SlideShare a Scribd company logo
1 of 17
Developement and
Implementation of an MPEG1
Layer III Decoder on x86 and
TMS320C6711 platforms
Farina Simone
(Braidotti Enrico)
DECODING PROCESS
ALIAS RECONSTRUCTION
It is performed only when using long blocks: this means
only when using pure long blocks or mixed blocks.
Lets see what long/short blocks are
Not encoded signal
Same signal encoded
using long blocks
Same signal encoded
using short blocks
HYBRID SYNTHESIS
• IMDCT (Inverse Modified Discrete Cosine Transform)
Subbands are backward transformed separately depending on
block length.
• 6-point IMDCT
When short blocks are used (pre-echoes masking)
• 18-point IMDCT
When long blocks are used
HYBRID SYNTHESIS
• Fast IMDCT algorithm (Szu-Wei Lee )
Based on simmetric properties of cosine function
It needs a rearranging stage to restore values to their original
positions
Drastically reduces number of operations if compared to direct
implementation
× (short/long) + (short/long)
Direct Implementation 216 / 648 180 / 612
Fast IMDCT (Szu-Wei Lee) 33 / 43 69 / 115
Improvement 84.7 % / 93.3 % 61.7 % / 81.2 %
HYBRID SYNTHESIS
• Windowing
Once transformed, subbands are windowed according to value of
block_type (subbands with short blocks are separately
transformed for each window and then overlapped)
• Overlap-adding
First half of transformed blocks is overlapped with second half of
the corresponding blocks in the previous granule
FREQUENCY INVERSION
Every second sample in every second subband has to be
multiplied by -1.
SYNTHESIS POLYPHASE FILTERBANK
This process
produces 32
PCM audio
samples.
• 576 / granule
• 1152 / frame
(equal to 26 ms
of audio @ 44,1
kHz)
Composed of
several steps, it
turns out to be
the most
time-consuming
stage of the
overall decoding
process
SYNTHESIS POLYPHASE FILTERBANK
• Polyphase Matrixing
It is a cosine-like transform (non standard )
The direct computation involves a 64×32 matrix and requires
almost ¼ of decoding time
Needs optimization to perform real-time decoding
• K. Konstantinides’ algorithm
• 32-point Fast DCT (B.G.Lee)
SYNTHESIS POLYPHASE FILTERBANK
• Konstantinides’ Algorithm
SYNTHESIS POLYPHASE FILTERBANK
• FCT Algorithm (Byeong-Gi Lee )
Using trigonometrical properties a 2
M
DCT can be performed by
2
M-1
2-point DCTs
Direct computation
× = N ²
+ = N· ( N-1 )
FCT
× = N/2 · log2
( N )
+ < 3· N/2 · log2
( N )
WAVE STANDARD
• Individuated by a 44-byte header, holds information about:
• sampling frequency
• number of channels
• . . .
• Uncompressed PCM
audio samples
(normally with 16
bits/sample resolution)
stored in following way:
Istante di
campionamento
Canale
0 1 (Left)
2 (Right)
1 1 (Left)
2 (Right)
2 1 (Left)
2 (Right)
PERFORMANCE ANALYSIS
• PC Performances
The decoder, without optimization, works in real time on the following CPUs:
The decoder, with optimization , reaches 17,5× on Pentium IV CPU
PERFORMANCE ANALYSIS
• C6711 DSK Performances
The decoder, without optimization, doesn’t work in real time on the board
• Parallel port is used for data transfer and it’s very very slow
• Most algorithms need optimization (only Huffman Decoding is optimized)
• Code needs some ASM optimization to use the full-potential of the board
architecture
Whole decoding process (except data transfer TO external hard disk) takes
about 10 times more than needed to work in real time. With optimization it is an
easy goal to reach.
PERFORMANCE ANALYSIS
• Time-occupation of optimized processes on C6711 DSK:
PERFORMANCE ANALYSIS
• Time-occupation of optimized processes on C6711 DSK:
PERFORMANCE ANALYSIS
• Time-occupation of other processes on C6711 DSK:

More Related Content

What's hot

On chip crosstalk_avoidance_codec_design_using_fibonacci
On chip crosstalk_avoidance_codec_design_using_fibonacciOn chip crosstalk_avoidance_codec_design_using_fibonacci
On chip crosstalk_avoidance_codec_design_using_fibonaccibharath naidu
 
FPS 레벨 디자인_포트폴리오_폭파미션 Nuclear Power Plant
FPS 레벨 디자인_포트폴리오_폭파미션 Nuclear Power PlantFPS 레벨 디자인_포트폴리오_폭파미션 Nuclear Power Plant
FPS 레벨 디자인_포트폴리오_폭파미션 Nuclear Power Plantdongjidong
 
Cycle’s topological optimizations and the iterative decoding problem on gener...
Cycle’s topological optimizations and the iterative decoding problem on gener...Cycle’s topological optimizations and the iterative decoding problem on gener...
Cycle’s topological optimizations and the iterative decoding problem on gener...Usatyuk Vasiliy
 
Matrix transposition
Matrix transpositionMatrix transposition
Matrix transposition동호 이
 
FPGA Implementation of FIR Filter using Various Algorithms: A Retrospective
FPGA Implementation of FIR Filter using Various Algorithms: A RetrospectiveFPGA Implementation of FIR Filter using Various Algorithms: A Retrospective
FPGA Implementation of FIR Filter using Various Algorithms: A RetrospectiveIJORCS
 
Design and Implementation of Area Optimized, Low Complexity CMOS 32nm Technol...
Design and Implementation of Area Optimized, Low Complexity CMOS 32nm Technol...Design and Implementation of Area Optimized, Low Complexity CMOS 32nm Technol...
Design and Implementation of Area Optimized, Low Complexity CMOS 32nm Technol...IJERA Editor
 
High Performance Pedestrian Detection On TEGRA X1
High Performance Pedestrian Detection On TEGRA X1High Performance Pedestrian Detection On TEGRA X1
High Performance Pedestrian Detection On TEGRA X1NVIDIA
 
Temporal_video_noise_reduction
Temporal_video_noise_reductionTemporal_video_noise_reduction
Temporal_video_noise_reductionShereef Shehata
 
vedic mathematics based MAC unit
vedic mathematics based MAC unitvedic mathematics based MAC unit
vedic mathematics based MAC unitNavya Shree
 

What's hot (13)

On chip crosstalk_avoidance_codec_design_using_fibonacci
On chip crosstalk_avoidance_codec_design_using_fibonacciOn chip crosstalk_avoidance_codec_design_using_fibonacci
On chip crosstalk_avoidance_codec_design_using_fibonacci
 
FPS 레벨 디자인_포트폴리오_폭파미션 Nuclear Power Plant
FPS 레벨 디자인_포트폴리오_폭파미션 Nuclear Power PlantFPS 레벨 디자인_포트폴리오_폭파미션 Nuclear Power Plant
FPS 레벨 디자인_포트폴리오_폭파미션 Nuclear Power Plant
 
Cycle’s topological optimizations and the iterative decoding problem on gener...
Cycle’s topological optimizations and the iterative decoding problem on gener...Cycle’s topological optimizations and the iterative decoding problem on gener...
Cycle’s topological optimizations and the iterative decoding problem on gener...
 
Matrix transposition
Matrix transpositionMatrix transposition
Matrix transposition
 
What is I/Q phase
What is I/Q phaseWhat is I/Q phase
What is I/Q phase
 
FPGA Implementation of FIR Filter using Various Algorithms: A Retrospective
FPGA Implementation of FIR Filter using Various Algorithms: A RetrospectiveFPGA Implementation of FIR Filter using Various Algorithms: A Retrospective
FPGA Implementation of FIR Filter using Various Algorithms: A Retrospective
 
Design and Implementation of Area Optimized, Low Complexity CMOS 32nm Technol...
Design and Implementation of Area Optimized, Low Complexity CMOS 32nm Technol...Design and Implementation of Area Optimized, Low Complexity CMOS 32nm Technol...
Design and Implementation of Area Optimized, Low Complexity CMOS 32nm Technol...
 
High Performance Pedestrian Detection On TEGRA X1
High Performance Pedestrian Detection On TEGRA X1High Performance Pedestrian Detection On TEGRA X1
High Performance Pedestrian Detection On TEGRA X1
 
Temporal_video_noise_reduction
Temporal_video_noise_reductionTemporal_video_noise_reduction
Temporal_video_noise_reduction
 
Chomsky Normal Form
Chomsky Normal FormChomsky Normal Form
Chomsky Normal Form
 
vedic mathematics based MAC unit
vedic mathematics based MAC unitvedic mathematics based MAC unit
vedic mathematics based MAC unit
 
Uic Montone Thesis
Uic Montone ThesisUic Montone Thesis
Uic Montone Thesis
 
UIC Thesis Montone
UIC Thesis MontoneUIC Thesis Montone
UIC Thesis Montone
 

Viewers also liked (12)

Psicologia general
Psicologia generalPsicologia general
Psicologia general
 
C_Thompson Script Scholarship Essay
C_Thompson Script Scholarship EssayC_Thompson Script Scholarship Essay
C_Thompson Script Scholarship Essay
 
Digital storytelling
Digital storytellingDigital storytelling
Digital storytelling
 
791
791791
791
 
Wallapod
WallapodWallapod
Wallapod
 
Kegiatan
KegiatanKegiatan
Kegiatan
 
Preesentacion del portafolio
Preesentacion del portafolioPreesentacion del portafolio
Preesentacion del portafolio
 
Summary p4
Summary p4Summary p4
Summary p4
 
Design Secrets of Slide Plates
Design Secrets of Slide PlatesDesign Secrets of Slide Plates
Design Secrets of Slide Plates
 
Going Live with Ustream
Going Live with UstreamGoing Live with Ustream
Going Live with Ustream
 
Design, Application & Installation of Pipe Clamps Webinar
Design, Application & Installation of Pipe Clamps WebinarDesign, Application & Installation of Pipe Clamps Webinar
Design, Application & Installation of Pipe Clamps Webinar
 
Guía de miscelánea de ciencias i
Guía de miscelánea de ciencias   iGuía de miscelánea de ciencias   i
Guía de miscelánea de ciencias i
 

Similar to Presentation_1367231547532

Efficient Implementation of Low Power 2-D DCT Architecture
Efficient Implementation of Low Power 2-D DCT ArchitectureEfficient Implementation of Low Power 2-D DCT Architecture
Efficient Implementation of Low Power 2-D DCT ArchitectureIJMER
 
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...ijceronline
 
HC05.5.3-Bolton-SGS-Thomson-MPEG
HC05.5.3-Bolton-SGS-Thomson-MPEGHC05.5.3-Bolton-SGS-Thomson-MPEG
HC05.5.3-Bolton-SGS-Thomson-MPEGDavid Parre
 
El text.tokuron a(2019).jung190711
El text.tokuron a(2019).jung190711El text.tokuron a(2019).jung190711
El text.tokuron a(2019).jung190711RCCSRENKEI
 
Fingerprint EVM
Fingerprint EVMFingerprint EVM
Fingerprint EVMPranay Raj
 
디지털통신 8
디지털통신 8디지털통신 8
디지털통신 8KengTe Liao
 
Course-Notes__Advanced-DSP.pdf
Course-Notes__Advanced-DSP.pdfCourse-Notes__Advanced-DSP.pdf
Course-Notes__Advanced-DSP.pdfShreeDevi42
 
Advanced_DSP_J_G_Proakis.pdf
Advanced_DSP_J_G_Proakis.pdfAdvanced_DSP_J_G_Proakis.pdf
Advanced_DSP_J_G_Proakis.pdfHariPrasad314745
 
Design of High Performance 8,16,32-bit Vedic Multipliers using SCL PDK 180nm ...
Design of High Performance 8,16,32-bit Vedic Multipliers using SCL PDK 180nm ...Design of High Performance 8,16,32-bit Vedic Multipliers using SCL PDK 180nm ...
Design of High Performance 8,16,32-bit Vedic Multipliers using SCL PDK 180nm ...Angel Yogi
 
TMPA910CRAXBG 32-Bit Display MCU
TMPA910CRAXBG 32-Bit Display MCUTMPA910CRAXBG 32-Bit Display MCU
TMPA910CRAXBG 32-Bit Display MCUPremier Farnell
 
Digital communication
Digital communicationDigital communication
Digital communicationmeashi
 
The evolution of TMS, family of DSP\'s
The evolution of TMS, family of DSP\'sThe evolution of TMS, family of DSP\'s
The evolution of TMS, family of DSP\'sRitul Sonania
 
DSP_2018_FOEHU - Lec 07 - IIR Filter Design
DSP_2018_FOEHU - Lec 07 - IIR Filter DesignDSP_2018_FOEHU - Lec 07 - IIR Filter Design
DSP_2018_FOEHU - Lec 07 - IIR Filter DesignAmr E. Mohamed
 
Iaetsd designing of cmos image sensor test-chip and its characterization
Iaetsd designing of cmos image sensor test-chip and its characterizationIaetsd designing of cmos image sensor test-chip and its characterization
Iaetsd designing of cmos image sensor test-chip and its characterizationIaetsd Iaetsd
 

Similar to Presentation_1367231547532 (20)

Efficient Implementation of Low Power 2-D DCT Architecture
Efficient Implementation of Low Power 2-D DCT ArchitectureEfficient Implementation of Low Power 2-D DCT Architecture
Efficient Implementation of Low Power 2-D DCT Architecture
 
Lecture 18 (5)
Lecture 18 (5)Lecture 18 (5)
Lecture 18 (5)
 
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
 
HC05.5.3-Bolton-SGS-Thomson-MPEG
HC05.5.3-Bolton-SGS-Thomson-MPEGHC05.5.3-Bolton-SGS-Thomson-MPEG
HC05.5.3-Bolton-SGS-Thomson-MPEG
 
Sampling.pptx
Sampling.pptxSampling.pptx
Sampling.pptx
 
Deblocking_Filter_v2
Deblocking_Filter_v2Deblocking_Filter_v2
Deblocking_Filter_v2
 
El text.tokuron a(2019).jung190711
El text.tokuron a(2019).jung190711El text.tokuron a(2019).jung190711
El text.tokuron a(2019).jung190711
 
26_Fan.pdf
26_Fan.pdf26_Fan.pdf
26_Fan.pdf
 
Fingerprint EVM
Fingerprint EVMFingerprint EVM
Fingerprint EVM
 
디지털통신 8
디지털통신 8디지털통신 8
디지털통신 8
 
Course-Notes__Advanced-DSP.pdf
Course-Notes__Advanced-DSP.pdfCourse-Notes__Advanced-DSP.pdf
Course-Notes__Advanced-DSP.pdf
 
Advanced_DSP_J_G_Proakis.pdf
Advanced_DSP_J_G_Proakis.pdfAdvanced_DSP_J_G_Proakis.pdf
Advanced_DSP_J_G_Proakis.pdf
 
Design of High Performance 8,16,32-bit Vedic Multipliers using SCL PDK 180nm ...
Design of High Performance 8,16,32-bit Vedic Multipliers using SCL PDK 180nm ...Design of High Performance 8,16,32-bit Vedic Multipliers using SCL PDK 180nm ...
Design of High Performance 8,16,32-bit Vedic Multipliers using SCL PDK 180nm ...
 
Data compression
Data compressionData compression
Data compression
 
TMPA910CRAXBG 32-Bit Display MCU
TMPA910CRAXBG 32-Bit Display MCUTMPA910CRAXBG 32-Bit Display MCU
TMPA910CRAXBG 32-Bit Display MCU
 
Digital communication
Digital communicationDigital communication
Digital communication
 
The evolution of TMS, family of DSP\'s
The evolution of TMS, family of DSP\'sThe evolution of TMS, family of DSP\'s
The evolution of TMS, family of DSP\'s
 
Tdm fdm
Tdm fdmTdm fdm
Tdm fdm
 
DSP_2018_FOEHU - Lec 07 - IIR Filter Design
DSP_2018_FOEHU - Lec 07 - IIR Filter DesignDSP_2018_FOEHU - Lec 07 - IIR Filter Design
DSP_2018_FOEHU - Lec 07 - IIR Filter Design
 
Iaetsd designing of cmos image sensor test-chip and its characterization
Iaetsd designing of cmos image sensor test-chip and its characterizationIaetsd designing of cmos image sensor test-chip and its characterization
Iaetsd designing of cmos image sensor test-chip and its characterization
 

More from Alexander Nevidimov

чек за платеж за телефон
чек за платеж за телефончек за платеж за телефон
чек за платеж за телефонAlexander Nevidimov
 
чек за платеж за телефон
чек за платеж за телефончек за платеж за телефон
чек за платеж за телефонAlexander Nevidimov
 

More from Alexander Nevidimov (20)

чек за платеж за телефон
чек за платеж за телефончек за платеж за телефон
чек за платеж за телефон
 
чек за платеж за телефон
чек за платеж за телефончек за платеж за телефон
чек за платеж за телефон
 
Presentation_1376917645876
Presentation_1376917645876Presentation_1376917645876
Presentation_1376917645876
 
Presentation_1376678601814
Presentation_1376678601814Presentation_1376678601814
Presentation_1376678601814
 
Presentation_1376311255728
Presentation_1376311255728Presentation_1376311255728
Presentation_1376311255728
 
Presentation_1376222064850
Presentation_1376222064850Presentation_1376222064850
Presentation_1376222064850
 
Presentation_1376220985856
Presentation_1376220985856Presentation_1376220985856
Presentation_1376220985856
 
Presentation_1376220236996
Presentation_1376220236996Presentation_1376220236996
Presentation_1376220236996
 
Presentation_1376218980392
Presentation_1376218980392Presentation_1376218980392
Presentation_1376218980392
 
Presentation_1376168115602
Presentation_1376168115602Presentation_1376168115602
Presentation_1376168115602
 
Presentation_1375882767439
Presentation_1375882767439Presentation_1375882767439
Presentation_1375882767439
 
Presentation_1375882705328
Presentation_1375882705328Presentation_1375882705328
Presentation_1375882705328
 
Presentation_1375280857464
Presentation_1375280857464Presentation_1375280857464
Presentation_1375280857464
 
Presentation_1375280653597
Presentation_1375280653597Presentation_1375280653597
Presentation_1375280653597
 
Presentation_1374052137363
Presentation_1374052137363Presentation_1374052137363
Presentation_1374052137363
 
Presentation_1373778041831
Presentation_1373778041831Presentation_1373778041831
Presentation_1373778041831
 
Presentation_1373190655210
Presentation_1373190655210Presentation_1373190655210
Presentation_1373190655210
 
Presentation_1372848115982
Presentation_1372848115982Presentation_1372848115982
Presentation_1372848115982
 
Presentation_1372103147097
Presentation_1372103147097Presentation_1372103147097
Presentation_1372103147097
 
Presentation_1371997361000
Presentation_1371997361000Presentation_1371997361000
Presentation_1371997361000
 

Presentation_1367231547532

  • 1. Developement and Implementation of an MPEG1 Layer III Decoder on x86 and TMS320C6711 platforms Farina Simone (Braidotti Enrico)
  • 3. ALIAS RECONSTRUCTION It is performed only when using long blocks: this means only when using pure long blocks or mixed blocks. Lets see what long/short blocks are Not encoded signal Same signal encoded using long blocks Same signal encoded using short blocks
  • 4. HYBRID SYNTHESIS • IMDCT (Inverse Modified Discrete Cosine Transform) Subbands are backward transformed separately depending on block length. • 6-point IMDCT When short blocks are used (pre-echoes masking) • 18-point IMDCT When long blocks are used
  • 5. HYBRID SYNTHESIS • Fast IMDCT algorithm (Szu-Wei Lee ) Based on simmetric properties of cosine function It needs a rearranging stage to restore values to their original positions Drastically reduces number of operations if compared to direct implementation × (short/long) + (short/long) Direct Implementation 216 / 648 180 / 612 Fast IMDCT (Szu-Wei Lee) 33 / 43 69 / 115 Improvement 84.7 % / 93.3 % 61.7 % / 81.2 %
  • 6. HYBRID SYNTHESIS • Windowing Once transformed, subbands are windowed according to value of block_type (subbands with short blocks are separately transformed for each window and then overlapped) • Overlap-adding First half of transformed blocks is overlapped with second half of the corresponding blocks in the previous granule
  • 7. FREQUENCY INVERSION Every second sample in every second subband has to be multiplied by -1.
  • 8. SYNTHESIS POLYPHASE FILTERBANK This process produces 32 PCM audio samples. • 576 / granule • 1152 / frame (equal to 26 ms of audio @ 44,1 kHz) Composed of several steps, it turns out to be the most time-consuming stage of the overall decoding process
  • 9. SYNTHESIS POLYPHASE FILTERBANK • Polyphase Matrixing It is a cosine-like transform (non standard ) The direct computation involves a 64×32 matrix and requires almost ¼ of decoding time Needs optimization to perform real-time decoding • K. Konstantinides’ algorithm • 32-point Fast DCT (B.G.Lee)
  • 10. SYNTHESIS POLYPHASE FILTERBANK • Konstantinides’ Algorithm
  • 11. SYNTHESIS POLYPHASE FILTERBANK • FCT Algorithm (Byeong-Gi Lee ) Using trigonometrical properties a 2 M DCT can be performed by 2 M-1 2-point DCTs Direct computation × = N ² + = N· ( N-1 ) FCT × = N/2 · log2 ( N ) + < 3· N/2 · log2 ( N )
  • 12. WAVE STANDARD • Individuated by a 44-byte header, holds information about: • sampling frequency • number of channels • . . . • Uncompressed PCM audio samples (normally with 16 bits/sample resolution) stored in following way: Istante di campionamento Canale 0 1 (Left) 2 (Right) 1 1 (Left) 2 (Right) 2 1 (Left) 2 (Right)
  • 13. PERFORMANCE ANALYSIS • PC Performances The decoder, without optimization, works in real time on the following CPUs: The decoder, with optimization , reaches 17,5× on Pentium IV CPU
  • 14. PERFORMANCE ANALYSIS • C6711 DSK Performances The decoder, without optimization, doesn’t work in real time on the board • Parallel port is used for data transfer and it’s very very slow • Most algorithms need optimization (only Huffman Decoding is optimized) • Code needs some ASM optimization to use the full-potential of the board architecture Whole decoding process (except data transfer TO external hard disk) takes about 10 times more than needed to work in real time. With optimization it is an easy goal to reach.
  • 15. PERFORMANCE ANALYSIS • Time-occupation of optimized processes on C6711 DSK:
  • 16. PERFORMANCE ANALYSIS • Time-occupation of optimized processes on C6711 DSK:
  • 17. PERFORMANCE ANALYSIS • Time-occupation of other processes on C6711 DSK: