View
229
Download
0
Tags:
Embed Size (px)
Citation preview
Professor A G
Constantinides© 1
AGC
DSP
AGC
DSP
Digital Signal Processing & Digital Signal Processing & Digital FiltersDigital Filters
An Introductory CourseBy
Professor A G ConstantinidesMSc, EE4, ISE4, PhD
Professor A G
Constantinides© 2
AGC
DSP
AGC
DSP
Digital Signal Processing & Digital Signal Processing & Digital FiltersDigital Filters
Contents
1-Introduction1) Introduction to Digital Signal Processing Review of background DSP Review of mathematical methods Review of discrete-time random
processes and linear systems
Professor A G
Constantinides© 3
AGC
DSP
AGC
DSP
Digital Signal Processing & Digital Signal Processing & Digital FiltersDigital Filters
2) Multirate techniques and wavelets Introduction to short-time Fourier analysis Filter-banks and overlap-add methods of analysis
and synthesis Introduction to generalised time-frequency
representation Wavelet analysis Multirate signal processing Interpolation and decimation Efficient filter structures for interpolation and
decimation
Professor A G
Constantinides© 4
AGC
DSP
AGC
DSP
Digital Signal Processing & Digital Signal Processing & Digital FiltersDigital Filters
3) Classical spectrum estimation methods Power spectrum, power spectral density functions,
random processes and linear systems Introduction to statistical estimation and estimators Biased and unbiased estimators Einstein/Wiener Khintchine Theorem Estimation of autocorrelations Means and variances of periodograms Smoothed spectral estimates, leakage
Professor A G
Constantinides© 5
AGC
DSP
AGC
DSP
Digital Signal Processing & Digital Signal Processing & Digital FiltersDigital Filters
4) Modern spectrum estimation methods Introduction to modern spectral estimation:
Principles and approaches Cramer-Rao Lower Bound (CRLB) and Efficient
estimators The Maximum Entropy Method (MEM) or
Autoregressive Power Spectrum Estimation: Principles.
The MEM equations and Levinson/Durbin algorithm
Professor A G
Constantinides© 6
AGC
DSP
AGC
DSP
Digital Signal Processing & Digital Signal Processing & Digital FiltersDigital Filters
4) Modern spectrum estimation methods (continued)
Introduction to Linear Prediction Linear Predictive Coding using covariances
and correlations Cholesky decomposition Lattice Filters Linear Prediction of Speech Signals
Professor A G
Constantinides© 7
AGC
DSP
AGC
DSP
Digital Signal Processing & Digital Signal Processing & Digital FiltersDigital Filters
5) Adaptive signal processing Introduction to adaptive signal processing Objective measures of goodness Least squares and consequences Steepest descent The LMS and RLS algorithms Kalman Filters
Professor A G
Constantinides© 8
AGC
DSP
AGC
DSP
Digital Signal Processing & Digital Signal Processing & Digital FiltersDigital Filters
6) Applications Communications Biomedical Seismic Audio/Music
Professor A G
Constantinides© 9
AGC
DSP
AGC
DSP
DIGITAL FILTERSDIGITAL FILTERSDigital Filters In this course you will learn: How to choose an appropriate filter
response. Why Butterworth responses are
maximally flat. Why Chebyshev and Elliptic responses
are equiripple. When to choose an IIR and when an FIR
filter
Professor A G
Constantinides© 10
AGC
DSP
AGC
DSP
DIGITAL FILTERSDIGITAL FILTERS How do you design FIR and IIR filters from
specifications on amplitude performance? What are multirate systems and their
properties? What is interpolation / Upsampling and Decimation / Downsampling?
How do you design efficient Decimation and Interpolation systems?
What are frequency transformations and how do you design these?
How accurate is the DFT as a spectrum estimator?
Professor A G
Constantinides© 11
AGC
DSP
AGC
DSP
DIGITAL FILTERSDIGITAL FILTERS What are short FFT algorithms? How do you choose the required
wordlength? What are Fast Convolutions and
how are they realised? How do you deal with a DSP
problem in practice?
Professor A G
Constantinides© 12
AGC
DSP
AGC
DSP
Course contentCourse contentAssumed DSP background
DSP Background folder 1-Introduction 2-z transform 3-transfer functions 4-Signal Flow Graphs 5-digital filters intro
Professor A G
Constantinides© 13
AGC
DSP
AGC
DSP
Course contentCourse content
2-Digital Filter Design 1-Digital Filters (FIR) 2-Digital Filters (IIR)
3-Multirate1-Interpolation_Decimation
Professor A G
Constantinides© 14
AGC
DSP
AGC
DSP
Course contentCourse content
4-Tranforms 1-DFT 2-DFT_one2two 3-general transforms 4-Wavelets
5-Finite Wordlength 1-Finite Wordlength
Professor A G
Constantinides© 15
AGC
DSP
AGC
DSP
Course contentCourse content6-Spectrum Estimation (Assumed
background in Mathematical Background folder)
1-Fourier transform & DFT 2-FFT-based Power Spectrum Estimation 3-Modern Spectrum Estimation 4-Intro-Estimation 5-Eigen-based methods 6-A Prediction Problem
Professor A G
Constantinides© 16
AGC
DSP
AGC
DSP
Course contentCourse content
7-Adaptive Signal Processing 1-Adaptive Signal Processing
8-Applications 1-Applications 2-Applications
Professor A G
Constantinides© 17
AGC
DSP
AGC
DSP
Digital Signal Processing & Digital Signal Processing & Digital FiltersDigital Filters
BOOKS Main Course text books: Digital
Signal Processing: A computer Based Approach, S K Mitra, McGraw Hill
Mathematical Methods and Algorithms for Signal Processing, Todd Moon, Addison Wesley
Other books: Digital Signal Processing, Roberts &
Mullis, Addison Wesley Digital Filters, Antoniou, McGraw Hill
Professor A G
Constantinides© 18
AGC
DSP
AGC
DSP
DIGITAL FILTERSDIGITAL FILTERSAnalogue Vs Digital Signal Processing Reliability:Analogue system performance degrades
due to: Long term drift (ageing) Short term drift (temperature?) Sensitivity to voltage instability. Batch-to-Batch component variation. High discrete component count
Interconnection failures
Professor A G
Constantinides© 19
AGC
DSP
AGC
DSP
DIGITAL FILTERSDIGITAL FILTERSDigital Systems: No short or long term drifts. Relative immunity to minor power
supply variations. Virtually identical components. IC’s have > 15 year lifetime Development costs System changes at
design/development stage only software changes.
Digital system simulation is realistic.
Professor A G
Constantinides© 20
AGC
DSP
AGC
DSP
DIGITAL FILTERSDIGITAL FILTERS
Power aspects Size Dissipation
DSP chips available as well as ASIC/FPGA realisations
Professor A G
Constantinides© 21
AGC
DSP
AGC
DSP
ApplicationsApplications
Radar systems & Sonar systems Doppler filters. Clutter Suppression. Matched filters. Target tracking. Identification
Professor A G
Constantinides© 22
AGC
DSP
AGC
DSP
DIGITAL FILTERSDIGITAL FILTERS
Image Processing Image data compression. Image filtering. Image enhancement. Spectral Analysis. Scene Analysis / Pattern
recognition.
Professor A G
Constantinides© 23
AGC
DSP
AGC
DSP
DIGITAL FILTERSDIGITAL FILTERS
Biomedical Signal Analysis Spatial image enhancement. (X-
rays) Spectral Analysis. 3-D reconstruction from projections. Digital filtering and Data
compression.
Professor A G
Constantinides© 24
AGC
DSP
AGC
DSP
DIGITAL FILTERSDIGITAL FILTERS
Music Music recording. Multi-track “mixing”. CD and DAT. Filtering / Synthesis / Special
effects.
Professor A G
Constantinides© 25
AGC
DSP
AGC
DSP
DIGITAL FILTERSDIGITAL FILTERS
Seismic Signal Analysis
Bandpass Filtering for S/N improvement.
Predictive deconvolution to extract reverberation characteristics.
Optimal filtering. (Wiener and Kalman.)
Professor A G
Constantinides© 26
AGC
DSP
AGC
DSP
DIGITAL FILTERSDIGITAL FILTERSTelecommunications and Consumer
ProductsThese are the largest and most
pervasive applications of DSP and Digital Filtering
Mobile Communications Digital Recording Digital Cameras Blue Tooth or similar