rate does the
processor execute instructions) Complexity of Instruction Set
CISC : Complex Instruction Set
ComputerRISC : Reduced Instruction SetComputer
-
7/27/2019 ADSP Lec 01
41/52
Embedded Systems Characteristics Real-Time
Real, defined timing requirements for particular
actions to be accomplished
Event DrivenActions of the system are in response to events,not a predefined sequence.
Resource constrainedMemory Size, speed, power constrained
Special purposeDevice must only perform certain well defined
tasks
-
7/27/2019 ADSP Lec 01
42/52
Embedded System Example Events :
Button Press
Knob Turned
New Sample needed
by D/A converter
Data block availablefrom CD drive
-
7/27/2019 ADSP Lec 01
43/52
Design Options for Digital Systems Special Purpose Hardware
Custom ICs / ASICs
Software Programmable ProcessorPentium, PowerPC, etc
FPGA possibly with embeddedgeneral purpose microprocessor)Xilinx, Altera, etc
DSPTI, ADSP, etc
-
7/27/2019 ADSP Lec 01
44/52
Comparison of OptionsSpecific HW Gen Purpose HW
NRE/Dev Cost
Speed
Flexibility
Time to Market
Production Cost
-
7/27/2019 ADSP Lec 01
45/52
Embedded SW Design Flow Develop Code for a Target processor
Since target is minimal (not muchmemory, I/Oetc. Code developmentdone on a separate machine. (e.g a PC)
Cross Compiler / Assembler Simulator
Code then run in the target system andobserved. Debug support programmedinto the software
-
7/27/2019 ADSP Lec 01
46/52
Emulation / Debugging In-Circuit Emulator
Debug Kernel BIOS
JTAG Emulation
Interactively Run Code
Breakpoints
Single Step
Watch Variables
Observe interaction with rest of system Development environment is frequently
processor specific
-
7/27/2019 ADSP Lec 01
47/52
TI TMS320C6713 DSP
-
7/27/2019 ADSP Lec 01
48/52
TI TMS320C6713 DSP Features
DMA Controller Serial Ports (I/O)
Multiple Computation Units
Cache On-chip PLL
Host Port Interface
Timers
Floating Point Units
-
7/27/2019 ADSP Lec 01
49/52
Basic Numbering Formats
Three main numbering formats:unsigned representation
2s complement representation (signed)
floating point representations Fixed point representations of
fractions
Saturating arithmeticMultiplication of fractions
-
7/27/2019 ADSP Lec 01
50/52
Basic Numbering Formats
-
7/27/2019 ADSP Lec 01
51/52
-
7/27/2019 ADSP Lec 01
52/52