基于压缩感知的多声道信号重构与改进  

Compressed Sensing Based Multi-channel Audio Signal Reconstruction and Improvement

作　　者：王映雪 刘昱 段继忠 翟丽 Wang Yingxue;Liu Yu;Duan Jizhong;Zhai Li(School of Microelectronics,Tianjin University,Tianjin 300072,China;Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,China)

机构地区：[1]天津大学微电子学院,天津300072 [2]昆明理工大学信息工程与自动化学院,云南昆明650500

出　　处：《南开大学学报（自然科学版）》2020年第4期41-46,共6页Acta Scientiarum Naturalium Universitatis Nankaiensis

基　　金：国家自然科学基金(61101226,61771338);The Scientific Research Foundation of KUST(KKSY201703017)。

摘　　要：多声道环绕声技术的发展和应用对音频信号的存储、传输和处理提出了新的要求.为了在显著降低音频数据量的同时保障多声道环绕声的质量,将多声道音频技术与压缩感知技术(CS)相结合,分别提出二维和一维两种CS重构方案来对多声道音频数据进行采样和重构.此外,针对低CS采样率的情况,提出了一种可以改进CS重构信号质量的新算法.实验结果表明,新算法不仅能显著提高音频信号的重构质量还能大大缩短重构时间,尤其在低采样率时提升显著.With the rapid development and application of surround sound systems,it becomes more and more difficult to store,transmit or process multi-channel audio signals.In order to guarantee the audio quality as well as to reduce audio sampling data,we applied compressed sensing(CS)to processing multi-channel surround sound.Two CS schemes including 1D and 2D were proposed for reconstructing multi-channel audio.In terms of low CS sampling rates,we presented a new reconstruction algorithm.The experimental results illustrate that the proposed algorithm not only improves the reconstruction quality but also reduces the reconstruction time,especially in conditions of low CS sampling rates.

关 键 词：音频处理 多声道环绕声 压缩感知 重构算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]