语音信号处理中鲁棒性压缩感知关键技术
发布时间：2018-11-21 13:45
【摘要】：压缩感知技术在许多领域都有广阔的应用前景。压缩感知关键技术主要包括稀疏矩阵的选取,观测矩阵的构造以及重构算法的设计。语音信号不同于一般信号,通常在观测矩阵和重构算法中有特殊的结构特征。在实际的应用中,噪声难以避免,而压缩感知系统中,重构系统是非线性的,且对噪声敏感,因而研究具有抗噪能力的鲁棒性压缩感知系统具有重要意义,也是压缩感知技术能否真正实用的关键之一。本文首先介绍了压缩感知的基本概念,然后分析各种噪声对压缩感知带来的影响,侧重从语音信号的观测矩阵和重构技术入手,介绍具有鲁棒性的压缩感知投影算子和重构算法,最后对未来可能的研究方向进行了展望。
[Abstract]:Compression sensing technology has a broad application prospect in many fields. The key techniques of compression sensing include the selection of sparse matrix, the construction of observation matrix and the design of reconstruction algorithm. Speech signals are different from general signals and usually have special structural characteristics in observation matrix and reconstruction algorithm. It is difficult to avoid noise in practical applications, but in compression sensing systems, the reconstruction system is nonlinear and sensitive to noise, so it is important to study robust compression sensing systems with anti-noise capability. It is also one of the keys to the real practicability of compression sensing technology. This paper first introduces the basic concept of compression perception, then analyzes the influence of various noises on compression perception, focusing on the speech signal observation matrix and reconstruction technology. The compaction perceptual projection operator and reconstruction algorithm with robustness are introduced. Finally, the possible research directions in the future are prospected.
【作者单位】： 南京邮电大学宽带无线通信与传感网技术教育部重点实验室;
【基金】：国家自然科学基金(61671252,61271335)资助项目 南京邮电大学校科研基金(NY214191)资助项目
【分类号】：TN912.3
[Abstract]:Compression sensing technology has a broad application prospect in many fields. The key techniques of compression sensing include the selection of sparse matrix, the construction of observation matrix and the design of reconstruction algorithm. Speech signals are different from general signals and usually have special structural characteristics in observation matrix and reconstruction algorithm. It is difficult to avoid noise in practical applications, but in compression sensing systems, the reconstruction system is nonlinear and sensitive to noise, so it is important to study robust compression sensing systems with anti-noise capability. It is also one of the keys to the real practicability of compression sensing technology. This paper first introduces the basic concept of compression perception, then analyzes the influence of various noises on compression perception, focusing on the speech signal observation matrix and reconstruction technology. The compaction perceptual projection operator and reconstruction algorithm with robustness are introduced. Finally, the possible research directions in the future are prospected.
【作者单位】： 南京邮电大学宽带无线通信与传感网技术教育部重点实验室;
【基金】：国家自然科学基金(61671252,61271335)资助项目 南京邮电大学校科研基金(NY214191)资助项目
【分类号】：TN912.3
【相似文献】
相关期刊论文 前10条
1 李志刚;;一种快速的压缩感知信号重构算法[J];信息技术;2013年06期
2 陈勤;邹志兵;张e，
本文编号：2347149
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2347149.html
