基于压缩感知的语音加密

Speech encryption based on compressed sensing

语音通信是人们获取和交换信息的一种重要方式。随着社会信息化的发展,语音通信的机密性越来越受到重视。传统的语音加密方式都遵循经典的香农-奈奎斯特采样定理,在采样时会采集一些冗余语音数据,浪费了采样资源。为了节约采样资源并简化语音加密流程,采用压缩感知理论对语音进行加密。该方法不仅可以对语音进行加密而且也同时实现了语音的有效采样。仿真实验表明,利用压缩感知进行语音加密是切实有效的,其既可以节约采样资源又节省了存储空间。

Speech communication is an important way for people to acquire and exchange information. With the development of social information, the confidentiality of speech communication becomes more and more attractive. Traditional speech encryption methods obey the classical Shannon-Nyquist sampling theorem. When sampling with Shannon-Nyquist theorem, some redundant speech data will be acquired, which wastes the sampling resources. In order to save sampling resources and simplify the speech encryption process, compressed sensing theory was employed to encrypt speech in this paper. This method can not only realize speech encryption,but it can also sample speech effectively at the same time. The simulation results show that speech encryption with compressed sensing is effective and practical, which can save not only the sampling resources but also storage space.