摘要
本文针对含噪语音压缩感知在低信噪比时重构语音性能差的问题,提出了一种自适应快速重构算法。该算法将行阶梯观测矩阵与一种新型的快速重构算法结合,并根据含噪语音信号的信噪比自适应选择最佳重构参数,使得在重构语音的同时提高了重构信噪比。算法实现简单快速,且不需要预先计算信号的稀疏度。实验结果表明:在低信噪比时,自适应快速重构算法的重构性能优于基追踪算法和快速重构算法,且重构速度快于快速重构算法和基追踪算法。
An adaptive fast recovery algorithm is proposed to solve the problem of poor reconstruction performance for compressed sensing of noisy speech with low signal-to-noise ratio. This method combines row echelon measurement matrix and a new fast recovery algorithm,adaptively selects the optimal reconstruction parameters according to the signal-to-noise ratio of noisy speech signal,and enhances the signal-to-noise ratio while reconstructing the speech. The adaptive fast recovery algorithm is simple and fast and does not require pre-calculated signal sparsity. Simulation experiment results demonstrate that the proposed algorithm outperforms basis pursuit algorithm and fast reconstruction algorithm,and faster than basis pursuit algorithm and fast recovery algorithm.
引文
[1]Donoho D L.Compressed sensing[J].IEEE Transactions on Information Theory,2006,52(4):1289-1306.
[2]Candes E J,Romberg J,Tao T.Robust uncertainty principles:Exact signal reconstruction from highly incomplete frequency information[J].IEEE Transactions on Information Theory,2006,52(2):489-509.
[3]杨真真,杨震.语音压缩感知硬阈值梯度追踪重构算法[J].信号处理,2014,30(4):391-398.Yang Z Z,Yang Z.Hard Threshold Gradient Pursuit Reconstruction Algorithm for Speech Compressed Sensing[J].Journal of Signal Processing,2014,30(4):391-398.(in Chinese)
[4]Mingrui Yang,Frank de Hoog.Orthogonal Matching Pursuit with Thresholding and Its Application in Compressive Sensing[J].IEEE Transactions on Signal Processing,2015,63(20):5479-5486.
[5]孙林慧,杨震.语音压缩感知研究进展与展望[J].数据采集与处理,2015,30(2):275-288.Sun L H,Yang Z.Compressed Speech Sensing for Research Process and Prospect[J].Journal of Data Acquisitions and Processing,2015,30(2):275-288.(in Chinese)
[6]孙林慧,杨震.基于自适应基追踪去噪的含噪语音压缩感知[J].南京邮电大学学报,2011,31(5):1-6.Sun L H,Yang Z.Compressed sensing of noisy speech signal based on adaptive basis pursuit de-noising[J].Journal of Nanjing University of Posts and Telecommunications:Natural Science,2011,31(5):1-6.(in Chinese)
[7]杨真真,杨震.含噪语音压缩与重构的自适应共轭梯度投影算法[J].仪器仪表学报,2012,33(10):2200-2207.Yang Z Z,Yang Z.Adaptive conjugate gradient projection algorithm for noisy speech compression and reconstruction[J].Chinese Journal of Science Instrument,2012,33(10):2200-2207.(in Chinese)
[8]季云云,杨震.脉冲噪声环境下高斯稀疏信源贝叶斯压缩感知重构[J].电子学报,2013,41(2):363-370.Ji Y Y,Yang Z.Bayesian Compressed sensing for Gaussian Sparse Signals in the Presence of Impulsive Noise[J].Acta Electronica Sinica,2013,41(2):363-370.(in Chinese)
[9]叶蕾,杨震,王天荆,等.行阶梯观测矩阵,对偶仿射尺度内点重构算法下的语音压缩感知[J].电子学报,2012,40(3):429-434.Ye L,Yang Z,Wang T J,et al.Compressed sensing of speech signal based on row echelon measurement matrix and dual affine scaling interior point reconstruction method[J].Acta Electronica Sinica,2012,40(3):429-434.(in Chinese)
[10]Andrianiaina Ravelomanantsoa,Hassan Rabah,Amar Rouane.Compressed sensing:A Simple Deterministic Measurement Matrix and a Fast Recovery Algorithm[J].IEEE Transactions on Instrumentation and Measurement,2015,64(12):3405-3413.
[11]Golub G H,Loan C F V.Matrix Computations[M].Third Edition.London,UK:The Johns Hopkins University Press,1996.
[12]Hoda Dehghan,Richard M,Dansereau,Adrian D C.Restricted Isometry Property on Banded Block Toeplitz Matrices with Application to Multi-Channel Convolutive Source Separation[J].IEEE Transactions on Signal Processing,2015,63(21):5665-5676.