Welcome to
“Digital Signal Processing”
About this course
Digital Signal Processing (DSP, No.620102)
2 credits
40 hours, 4 each week, 10 weeks
Homework, final exam
Important course:
DSP is a very important course, and more and more
important.
DSP is important both in theoretical and engineering
areas.
Instructor
Assoc. Prof. Jing-Ran Lin （林静然）
Dept. Communication Engineering, SCIE, UESTC
Office: Room 244, District-B of Sci.& Res. Building
E-mail:
[email protected] Tel: 028-61831246
Digital Signal Processing
─ Introduction
Outline
Pre-requirements: Related Courses and
Background
Textbook and References
What is DSP?
Why Digital?
Applications of DSP Technology
Pre-requirements
The DSP related Tech. Domains
Communication, Radio, Radar, Navigation, IT, ... ...
Related Courses:
“Signals and Systems”, very important!
“Advanced Mathematics”, “Linear Algebra”
“Programming Language”
“The Principle of Micro-computer”, “Logic Circuit”,
“The Technique of DSP”
……
Textbook
Digital Signal Processing
─ A Computer-Based
Approach (3rd/4th Edition)
Sajit K. Mitra
McGraw Hill, 2012.8
Reference books
1. A.V.Oppenheim and R.W.Schafer, “Discrete Time Signal
Processing”, Pearson Education, 2007
2. Sen M. Kuo and Woon-Seng Gan, “Digital Signal Processing ─
Architectures, Implementations, and Applications”, Pearson
Education, 2003
3. G. Marven and G. Ewers, “A Simple Approach to Digital Signal
Processing”, Wiley, 1996
4. Saeed V. Vaseghi, “Advanced Digital Signal Processing and
Noise Reduction”, Wiley, 2009
5. Richard Newbold, “Practical Applications in Digital Signal
Processing”, Prentice Hall, 2012
What is DSP?
Digital Signal Processing
─ Theories, methods and algorithms about how to
process signals in digital form;
Digital Signal Processor
─ A kind of microprocessor used to implement digital
signal processing algorithms;
─ Architecture (Chips, Peripherals, Pipelines,
Instructions, Circuits, … … )
─ System (Hardware & Software Design and Debug …
…)
What is DSP?
DSP simply means processing signals in digital form.
Analog v.s. Digital
Analog Continuous in both time and magnitude
Digital Discrete in both time and magnitude
Most signals in real world are analog!
Needs some transformations when performing DSP
Step 1: Sample the analog signal at some discrete time points;
Step 2: Quantize the sample values either by rounding or truncation
DSP Solution
Why DSP is important?
The foundation of information technology is digitization.
The kernel of digitization is digital signal processing.
Most digital signal processing, especially real-time
processing are implemented by DSP processors or
ASICs based on DSP cores.
DSP technology becomes hot front edge and driver for
almost the entire semiconductor industry.
Why Digital?
Advantages over ASP:
More reliable, less sensitive to tolerances such as
temperature, noise, etc.;
Higher accuracy, can be integrated on a single chip;
Limit:
A/D D/A rates are not available in some applications;
Why Digital?
(1) Programmability:
Analog systems: Modify hardware design
Digital systems: Modify software
Examples:
Analog filter, digital filter, adaptive filter, … …
Why Digital?
(2) Precision:
Analog system, by components specification
Resistors have a tolerance of 5%;
Capacitors 20% or worse;
Digital system, by ADC bits, CPU word width (or
word-length) and algorithm;
Why Digital?
(3) Stability:
Analog system:
The characteristics of analog system components (e.g.,
resistors, capacitors and operational amplifiers) will
change along with temperature, humidity, etc.
Digital system:
Shows no variation with temperature throughout their
guaranteed operation range.
Why Digital?
(4) Anti-noise:
Why Digital?
(5) Repeatability:
Amplifier A computer A
Signal
generator
Amplifier B computer B
Why Digital?
(6) VLSI (Very Large Scale Integrated Circuit):
Why Digital?
(7) Error Correcting Codes:
Data retrieval and transmission systems suffer from
a number of potential forms of error.
With information in a digital or binary form, we may
easily build into the data stream additional
“redundant bits” that are used to detect when an
error has occurred.
Why Digital?
(8) Data Transmission and Storage:
The fidelity of the digital medium is greater than that
of the analog one.
The Internet, Compact Disc (CD) and Digital Video
Disc (DVD) brought information based on trouble-
free high-quality text, audio and video into office and
home.
Why Digital?
(9) Data Compression:
The information channels cost and transmission
bottlenecks make compression necessary for real-
time processing.
With analog compression some information lost.
A example is the bandwidth limiting applied to analog
telephone lines, which limits the bandwidth to 3kHz.
Digital tech. makes lossless compression possible.
We still need analog processing!
(1) Real-Time Processing:
Analog system ：
Besides the delay introduced by the circuit, the
processing is in real-time.
Digital system ：
Processing time is determined by the processor speed
and the adopted algorithms.
We still need analog processing!
(2) Processing very high frequency signals:
Analog system ：
may process microwave, mini-meter-wave, even light
wave signals.
Digital system ：
By the Nyquist Rule, the processing is limited by the S/H,
A/D and processor speed.
ADs with very high sampling rates are extremely
expensive!
We still need analog processing!
(3) Most signals in real world are analog
If we want to process these analog signals with
digital signal processing system, we must change
them into digital form first by mixed signal
processing.
Applications of DSP Technology
Communication
Wireless Communication
(Base stations and Mobile phones)
Fashion Cell-Phone
Diagram of GSM Cell-Phone
2G (GSM) 3G (CDMA)
Satellite Navigation
Satellite Navigation
GPS
(Global Positioning System)
GPS
(Global Positioning System)
GIS
(Geography Information System)
Satellite Navigation
System Testing
BD
(BeiDou or Compass)
Test in Zhuhai
Test in ChengDu
Measurement and Control
Virtual Instruments
(HP89441 Vector Analyzer)
Shenzhou-X docked with Tiangong-I
Automotive Electronics
Military and Safety
Air Early Warning (AEW)
Radar/Sonar
Cruise Missile
Smart Bomb
Pilotless Aircraft
UAV (Unmanned Aerial Vehicle)
Consumer Electronics
DC (Digital Camera)
DV (Digital Video)
MP3
(MPEG Level-3 Audio Decoder)
HDTV (High Definition TV)
Home Theater
IPTV and IP Phone
Medical
Ultrasound
MRI
(Magnetic Resonance Imaging)
CT (Computed Tomography)
Gamma knife
Hearing Aids
Others ……
Drowsy Driving Alert System
Digital Watermarking
Video Surveillance System
Fingerprint distinguish
Pattern Recognition
Image & Video Reparation
Topics
Review of the continuous signal and systems
Discrete signal and systems in time domain
Discrete signal and systems in transform domain
Digital Processing of Continuous-Time Signals
Digital Filters
– Digital Filter Structures
– Digital Filter Design (FIR and IIR)
Analysis of Finite Word-length Effects
Multi-rate Digital Signal Processing
Jing-Ran Lin
B244, Sci. & Res. Building
Tel: 028-61831246
Email: [email protected]
