分布式音频编码中虚拟信道模型研究
|
摘要
随着当今社会的发展，大量的无线音频出现并逐渐发展，分布式音频编码应运而生。在分布式音频编码系统中，解码端生成的边信息Y与编码端的音频源X构成了一个虚拟依赖信道。
由于时空域相关噪声的存在，分布式音频编码虚拟信道建模非常困难。于是针对如何降低时空域相关噪声这一问题，本文在已有分布式音频编码虚拟信道建模的基础上，提出一种基于小波变换分布式音频编码多径虚拟依赖信道模型的方案。所做的主要工作及取得的创新成果主要体现在以下几个方面：
首先，分析了小波变换分布式音频编码虚拟依赖信道模型，对小波变换的LL低频子带采用Wyner-Ziv，而对高频子带采用SPIHT方法编码。在该模型基础上提出了一种基于小波变换分布式音频编码多径虚拟依赖信道模型的方案，并进行了建模：音频源X经过小波变换后，分解为多个音频子源进行量化，而且每个音频子源对应一个虚拟信道，构成多径虚拟信道，其次，在这种基于小波变换分布式音频编码多径虚拟依赖信道模型中，结合小波变换和调制码率，运用数学公式计算此建模的噪声、编解码复杂度和延迟时间，得到的噪声、编解码复杂度和延迟时间都比已有的模型低。
最后，先对小波域多径虚拟依赖信道模型分布式音频编码系统性能进行仿真，再对小波域多径虚拟信道模型分布式音频编码系统与小波域虚拟信道模型分布式音频编码系统的仿真结果相比较,充分验证了小波域多径虚拟依赖信道模型分布式音频编码系统下编码简单，证明了该系统的有效性和可行性。
A lot of wireless audio appear and develop gradually with nowadays socialdevelopment, the distributed audio coding comes with the tide of fashion. TraditionalIn the distributed audio coding system,the side information Y born at sender andaudio source X at at receiver compose a virtual channel.
Because the existence of the area in time and space related noise, the distributed audio coding virtual channel model is very difficult.Hence,in order to solve the problem that how to low related noise of the time and space,this paper th-at hasing already the distributed audio coding virtual channel model brings up a kind scheme according to DWT distributed audio coding many virtual chann-els model. he main worrs and some innovative achievements reflect mainly in thefollowing asTpects:
Firstly,we analyze DWT distributed audio coding virtual channel model.Takethe adoption Wyner-Ziv to the LL low frequency sonbands of DWT, but tohight sonbands take to adopt SPIHT method encoding. In the model foundationup,we bring up a kind scheme according to DWT distributed audio codingmany virtual chann-els model:adopt to DWT, resolve to several audio sonsources,each audio audio sonsource to constitute many virtual channels in responseto a virtual channel.
Secondly, combine DWT and make a code rate, combine the noise, pla it offormula analysis this model decoding complications and delay time.
Finally,we emulate DWT distributed audio coding many virtual channels model performance.We emulate DWT distributed audio coding many virtual channelsmodel that DWT distributed audio coding virtual channel model as a result.It Well verifie s DWT distributed audio coding many virtual channels modeldepe-nd on a letter way the model distribute type audio frequency code systembottom code in brief.They prove the usefulness and possibility of the system.
由于时空域相关噪声的存在，分布式音频编码虚拟信道建模非常困难。于是针对如何降低时空域相关噪声这一问题，本文在已有分布式音频编码虚拟信道建模的基础上，提出一种基于小波变换分布式音频编码多径虚拟依赖信道模型的方案。所做的主要工作及取得的创新成果主要体现在以下几个方面：
首先，分析了小波变换分布式音频编码虚拟依赖信道模型，对小波变换的LL低频子带采用Wyner-Ziv，而对高频子带采用SPIHT方法编码。在该模型基础上提出了一种基于小波变换分布式音频编码多径虚拟依赖信道模型的方案，并进行了建模：音频源X经过小波变换后，分解为多个音频子源进行量化，而且每个音频子源对应一个虚拟信道，构成多径虚拟信道，其次，在这种基于小波变换分布式音频编码多径虚拟依赖信道模型中，结合小波变换和调制码率，运用数学公式计算此建模的噪声、编解码复杂度和延迟时间，得到的噪声、编解码复杂度和延迟时间都比已有的模型低。
最后，先对小波域多径虚拟依赖信道模型分布式音频编码系统性能进行仿真，再对小波域多径虚拟信道模型分布式音频编码系统与小波域虚拟信道模型分布式音频编码系统的仿真结果相比较,充分验证了小波域多径虚拟依赖信道模型分布式音频编码系统下编码简单，证明了该系统的有效性和可行性。
A lot of wireless audio appear and develop gradually with nowadays socialdevelopment, the distributed audio coding comes with the tide of fashion. TraditionalIn the distributed audio coding system,the side information Y born at sender andaudio source X at at receiver compose a virtual channel.
Because the existence of the area in time and space related noise, the distributed audio coding virtual channel model is very difficult.Hence,in order to solve the problem that how to low related noise of the time and space,this paper th-at hasing already the distributed audio coding virtual channel model brings up a kind scheme according to DWT distributed audio coding many virtual chann-els model. he main worrs and some innovative achievements reflect mainly in thefollowing asTpects:
Firstly,we analyze DWT distributed audio coding virtual channel model.Takethe adoption Wyner-Ziv to the LL low frequency sonbands of DWT, but tohight sonbands take to adopt SPIHT method encoding. In the model foundationup,we bring up a kind scheme according to DWT distributed audio codingmany virtual chann-els model:adopt to DWT, resolve to several audio sonsources,each audio audio sonsource to constitute many virtual channels in responseto a virtual channel.
Secondly, combine DWT and make a code rate, combine the noise, pla it offormula analysis this model decoding complications and delay time.
Finally,we emulate DWT distributed audio coding many virtual channels model performance.We emulate DWT distributed audio coding many virtual channelsmodel that DWT distributed audio coding virtual channel model as a result.It Well verifie s DWT distributed audio coding many virtual channels modeldepe-nd on a letter way the model distribute type audio frequency code systembottom code in brief.They prove the usefulness and possibility of the system.
引文
[1] Painter T-8paniasA,“Perce-ptual codingofdigital audio”,ProcIEEE,2000,88(4):45-1515.
[2] Ted Painter·Andfea8 Spanias,“A review ofalgorithms for pcrcoptual codingofcUgital audio signals”,Proceedings of International Confcrenco on DistalSiIal Processing(DSP),1997.179-205.
[3] Oohama Y.Gaussian multiterminal source coding.IEEE Trans.In-form.Theory,1997,43:1912-1923.
[4] Lam Y.H,Stewart R.W,“Perception-based residual analysis.Synthesis system”,Proc ICASSP 1999.USA:IEEE,PP.989-992,1999.
[5] Pumhagen H,“Advances in parametric audio coding”,1999 IEEE Workshopon Applications of Signal Processing to Audio and Acoustics,PP.31-34,Published:1999.
[6] Nomura T.1wadare,M.Serizawa.“A bitrate and bandwidth scalable CELP coder”,Proceedings of the 1998 IEEE International Conference on Acoustics,Speechand Signal Processing,1998,v01.1,PP.341-344,Published:1998.
[7] Pan D.“A tutorial on MPEG/audio compression”.IEEE Multimcdia,V01.22,pp.60-74,PuMished:Summer 1995.
[8] Lam Y．H，Stewart R．W．Perception-based residual analysis-synthesis system．Proc ICASSP 1999．USA：IEEE．1999．989-992．
[9] Pumhagan H.Advances in parametric audio coding.1999 mEE Workshop onApplications of Signal Processing to Audioand Acoustics,Put'lished:1999,Page(s):31-34.
[10] Nomura,T.1wadarc,M.Serizawa,M.A bitrate and band-width scalable CELPcoder.Proceedings ofthe 1998 IEEE International Conference 011 Acoustics,Speech and Signal Processing,1998,Published:1998,Volume:1,Page(s):341-344.
[11] ISO/IEC JTCl/SC29,wGll.Coding of moving piclolrcs and audio-MPEG-2Advanced Audio Coding AAC.ISO/IEC 13818-7 International Standard,1997.
[l2] Pan,D.A mtorial on MPEG/audio compression.IEEE Multimedia,Published:Summer 1995,Volume:22,Page(s):60-74.
[13] ISO/IEC International Standard 11172“Coding of Moving Pictures and Associated Audio for Digital Storage Media at up to about 1.5Mbit/s-Part 3:Audio”.
[14] ISO/IEC Intgwnational Standard 13818“generic Coding of Moving Pictureand Associated Audio Information·Part7:Advanced Audio Coding”.
[15] J D Slepian.J K Wolf.Noiseless coding of correlated information SOUrccS01.IEEE Transactions on Information Theory(S0018-944s),1973,rr-19:471-480.
[16] A D Wyner,J ziv.The rate·distortion function for sonrc￡coding with sideinformation at the decoder[J].IEEE Transactions on Information Theory(S0018-9448),1976,IT-22(1):1-10.
[17] B.Girod,A.Aaron,S.Rane et al，Distributed video coding,Proceedings of theIEEE,2005,93(1):71-83.
[18] S.S.Pradhan,K Ramchandran.Distributed Source Coding Using Syndromes(DISCUS):Design and Construction.Proceedings of the Data Compression Conference(DCC),1999,158-167.
[19] J.Bajesy,E Mitran.Coding for the Slepian-Wolf Problem with turbo codes.Proc.IEEE Global Communications Conference,2001,2(4):1400-1404.
[20] J.Garcia-Ffias.Compression of correlated binary sources using turbo codes.IEEE Communications Letters,2001,5(10):417-419.
[21] A.Aaron,B.Girod.Compression still side information using turbo c,odes.Proc.IEEE Data Compression Conference,2002,252-261.
[22] A.Liveris,Z.Xiong,C.Georghiades.Compression ofbinary source with side information at the decoder using LDPC codes.IEEE Communications Letters,2002,6(10):440-442.
[23] Y Yang,S.Cheng,Z.Xiong.Wyner-Ziv coding based OIl TCQ and LDPC codes.Proc.Asilomar Conference On signals ,Systems and Computers,2003,V01.1:825-829.
[24] Z.Xiong,A.Liveris,S.Cheng.Nested quantization and Slepian-wolf coding:AWyner-Ziv coding paradigm for i.i.d sources.Proc.IEEE Workshop 011 Statistical Signal Processing(SSP),2003,399-402.
[25] D.Rebollo-Moncdero,R Zhang,B.Giro&Design of optimal quantizers for distributed source coding.Proc.IEEE Data Compression Conference,2003,13-22.
[26] I Purl,k Ramchandran.PRISM:A 11eW robust video coding architecture based 011 distributed compression principles.Proc.Allerton Conference 011 Communication,Control and Computer,2002.
[27] IL Puri,k Ramehandran.PRISM:A"rcvcxl multimedia coding paradigm．Proc．IEEE International Confefence 01"1 Image Processing ,2003,617-620.
[28] A.Aaron,E.Seaon,B.Girod.Towards practical Wyner-Ziv coding of video.Proc.IEEE International Conference on Image Processing,2003,V01.2:869-872.
[29] D.Rebollo-Monedcro,A.Aaron,B.Girod.Transforms for high rate distributed sourc:e coding.Proc.Asilomar Conference on Signals,Systems and Computers ,2003,V01.1:850-854.
[30] A.Aaron,S.Rane,E.SeRon,B.Oirod.Transform-domain Wyner-Ziv codec for video.Pmc.SPIE Visual Communications and Image Processing,2004.
[31] A.Aaron,S.Rane,B.Girod.Wyner-Ziv video coding with hash based motion compensationat the receiver.Proc.IEEE International Conference on Image Processing,2004.
[32] S.S.Pradhan,K Ramchandran．Distributed Source Coding Using Syndromes(DISCUS)：Design and Construction.IEEE Transactions on Information Theory,2003,V01.49:626-643.
[33] John P.The design of nonuniform modulated filterbanks.IEEE Transactionson Signal Processing,1995,43(11):2550-2560.
[34] Gallager R.Low density parity cheek codes[J].IRE Transactions on Information Theory,1962,8:21-28.
[35] John P.The design of nonuniform modulated filterbanks.IEEE Transactionson Signal Processing, 1995, 43(11):2550-2560.
[36]陈熹.小波变换与图像压缩的探讨[J].教育探究.2006(4):17-20
[37] ISHWAR P,PRABHAKARAN V M, RAM CHANDAN K.Towards a theoy for video coding using distributed compression principles[C].InternationalConference on Image Processing,2003(2):687-690.
[38] WEERAKKODY W A R J, FERNANDO WAC,ADIKARI AB,et al..4-PSKTTCM for Wyner-Ziv frame coding in DVC[J].IEICE Electronics Express,2006,3(8):178-183.
[39] AARON A,ZHANG R,GIROD B.Wyner-Ziv coding of motion video[C].AsilomarConference on Signals,Systems and Computers,Pacific Grove,2002:35.
[40] CHIEN W J,KARAM L J.Distributed video coding with 3D recursive searchblock matching[C].IEEE International Symposium on Circuits and Systems,2006.
[41] SSCENSO J,BRITES C,PEREIRA F.Motion compensated refinement for lowcomplexity pixel based distributed video coding[J].IEEE International C-onference on Advanced Video and SignalBased Surveillance,2005:593-598.
[42] BRITES C,ASCENSO J,PEREIRA F.Improve transform domain Wyner-Zivvideo coding performance[C].IEEE International Conference on Acoustics,Speechand Signal Processing,2006:525-528.
[43]卿粼波，何小海，吕瑞等分布式视频编码中边信息的多策略优化[J].四川大学学报(工程科学版)，2008,40(1):138-143.
[44] Guo Xu.Video Coding Based on Set Partitioning in Hierarchical Tree[C]//Proc.of IEEE Int’l Conf.on Image Processing.[S.l.]:IEEE Press,2006.
[45] MAJUMDA A,BAMCHANDRAN K,KOZINTSEV I.Dis tributed coding forwireless audio sensors[C]//Proceedings of IEEE Workshop on Applicationsof Signal Processing to Audio and Acoustics:IEEE Press,2003:209-212.
[46] OOHAMA Y.Multiterminal source coding for correlated memoryless Gaussian sources with several sidetion for the quadratic Gaussian CEO problem[J].IEEE JSAC,2004,22:977-987.
[47] SARENSEN J A.High-fidelity multichannel audio coding[J].IEEE Signal Processing Magazine,2005,22(5):150-153.
[48] FALLER C.Parametric multichannel audio coding：synthesis of coherence cues[J].IEEE Trans.on Audio Speech Language Processing,2006,14(1):299-310.
[2] Ted Painter·Andfea8 Spanias,“A review ofalgorithms for pcrcoptual codingofcUgital audio signals”,Proceedings of International Confcrenco on DistalSiIal Processing(DSP),1997.179-205.
[3] Oohama Y.Gaussian multiterminal source coding.IEEE Trans.In-form.Theory,1997,43:1912-1923.
[4] Lam Y.H,Stewart R.W,“Perception-based residual analysis.Synthesis system”,Proc ICASSP 1999.USA:IEEE,PP.989-992,1999.
[5] Pumhagen H,“Advances in parametric audio coding”,1999 IEEE Workshopon Applications of Signal Processing to Audio and Acoustics,PP.31-34,Published:1999.
[6] Nomura T.1wadare,M.Serizawa.“A bitrate and bandwidth scalable CELP coder”,Proceedings of the 1998 IEEE International Conference on Acoustics,Speechand Signal Processing,1998,v01.1,PP.341-344,Published:1998.
[7] Pan D.“A tutorial on MPEG/audio compression”.IEEE Multimcdia,V01.22,pp.60-74,PuMished:Summer 1995.
[8] Lam Y．H，Stewart R．W．Perception-based residual analysis-synthesis system．Proc ICASSP 1999．USA：IEEE．1999．989-992．
[9] Pumhagan H.Advances in parametric audio coding.1999 mEE Workshop onApplications of Signal Processing to Audioand Acoustics,Put'lished:1999,Page(s):31-34.
[10] Nomura,T.1wadarc,M.Serizawa,M.A bitrate and band-width scalable CELPcoder.Proceedings ofthe 1998 IEEE International Conference 011 Acoustics,Speech and Signal Processing,1998,Published:1998,Volume:1,Page(s):341-344.
[11] ISO/IEC JTCl/SC29,wGll.Coding of moving piclolrcs and audio-MPEG-2Advanced Audio Coding AAC.ISO/IEC 13818-7 International Standard,1997.
[l2] Pan,D.A mtorial on MPEG/audio compression.IEEE Multimedia,Published:Summer 1995,Volume:22,Page(s):60-74.
[13] ISO/IEC International Standard 11172“Coding of Moving Pictures and Associated Audio for Digital Storage Media at up to about 1.5Mbit/s-Part 3:Audio”.
[14] ISO/IEC Intgwnational Standard 13818“generic Coding of Moving Pictureand Associated Audio Information·Part7:Advanced Audio Coding”.
[15] J D Slepian.J K Wolf.Noiseless coding of correlated information SOUrccS01.IEEE Transactions on Information Theory(S0018-944s),1973,rr-19:471-480.
[16] A D Wyner,J ziv.The rate·distortion function for sonrc￡coding with sideinformation at the decoder[J].IEEE Transactions on Information Theory(S0018-9448),1976,IT-22(1):1-10.
[17] B.Girod,A.Aaron,S.Rane et al，Distributed video coding,Proceedings of theIEEE,2005,93(1):71-83.
[18] S.S.Pradhan,K Ramchandran.Distributed Source Coding Using Syndromes(DISCUS):Design and Construction.Proceedings of the Data Compression Conference(DCC),1999,158-167.
[19] J.Bajesy,E Mitran.Coding for the Slepian-Wolf Problem with turbo codes.Proc.IEEE Global Communications Conference,2001,2(4):1400-1404.
[20] J.Garcia-Ffias.Compression of correlated binary sources using turbo codes.IEEE Communications Letters,2001,5(10):417-419.
[21] A.Aaron,B.Girod.Compression still side information using turbo c,odes.Proc.IEEE Data Compression Conference,2002,252-261.
[22] A.Liveris,Z.Xiong,C.Georghiades.Compression ofbinary source with side information at the decoder using LDPC codes.IEEE Communications Letters,2002,6(10):440-442.
[23] Y Yang,S.Cheng,Z.Xiong.Wyner-Ziv coding based OIl TCQ and LDPC codes.Proc.Asilomar Conference On signals ,Systems and Computers,2003,V01.1:825-829.
[24] Z.Xiong,A.Liveris,S.Cheng.Nested quantization and Slepian-wolf coding:AWyner-Ziv coding paradigm for i.i.d sources.Proc.IEEE Workshop 011 Statistical Signal Processing(SSP),2003,399-402.
[25] D.Rebollo-Moncdero,R Zhang,B.Giro&Design of optimal quantizers for distributed source coding.Proc.IEEE Data Compression Conference,2003,13-22.
[26] I Purl,k Ramchandran.PRISM:A 11eW robust video coding architecture based 011 distributed compression principles.Proc.Allerton Conference 011 Communication,Control and Computer,2002.
[27] IL Puri,k Ramehandran.PRISM:A"rcvcxl multimedia coding paradigm．Proc．IEEE International Confefence 01"1 Image Processing ,2003,617-620.
[28] A.Aaron,E.Seaon,B.Girod.Towards practical Wyner-Ziv coding of video.Proc.IEEE International Conference on Image Processing,2003,V01.2:869-872.
[29] D.Rebollo-Monedcro,A.Aaron,B.Girod.Transforms for high rate distributed sourc:e coding.Proc.Asilomar Conference on Signals,Systems and Computers ,2003,V01.1:850-854.
[30] A.Aaron,S.Rane,E.SeRon,B.Oirod.Transform-domain Wyner-Ziv codec for video.Pmc.SPIE Visual Communications and Image Processing,2004.
[31] A.Aaron,S.Rane,B.Girod.Wyner-Ziv video coding with hash based motion compensationat the receiver.Proc.IEEE International Conference on Image Processing,2004.
[32] S.S.Pradhan,K Ramchandran．Distributed Source Coding Using Syndromes(DISCUS)：Design and Construction.IEEE Transactions on Information Theory,2003,V01.49:626-643.
[33] John P.The design of nonuniform modulated filterbanks.IEEE Transactionson Signal Processing,1995,43(11):2550-2560.
[34] Gallager R.Low density parity cheek codes[J].IRE Transactions on Information Theory,1962,8:21-28.
[35] John P.The design of nonuniform modulated filterbanks.IEEE Transactionson Signal Processing, 1995, 43(11):2550-2560.
[36]陈熹.小波变换与图像压缩的探讨[J].教育探究.2006(4):17-20
[37] ISHWAR P,PRABHAKARAN V M, RAM CHANDAN K.Towards a theoy for video coding using distributed compression principles[C].InternationalConference on Image Processing,2003(2):687-690.
[38] WEERAKKODY W A R J, FERNANDO WAC,ADIKARI AB,et al..4-PSKTTCM for Wyner-Ziv frame coding in DVC[J].IEICE Electronics Express,2006,3(8):178-183.
[39] AARON A,ZHANG R,GIROD B.Wyner-Ziv coding of motion video[C].AsilomarConference on Signals,Systems and Computers,Pacific Grove,2002:35.
[40] CHIEN W J,KARAM L J.Distributed video coding with 3D recursive searchblock matching[C].IEEE International Symposium on Circuits and Systems,2006.
[41] SSCENSO J,BRITES C,PEREIRA F.Motion compensated refinement for lowcomplexity pixel based distributed video coding[J].IEEE International C-onference on Advanced Video and SignalBased Surveillance,2005:593-598.
[42] BRITES C,ASCENSO J,PEREIRA F.Improve transform domain Wyner-Zivvideo coding performance[C].IEEE International Conference on Acoustics,Speechand Signal Processing,2006:525-528.
[43]卿粼波，何小海，吕瑞等分布式视频编码中边信息的多策略优化[J].四川大学学报(工程科学版)，2008,40(1):138-143.
[44] Guo Xu.Video Coding Based on Set Partitioning in Hierarchical Tree[C]//Proc.of IEEE Int’l Conf.on Image Processing.[S.l.]:IEEE Press,2006.
[45] MAJUMDA A,BAMCHANDRAN K,KOZINTSEV I.Dis tributed coding forwireless audio sensors[C]//Proceedings of IEEE Workshop on Applicationsof Signal Processing to Audio and Acoustics:IEEE Press,2003:209-212.
[46] OOHAMA Y.Multiterminal source coding for correlated memoryless Gaussian sources with several sidetion for the quadratic Gaussian CEO problem[J].IEEE JSAC,2004,22:977-987.
[47] SARENSEN J A.High-fidelity multichannel audio coding[J].IEEE Signal Processing Magazine,2005,22(5):150-153.
[48] FALLER C.Parametric multichannel audio coding：synthesis of coherence cues[J].IEEE Trans.on Audio Speech Language Processing,2006,14(1):299-310.