Enviar pesquisa
Carregar
quantization
•
1 gostou
•
1,637 visualizações
Aniruddh Tyagi
Seguir
Tecnologia
Diversão e humor
Denunciar
Compartilhar
Denunciar
Compartilhar
1 de 33
Baixar agora
Baixar para ler offline
Recomendados
quantization
quantization
aniruddh Tyagi
Quantization
Quantization
wtyru1989
Sparsity and Compressed Sensing
Sparsity and Compressed Sensing
Gabriel Peyré
Discrete Fourier Transform
Discrete Fourier Transform
Shahryar Ali
WAVELET-PACKET-BASED ADAPTIVE ALGORITHM FOR SPARSE IMPULSE RESPONSE IDENTIFI...
WAVELET-PACKET-BASED ADAPTIVE ALGORITHM FOR SPARSE IMPULSE RESPONSE IDENTIFI...
bermudez_jcm
A Classification Problem of Credit Risk Rating Investigated and Solved by Opt...
A Classification Problem of Credit Risk Rating Investigated and Solved by Opt...
SSA KPI
Decimation in time and frequency
Decimation in time and frequency
SARITHA REDDY
An Introduction to Hidden Markov Model
An Introduction to Hidden Markov Model
Shih-Hsiang Lin
Recomendados
quantization
quantization
aniruddh Tyagi
Quantization
Quantization
wtyru1989
Sparsity and Compressed Sensing
Sparsity and Compressed Sensing
Gabriel Peyré
Discrete Fourier Transform
Discrete Fourier Transform
Shahryar Ali
WAVELET-PACKET-BASED ADAPTIVE ALGORITHM FOR SPARSE IMPULSE RESPONSE IDENTIFI...
WAVELET-PACKET-BASED ADAPTIVE ALGORITHM FOR SPARSE IMPULSE RESPONSE IDENTIFI...
bermudez_jcm
A Classification Problem of Credit Risk Rating Investigated and Solved by Opt...
A Classification Problem of Credit Risk Rating Investigated and Solved by Opt...
SSA KPI
Decimation in time and frequency
Decimation in time and frequency
SARITHA REDDY
An Introduction to Hidden Markov Model
An Introduction to Hidden Markov Model
Shih-Hsiang Lin
Hmm viterbi
Hmm viterbi
Digvijay Singh
Fft presentation
Fft presentation
ilker Şin
fourier representation of signal and systems
fourier representation of signal and systems
Sugeng Widodo
Correlative level coding
Correlative level coding
srkrishna341
Two dimensional Pool Boiling
Two dimensional Pool Boiling
RobvanGils
Dsp U Lec07 Realization Of Discrete Time Systems
Dsp U Lec07 Realization Of Discrete Time Systems
taha25
Isi and nyquist criterion
Isi and nyquist criterion
srkrishna341
Elementary Landscape Decomposition of the Quadratic Assignment Problem
Elementary Landscape Decomposition of the Quadratic Assignment Problem
jfrchicanog
Dsp U Lec04 Discrete Time Signals & Systems
Dsp U Lec04 Discrete Time Signals & Systems
taha25
L. Perivolaropoulos, Topological Quintessence
L. Perivolaropoulos, Topological Quintessence
SEENET-MTP
Non newtonian models
Non newtonian models
MARIO CERDA CORTES
Matched filter
Matched filter
srkrishna341
Hidden Markov Models
Hidden Markov Models
Vu Pham
Mapping Ash Tree Colonization in an Agricultural Moutain Landscape_ Investiga...
Mapping Ash Tree Colonization in an Agricultural Moutain Landscape_ Investiga...
grssieee
Cognitive radio
Cognitive radio
Rached Abdel
Dsp U Lec06 The Z Transform And Its Application
Dsp U Lec06 The Z Transform And Its Application
taha25
Alba ffs flim 2012
Alba ffs flim 2012
benbarbieri
z transforms
z transforms
Shahbaz Goshtasebi
Signal fundamentals
Signal fundamentals
Lalit Kanoje
Solved problems
Solved problems
Mulugeta Ashango
Quantization encodingexample
Quantization encodingexample
Mody Farouk
PCM (Pulse Code Modulation)
PCM (Pulse Code Modulation)
Syauqina Idzni Adzhani
Mais conteúdo relacionado
Mais procurados
Hmm viterbi
Hmm viterbi
Digvijay Singh
Fft presentation
Fft presentation
ilker Şin
fourier representation of signal and systems
fourier representation of signal and systems
Sugeng Widodo
Correlative level coding
Correlative level coding
srkrishna341
Two dimensional Pool Boiling
Two dimensional Pool Boiling
RobvanGils
Dsp U Lec07 Realization Of Discrete Time Systems
Dsp U Lec07 Realization Of Discrete Time Systems
taha25
Isi and nyquist criterion
Isi and nyquist criterion
srkrishna341
Elementary Landscape Decomposition of the Quadratic Assignment Problem
Elementary Landscape Decomposition of the Quadratic Assignment Problem
jfrchicanog
Dsp U Lec04 Discrete Time Signals & Systems
Dsp U Lec04 Discrete Time Signals & Systems
taha25
L. Perivolaropoulos, Topological Quintessence
L. Perivolaropoulos, Topological Quintessence
SEENET-MTP
Non newtonian models
Non newtonian models
MARIO CERDA CORTES
Matched filter
Matched filter
srkrishna341
Hidden Markov Models
Hidden Markov Models
Vu Pham
Mapping Ash Tree Colonization in an Agricultural Moutain Landscape_ Investiga...
Mapping Ash Tree Colonization in an Agricultural Moutain Landscape_ Investiga...
grssieee
Cognitive radio
Cognitive radio
Rached Abdel
Dsp U Lec06 The Z Transform And Its Application
Dsp U Lec06 The Z Transform And Its Application
taha25
Alba ffs flim 2012
Alba ffs flim 2012
benbarbieri
z transforms
z transforms
Shahbaz Goshtasebi
Signal fundamentals
Signal fundamentals
Lalit Kanoje
Solved problems
Solved problems
Mulugeta Ashango
Mais procurados
(20)
Hmm viterbi
Hmm viterbi
Fft presentation
Fft presentation
fourier representation of signal and systems
fourier representation of signal and systems
Correlative level coding
Correlative level coding
Two dimensional Pool Boiling
Two dimensional Pool Boiling
Dsp U Lec07 Realization Of Discrete Time Systems
Dsp U Lec07 Realization Of Discrete Time Systems
Isi and nyquist criterion
Isi and nyquist criterion
Elementary Landscape Decomposition of the Quadratic Assignment Problem
Elementary Landscape Decomposition of the Quadratic Assignment Problem
Dsp U Lec04 Discrete Time Signals & Systems
Dsp U Lec04 Discrete Time Signals & Systems
L. Perivolaropoulos, Topological Quintessence
L. Perivolaropoulos, Topological Quintessence
Non newtonian models
Non newtonian models
Matched filter
Matched filter
Hidden Markov Models
Hidden Markov Models
Mapping Ash Tree Colonization in an Agricultural Moutain Landscape_ Investiga...
Mapping Ash Tree Colonization in an Agricultural Moutain Landscape_ Investiga...
Cognitive radio
Cognitive radio
Dsp U Lec06 The Z Transform And Its Application
Dsp U Lec06 The Z Transform And Its Application
Alba ffs flim 2012
Alba ffs flim 2012
z transforms
z transforms
Signal fundamentals
Signal fundamentals
Solved problems
Solved problems
Destaque
Quantization encodingexample
Quantization encodingexample
Mody Farouk
PCM (Pulse Code Modulation)
PCM (Pulse Code Modulation)
Syauqina Idzni Adzhani
Pcm
Pcm
srkrishna341
PWM
PWM
mulluraniket
Quantization
Quantization
Maj. Sanjaya Prasad
Sampling
Sampling
Muhammad Uzair Rasheed
Pulse code modulation
Pulse code modulation
Naveen Sihag
PULSE WIDTH MODULATION &DEMODULATION
PULSE WIDTH MODULATION &DEMODULATION
bharath405
Sampling theory
Sampling theory
DIPTENDU BASU
pulse modulation
pulse modulation
Herin Gala
PULSE CODE MODULATION (PCM)
PULSE CODE MODULATION (PCM)
vishnudharan11
Pulse code modulation
Pulse code modulation
Abhijay Sisodia
Pulse modulation
Pulse modulation
stk_gpg
Pulse modulation
Pulse modulation
A.Q Khan Institiute Of Technology
Destaque
(14)
Quantization encodingexample
Quantization encodingexample
PCM (Pulse Code Modulation)
PCM (Pulse Code Modulation)
Pcm
Pcm
PWM
PWM
Quantization
Quantization
Sampling
Sampling
Pulse code modulation
Pulse code modulation
PULSE WIDTH MODULATION &DEMODULATION
PULSE WIDTH MODULATION &DEMODULATION
Sampling theory
Sampling theory
pulse modulation
pulse modulation
PULSE CODE MODULATION (PCM)
PULSE CODE MODULATION (PCM)
Pulse code modulation
Pulse code modulation
Pulse modulation
Pulse modulation
Pulse modulation
Pulse modulation
Semelhante a quantization
Quantifying the call blending balance in two way communication retrial queues...
Quantifying the call blending balance in two way communication retrial queues...
wrogiest
Quadrature amplitude modulation qam transmitter
Quadrature amplitude modulation qam transmitter
Asaad Drake
Cosmological Perturbations and Numerical Simulations
Cosmological Perturbations and Numerical Simulations
Ian Huston
An automated and user-friendly optical tweezers for biomolecular investigat...
An automated and user-friendly optical tweezers for biomolecular investigat...
Dr. Pranav Rathi
Finite automata examples
Finite automata examples
ankitamakin
Finite automata examples
Finite automata examples
ankitamakin
Ee443 phase locked loop - presentation - schwappach and brandy
Ee443 phase locked loop - presentation - schwappach and brandy
Loren Schwappach
EC8553 Discrete time signal processing
EC8553 Discrete time signal processing
ssuser2797e4
Fourier transform
Fourier transform
Naveen Sihag
Chapter 5
Chapter 5
Imran Abdullah
Aquifer Parameter Estimation
Aquifer Parameter Estimation
C. P. Kumar
SOCG: Linear-Size Approximations to the Vietoris-Rips Filtration
SOCG: Linear-Size Approximations to the Vietoris-Rips Filtration
Don Sheehy
Ultrasound Modular Architecture
Ultrasound Modular Architecture
Jose Miguel Moreno
Short-time homomorphic wavelet estimation
Short-time homomorphic wavelet estimation
UT Technology
Signal Processing Course : Sparse Regularization of Inverse Problems
Signal Processing Course : Sparse Regularization of Inverse Problems
Gabriel Peyré
Dissertation Slides
Dissertation Slides
Sanghui Ahn
CVPR2010: Sparse Coding and Dictionary Learning for Image Analysis: Part 1: S...
CVPR2010: Sparse Coding and Dictionary Learning for Image Analysis: Part 1: S...
zukun
Tuto cvpr part1
Tuto cvpr part1
zukun
IGARSS2011 FR3.T08.3 BenDavid.pdf
IGARSS2011 FR3.T08.3 BenDavid.pdf
grssieee
Waveform_codingUNIT-II_DC_-PPT.pptx
Waveform_codingUNIT-II_DC_-PPT.pptx
KIRUTHIKAAR2
Semelhante a quantization
(20)
Quantifying the call blending balance in two way communication retrial queues...
Quantifying the call blending balance in two way communication retrial queues...
Quadrature amplitude modulation qam transmitter
Quadrature amplitude modulation qam transmitter
Cosmological Perturbations and Numerical Simulations
Cosmological Perturbations and Numerical Simulations
An automated and user-friendly optical tweezers for biomolecular investigat...
An automated and user-friendly optical tweezers for biomolecular investigat...
Finite automata examples
Finite automata examples
Finite automata examples
Finite automata examples
Ee443 phase locked loop - presentation - schwappach and brandy
Ee443 phase locked loop - presentation - schwappach and brandy
EC8553 Discrete time signal processing
EC8553 Discrete time signal processing
Fourier transform
Fourier transform
Chapter 5
Chapter 5
Aquifer Parameter Estimation
Aquifer Parameter Estimation
SOCG: Linear-Size Approximations to the Vietoris-Rips Filtration
SOCG: Linear-Size Approximations to the Vietoris-Rips Filtration
Ultrasound Modular Architecture
Ultrasound Modular Architecture
Short-time homomorphic wavelet estimation
Short-time homomorphic wavelet estimation
Signal Processing Course : Sparse Regularization of Inverse Problems
Signal Processing Course : Sparse Regularization of Inverse Problems
Dissertation Slides
Dissertation Slides
CVPR2010: Sparse Coding and Dictionary Learning for Image Analysis: Part 1: S...
CVPR2010: Sparse Coding and Dictionary Learning for Image Analysis: Part 1: S...
Tuto cvpr part1
Tuto cvpr part1
IGARSS2011 FR3.T08.3 BenDavid.pdf
IGARSS2011 FR3.T08.3 BenDavid.pdf
Waveform_codingUNIT-II_DC_-PPT.pptx
Waveform_codingUNIT-II_DC_-PPT.pptx
Mais de Aniruddh Tyagi
security vulnerabilities of dvb chipsets
security vulnerabilities of dvb chipsets
Aniruddh Tyagi
whitepaper_mpeg-if_understanding_mpeg4
whitepaper_mpeg-if_understanding_mpeg4
Aniruddh Tyagi
BUC BLOCK UP CONVERTER
BUC BLOCK UP CONVERTER
Aniruddh Tyagi
digital_set_top_box2
digital_set_top_box2
Aniruddh Tyagi
Discrete cosine transform
Discrete cosine transform
Aniruddh Tyagi
DCT
DCT
Aniruddh Tyagi
EBU_DVB_S2 READY TO LIFT OFF
EBU_DVB_S2 READY TO LIFT OFF
Aniruddh Tyagi
ADVANCED DVB-C,DVB-S STB DEMOD
ADVANCED DVB-C,DVB-S STB DEMOD
Aniruddh Tyagi
DVB_Arch
DVB_Arch
Aniruddh Tyagi
haffman coding DCT transform
haffman coding DCT transform
Aniruddh Tyagi
tyagi 's doc
tyagi 's doc
Aniruddh Tyagi
quantization_PCM
quantization_PCM
Aniruddh Tyagi
ECMG & EMMG protocol
ECMG & EMMG protocol
Aniruddh Tyagi
7015567A
7015567A
Aniruddh Tyagi
Basic of BISS
Basic of BISS
Aniruddh Tyagi
euler theorm
euler theorm
Aniruddh Tyagi
fundamentals_satellite_communication_part_1
fundamentals_satellite_communication_part_1
Aniruddh Tyagi
art_sklar7_reed-solomon
art_sklar7_reed-solomon
Aniruddh Tyagi
DVBSimulcrypt2
DVBSimulcrypt2
Aniruddh Tyagi
en_302769v010101v
en_302769v010101v
Aniruddh Tyagi
Mais de Aniruddh Tyagi
(20)
security vulnerabilities of dvb chipsets
security vulnerabilities of dvb chipsets
whitepaper_mpeg-if_understanding_mpeg4
whitepaper_mpeg-if_understanding_mpeg4
BUC BLOCK UP CONVERTER
BUC BLOCK UP CONVERTER
digital_set_top_box2
digital_set_top_box2
Discrete cosine transform
Discrete cosine transform
DCT
DCT
EBU_DVB_S2 READY TO LIFT OFF
EBU_DVB_S2 READY TO LIFT OFF
ADVANCED DVB-C,DVB-S STB DEMOD
ADVANCED DVB-C,DVB-S STB DEMOD
DVB_Arch
DVB_Arch
haffman coding DCT transform
haffman coding DCT transform
tyagi 's doc
tyagi 's doc
quantization_PCM
quantization_PCM
ECMG & EMMG protocol
ECMG & EMMG protocol
7015567A
7015567A
Basic of BISS
Basic of BISS
euler theorm
euler theorm
fundamentals_satellite_communication_part_1
fundamentals_satellite_communication_part_1
art_sklar7_reed-solomon
art_sklar7_reed-solomon
DVBSimulcrypt2
DVBSimulcrypt2
en_302769v010101v
en_302769v010101v
Último
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Alan Dix
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
naman860154
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
Radu Cotescu
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
BookNet Canada
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path Mount
Puma Security, LLC
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
Michael W. Hawkins
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
Delhi Call girls
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
Maria Levchenko
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024
Scott Keck-Warren
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
Rafal Los
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
soniya singh
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
hans926745
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 Presentation
Ridwan Fadjar
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
HampshireHUG
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j
Slack Application Development 101 Slides
Slack Application Development 101 Slides
praypatel2
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
2toLead Limited
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
ThousandEyes
Último
(20)
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food Manufacturing
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path Mount
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping Elbows
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 Presentation
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Slack Application Development 101 Slides
Slack Application Development 101 Slides
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
quantization
1.
Quantization
Yao Wang Polytechnic University, Brooklyn, NY11201 http://eeweb.poly.edu/~yao
2.
Outline • Review the
three process of A to D conversion • Quantization – Uniform – Non-uniform • Mu-law – Demo on quantization of audio signals – Sample Matlab codes • Binary encoding – Bit rate of digital signals • Advantage of digital representation ©Yao Wang, 2006 EE3414:Quantization 2
3.
Three Processes in
A/D Conversion Sampling Quanti- Binary zation Encoding xc(t) x[n] = xc(nT) x[n] c[n] Sampling Quantization Binary Period Interval codebook T Q • Sampling: take samples at time nT – T: sampling period; – fs = 1/T: sampling frequency • Quantization: map amplitude values into a set of discrete values kQ – Q: quantization interval or stepsize • Binary Encoding – Convert each quantized value into a binary codeword ©Yao Wang, 2006 EE3414:Quantization 3
4.
Analog to Digital
Conversion 1 T=0.1 Q=0.25 0.5 0 -0.5 A2D_plot.m -1 0 0.2 0.4 0.6 0.8 1 ©Yao Wang, 2006 EE3414:Quantization 4
5.
How to determine
T and Q? • T (or fs) depends on the signal frequency range – A fast varying signal should be sampled more frequently! – Theoretically governed by the Nyquist sampling theorem • fs > 2 fm (fm is the maximum signal frequency) • For speech: fs >= 8 KHz; For music: fs >= 44 KHz; • Q depends on the dynamic range of the signal amplitude and perceptual sensitivity – Q and the signal range D determine bits/sample R • 2R=D/Q • For speech: R = 8 bits; For music: R =16 bits; • One can trade off T (or fs) and Q (or R) – lower R -> higher fs; higher R -> lower fs • We considered sampling in last lecture, we discuss quantization in this lecture ©Yao Wang, 2006 EE3414:Quantization 5
6.
Uniform Quantization • Applicable
when the signal is in a finite range (fmin, fmax) • The entire data range is divided into L equal intervals of length Q (known as quantization interval or quantization step-size) Q=(fmax-fmin)/L • Interval i is mapped to the middle value of this interval • We store/send only the index of f − f min quantized value Index of quantized value = Qi ( f ) = Q Quantized value = Q ( f ) = Qi ( f )Q + Q / 2 + f min ©Yao Wang, 2006 EE3414:Quantization 6
7.
Special Case I:
Signal range is symmetric • (a) L=even, mid-rise Q(f)=floor(f/q)*q+q/2 L = even, Mid - Riser L = odd, Mid - Tread f Q f Qi ( f ) = floor ( ), Q ( f ) = Qi ( f ) * Q + Qi ( f ) = round ( ), Q( f ) = Qi ( f ) * Q Q 2 Q ©Yao Wang, 2006 EE3414:Quantization 7
8.
Special Case II:
Signal range starts at 0 f min = 0, B = f max , q = f max / L f Qi ( f ) = floor ( ) Q Q Q( f ) = Qi ( f ) * Q + 2 ©Yao Wang, 2006 EE3414:Quantization 8
9.
Example •
For the following sequence {1.2,-0.2,-0.5,0.4,0.89,1.3…}, Quantize it using a uniform quantizer in the range of (-1.5,1.5) with 4 levels, and write the quantized sequence. • Solution: Q=3/4=0.75. Quantizer is illustrated below. -1.125 -0.375 0.375 1.125 -1.5 -0.75 0 0.75 1.5 Yellow dots indicate the partition levels (boundaries between separate quantization intervals) Red dots indicate the reconstruction levels (middle of each interval) 1.2 fall between 0.75 and 1.5, and hence is quantized to 1.125 • Quantized sequence: {1.125,-0.375,-0.375,0.375,1.125,1.125} ©Yao Wang, 2006 EE3414:Quantization 9
10.
Effect of Quantization
Stepsize Q=0.25 Q=0.5 1.5 1.5 1 1 0.5 0.5 0 0 -0.5 -0.5 -1 -1 -1.5 -1.5 0 0.2 0.4 0 0.2 0.4 demo_sampling_quant.m ©Yao Wang, 2006 EE3414:Quantization 10
11.
Demo: Audio Quantization
0.02 original at 16 bit Original 0.015 quantized at 4 bit Mozart.wav 0.01 0.005 Quantized Mozart_q16.wav 0 -0.005 -0.01 2 2.002 2.004 2.006 2.008 2.01 2.012 2.014 2.016 2.018 2.02 4 x 10 ©Yao Wang, 2006 EE3414:Quantization 11
12.
Demo: Audio Quantization
(II) 0.02 0.02 original at 16 bit quantized at 6 bit original at 16 bit 0.015 quantized at 4 bit 0.015 0.01 0.01 0.005 0.005 0 0 -0.005 -0.005 -0.01 -0.01 2 2.002 2.004 2.006 2.008 2.01 2.012 2.014 2.016 2.018 2.02 2 2.002 2.004 2.006 2.008 2.01 2.012 2.014 2.016 2.018 2.02 4 4 x 10 x 10 Quantized Quantized Mozart_q16.wav Mozart_q64.wav ©Yao Wang, 2006 EE3414:Quantization 12
13.
Non-Uniform Quantization • Problems
with uniform quantization – Only optimal for uniformly distributed signal – Real audio signals (speech and music) are more concentrated near zeros – Human ear is more sensitive to quantization errors at small values • Solution – Using non-uniform quantization • quantization interval is smaller near zero ©Yao Wang, 2006 EE3414:Quantization 13
14.
Quantization: General Description
Quantization levels : L Partition values : bl Partition regions : Bl = [bl −1 , bl ) Reconstruction values : g l Quantized Index : Qi ( f ) = l , if f ∈ Bl Quantizer value : Q( f ) = g l , if f ∈ Bl ©Yao Wang, 2006 EE3414:Quantization 14
15.
Function Representation
Q ( f ) = gl , if f ∈ Bl ©Yao Wang, 2006 EE3414:Quantization 15
16.
Design of Non-Uniform
Quantizer • Directly design the partition and reconstruction levels • Non-linear mapping+uniform quantization µ-law quantization ©Yao Wang, 2006 EE3414:Quantization 16
17.
µ-Law Quantization
y =F [ x ] |x| log 1+µ X max .sign[ x] = X max log[1 + µ ] ©Yao Wang, 2006 EE3414:Quantization 17
18.
Implementation of µ-Law
Quantization (Direct Method) – Transform the signal using µ-law: x->y y =F [ x ] |x| log 1+µ X max = X max .sign[ x] log[1 + µ ] – Quantize the transformed value using a uniform quantizer: y->y^ – Transform the quantized value back using inverse µ-law: y^->x^ x =F −1[ y ] X log(1+ µ ) y = max 10 X max − 1 sign(y) µ ©Yao Wang, 2006 EE3414:Quantization 18
19.
Implementation of µ-Law
Quantization (Indirect Method) • Indirect Method: – Instead of applying the above computation to each sample, one can pre-design a quantization table (storing the partition and reconstruction levels) using the above procedure. The actual quantization process can then be done by a simple table look-up. – Applicable both for uniform and non-uniform quantizers – How to find the partition and reconstruction levels for mu-law quantizer • Apply inverse mu-law mapping to the partition and reconstruction levels of the uniform quantizer for y. • Note that the mu-law formula is designed so that if x ranges from (-x_max, x_max), then y also has the same range. ©Yao Wang, 2006 EE3414:Quantization 19
20.
Example •
For the following sequence {1.2,-0.2,-0.5,0.4,0.89,1.3…}, Quantize it using a mu-law quantizer in the range of (-1.5,1.5) with 4 levels, and write the quantized sequence. • Solution (indirect method): – apply the inverse formula to the partition and reconstruction levels found for the previous uniform quantizer example. Because the mu-law mapping is symmetric, we only need to find the inverse values for y=0.375,0.75,1.125 µ=9, x_max=1.5, 0.375->0.1297, 0.75->0.3604, 1.125->0.7706 – Then quantize each sample using the above partition and reconstruction levels. ©Yao Wang, 2006 EE3414:Quantization 20
21.
Example (cntd)
-1.125 -0.375 0.375 1.125 -1.5 -0.75 0 0.75 1.5 x =F −1[ y ] Inverse µ-law X log(1+ µ ) y = max 10 X max − 1 sign(y) µ -0.77 -0.13 0.13 0.77 -1.5 -0.36 0 0.36 1.5 • Original sequence: {1.2,-0.2,-0.5,0.4,0.89,1.3…} • Quantized sequence – {0.77,-0.13,-0.77,0.77,0.77,0.77} ©Yao Wang, 2006 EE3414:Quantization 21
22.
Uniform vs. µ-Law
Quantization Uniform: Q=0.5 µ-law: Q=0.5,µ=16 1.5 1.5 1 1 0.5 0.5 0 0 -0.5 -0.5 -1 -1 -1.5 -1.5 0 0.2 0.4 0 0.2 0.4 With µ-law, small values are represented more accurately, but large values are represented more coarsely. ©Yao Wang, 2006 EE3414:Quantization 22
23.
Uniform vs. µ-Law
for Audio 0.02 0.02 q32 q64 0.01 0.01 0 0 Mozart_q32.wav Mozart_q64.wav -0.01 -0.01 2 2.01 2.02 2 2.01 2.02 4 4 x 10 x 10 0.02 0.02 q32µ4 q32µ16 0.01 0.01 0 Mozart_q32_m4.wav 0 Mozart_q32_m16.wav -0.01 -0.01 2 2.01 2.02 2 2.01 2.02 4 4 x 10 x 10
24.
Evaluation of Quantizer
Performance • Ideally we want to measure the performance by how close is the quantized sound to the original sound to our ears -- Perceptual Quality • But it is very hard to come up with a objective measure that correlates very well with the perceptual quality • Frequently used objective measure – mean square error (MSE) between original and quantized samples or signal to noise ratio (SNR) 1 MSE : σ q = 2 N ∑ ( x(n) − x(n)) n ˆ 2 ( SNR(dB) : SNR = 10 log10 σ x / σ q 2 2 ) where N is the number of samples in the sequence. 1 σ x is the variance of the original signal, σ z2 = ∑ ( x(n)) 2 2 N n ©Yao Wang, 2006 EE3414:Quantization 24
25.
Sample Matlab Code Go
through “quant_uniform.m”, “quant_mulaw.m” ©Yao Wang, 2006 EE3414:Quantization 25
26.
Binary Encoding • Convert
each quantized level index into a codeword consisting of binary bits • Ex: natural binary encoding for 8 levels: – 000,001,010,011,100,101,110,111 • More sophisticated encoding (variable length coding) – Assign a short codeword to a more frequent symbol to reduce average bit rate – To be covered later ©Yao Wang, 2006 EE3414:Quantization 26
27.
Example 1: uniform
quantizer • For the following sequence {1.2,-0.2,-0.5,0.4,0.89,1.3…}, Quantize it using a uniform quantizer in the range of (-1.5,1.5) with 4 levels, and write the quantized sequence and the corresponding binary bitstream. • Solution: Q=3/4=0.75. Quantizer is illustrated below. • Codewords: 4 levels can be represented by 2 bits, 00, 01, 10, 11 codewords 00 01 10 11 Quantized values -1.125 -0.375 0.375 1.125 -1.5 -0.75 0 0.75 1.5 • Quantized value sequence: {1.125,-0.375,-0.375,0.375,1.125,1.125} • Bitstream representing quantized sequence: {11, 01, 01, 10, 11, 11} ©Yao Wang, 2006 EE3414:Quantization 27
28.
Example 2: mu-law
quantizer codewords 00 01 10 11 -1.125 -0.375 0.375 1.125 -1.5 -0.75 0 0.75 1.5 x =F −1[ y ] Inverse µ-law X log(1+ µ ) y = max 10 X max − 1 sign(y) codewords 00 01 10 11 µ -0.77 -0.13 0.13 0.77 -1.5 -0.36 0 0.36 1.5 • Original sequence: {1.2,-0.2,-0.5,0.4,0.89,1.3…} • Quantized sequence: {0.77,-0.13,-0.77,0.77,0.77,0.77} • Bitstream: {11,01,00,11,11,11} ©Yao Wang, 2006 EE3414:Quantization 28
29.
Bit Rate of
a Digital Sequence • Sampling rate: f_s sample/sec • Quantization resolution: B bit/sample, B=[log2(L)] • Bit rate: R=f_s B bit/sec • Ex: speech signal sampled at 8 KHz, quantized to 8 bit/sample, R=8*8 = 64 Kbps • Ex: music signal sampled at 44 KHz, quantized to 16 bit/sample, R=44*16=704 Kbps • Ex: stereo music with each channel at 704 Kbps: R=2*704=1.4 Mbps • Required bandwidth for transmitting a digital signal depends on the modulation technique. – To be covered later. • Data rate of a multimedia signal can be reduced significantly through lossy compression w/o affecting the perceptual quality. – To be covered later. ©Yao Wang, 2006 EE3414:Quantization 29
30.
Advantages of Digital
Representation (I) 1.5 More immune to noise added in original signal channel and/or received signal storage 1 The receiver applies a threshold to the received signal: 0.5 0 if x < 0.5 x= ˆ 1 if x ≥ 0.5 0 -0.5 0 10 20 30 40 50 60 ©Yao Wang, 2006 EE3414:Quantization 30
31.
Advantages of Digital
Representation (II) • Can correct erroneous bits and/or recover missing bits using “forward error correction” (FEC) technique – By adding “parity bits” after information bits, corrupted bits can be detected and corrected – Ex: adding a “check-sum” to the end of a digital sequence (“0” if sum=even, “1” if sum=odd). By computing check-sum after receiving the signal, one can detect single errors (in fact, any odd number of bit errors). – Used in CDs, DVDs, Internet, wireless phones, etc. ©Yao Wang, 2006 EE3414:Quantization 31
32.
What Should You
Know • Understand the general concept of quantization • Can perform uniform quantization on a given signal • Understand the principle of non-uniform quantization, and can perform mu-law quantization • Can perform uniform and mu-law quantization on a given sequence, generate the resulting quantized sequence and its binary representation • Can calculate bit rate given sampling rate and quantization levels • Know advantages of digital representation • Understand sample matlab codes for performing quantization (uniform and mu-law) ©Yao Wang, 2006 EE3414:Quantization 32
33.
References • Y. Wang,
Lab Manual for Multimedia Lab, Experiment on Speech and Audio Compression. Sec. 1-2.1. (copies provided). ©Yao Wang, 2006 EE3414:Quantization 33
Baixar agora