| 结合信源和信道的多级矢量量化联合优化算法 |
| 徐敬德, 崔慧娟, 唐昆 |
| 清华大学 电子工程系, 清华大学信息科学与技术国家实验室, 北京 100084 |
| Joint optimization algorithm combining source and channel coding for multi-stage vector quantization |
| XU Jingde, CUI Huijuan, TANG Kun |
| National Laboratory for Information Science and Technology, Department of Electronic Engineer, Tsinghua University, Beijing 100084, China |
摘要:
| |||
| 摘要针对传统的信源优化多级矢量量化抗误码性能较差的问题, 提出了一种结合信源和信道的多级矢量量化码本联合优化算法。该算法将码本联合优化与非等重信道保护相结合, 充分利用多级矢量量化中各级码字之间的相互作用关系和非等重信道保护的特性, 对各级码字进行非等重误码率的迭代优化来降低整个系统失真。在低速率语音编码中线谱频率参数的仿真测试表明: 与信道优化的多级矢量量化独立码本和非等重信道保护相结合的方案相比, 在8%误码率信道下该算法线谱频率参数的平均谱失真降低了0.1 dB; 与等重误码率的码本联合优化方案相比, 在各种误码率信道下该算法线谱频率参数的平均谱失真都有明显降低。 | |||
| 关键词 ：线谱频率,多级矢量量化,联合优化 | |||
| Abstract：Source optimized multi-stage vector quantization can have poor quality due to channel errors. This paper presents a joint optimization algorithm combining source and channel coding for multi-stage vector quantization. The joint optimization of the codebook with unequal error protection optimizes each codeword in each iteration using unequal bit error rates to reduce the distortion in the whole system by taking advantage of the connections between each codeword in each stage and the characteristics of unequal error protection. Simulations using linear spectral pairs of low bit speech coding show that the average spectral distortion is reduced 0.1 dB for an 8% bit error rate compared to combined channel optimized multi-stage individual vector quantization and unequal error protection. The average spectral distortion is much less than with using multi-stage codebook joint optimization of equal bit error rates. | |||
| Key words：linear spectral pairmulti-stage vector quantizationjoint optimization | |||
| 收稿日期: 2014-07-07 出版日期: 2015-09-30 | |||
| |||
| 通讯作者:唐昆,教授,E-mail:tangkun@tsinghua.edu.cnE-mail: tangkun@tsinghua.edu.cn | |||
| 引用本文: |
| 徐敬德, 崔慧娟, 唐昆. 结合信源和信道的多级矢量量化联合优化算法[J]. 清华大学学报（自然科学版）, 2015, 55(8): 826-830. XU Jingde, CUI Huijuan, TANG Kun. Joint optimization algorithm combining source and channel coding for multi-stage vector quantization. Journal of Tsinghua University(Science and Technology), 2015, 55(8): 826-830. |
| 链接本文: |
| http://jst.tsinghuajournals.com/CN/或 http://jst.tsinghuajournals.com/CN/Y2015/V55/I8/826 |
图表:
 |
| 表2 3种方案的平均谱失真/dB |
 |
| 表1 4种方案的平均谱失真/dB |
参考文献:
| [1] 鲍长春. 低比特率数字语音编码基础 [M].北京: 北京工业大学出版社, 2001.BAO Changchun. Low Bit Digital Speech Coding [M]. Beijing: Beijing Industry University Press, 2001. (in Chinese) [2] Quweider M K, Salari E. Efficient classification and codebook design for CVQ [J]. IEEE Proceedings Vision, Image and Signal Processing, 1996, 143(6): 344-352. [3] Chatterjee S, Sreenivas T V. Optimum switched split vector quantization of LSF parameters [J]. Signal Processing, 2008, 88(6) : 1528 -1538. [4] Murakami T, Asai K, Yamazaki E. Vector quantizer of video signals [J]. Electronics Letters, 1982, 55(3): 1005-1006. [5] Chan W Y, Gersho A. Enhanced multistage vector quantization with gonstrained storage [C]// The 24th Asilomar Conference on Circuits, Systems and Computers. London, UK: IEEE, 1990: 256-259. [6] Pan J S, Chu S C. Non-redundant VQ channel coding using Tabu search strategy [J]. Electronics Letters, 1996, 32(17): 133-138. [7] Pan J S, Lu Z M, Chu S C, et al. Non-redundant VQ channel coding using modified Tabu search approach with simulated annealing [C]// Int Conf Knowledge-Based Intelligent Information Engineering Systems. Adelaide, Australia: IEEE, 1999: 242 - 245. [8] Nariman F. A study of vector quantization for noisy channels [J]. IEEE Transactions on Information Theory, 1990, 36(4): 1220-1230. [9] Nam P, Nariman F, Takehiro M. A unified approach to tree-structured and multistage vector quantization for noisy channels [J]. IEEE Transactions on Information Theory, 1993, 39(3): 102-114. [10] Venkatesh K, David V A, Kwan K T. Optimal multistage vector quantization of LPC parameters over noisy channels [J]. IEEE Transactions on Speech and Audio Processing, 2004: 12(1): 856-864. [11] Yirong S, Goldsmith A J, Effros M. Joint design of vector quantizers and RCPC channel codes for Rayleigh fading channels [C]// IEEE Global Telecommunications Conference. Beijing, China: IEEE, 2000: 1611-1615. [12] Kim S J, Oh Y H. Spit vector quantization of LSF parameters with minimum of dLSF constraint [J]. IEEE Signal Processing Letters, 1999, 6(9): 227-229. |
相关文章:
| ||
