一种三阶多项式相位信号去噪的字典学习算法

发布时间：2019-02-11 12:14

【摘要】：在加性高斯白噪声的影响下,对于三阶多项式相位信号(CPS),经典的字典学习算法,如K-means Singular Value Decomposition(K-SVD),递归最小二乘字典学习算法(RLS-DLA)和K-means Singular Value Decomposition Denoising(K-SVDD)得到的学习字典,通过稀疏分解,不能有效去除信号的噪声。为此,该文提出了针对CPS去噪的字典学习算法。该算法首先利用RLS-DLA对的字典进行学习;其次采用非线性最小二乘(NLLS)法修改了该算法对字典更新的部分;最后对训练后的字典通过对信号的稀疏表示得到重构信号。对比其它的字典学习算法,该算法的信噪比(SNR)值明显高于其它算法,而均方误差(MSE)显著低于其它算法,具有明显的降噪效果。实验结果表明,采用该算法得到的字典通过稀疏分解,信号的平均信噪比比K-SVD,RLS-DLS和K-SVDD高出9.55 dB,13.94 dB和9.76 dB。
[Abstract]:Under the influence of additive Gao Si white noise, for third-order polynomial phase signal (CPS), classical dictionary learning algorithm, such as K-means Singular Value Decomposition (K-SVD, The learning dictionaries obtained by recursive least square dictionary learning algorithm (RLS-DLA) and K-means Singular Value Decomposition Denoising (K-SVDD) can not effectively remove the noise of signals by sparse decomposition. Therefore, a dictionary learning algorithm for CPS denoising is proposed. The algorithm first uses RLS-DLA to learn the dictionary, then modifies the updating part of the dictionary by using the nonlinear least square (NLLS) method. Finally, the trained dictionary is reconstructed by sparse representation of the signal. Compared with other dictionary learning algorithms, the signal-to-noise ratio (SNR) of this algorithm is significantly higher than that of other algorithms, while the mean square error (MSE) is significantly lower than that of other algorithms. The experimental results show that the average signal-to-noise ratio of the dictionary obtained by this algorithm is 9.55 dB,13.94 dB and 9.76 dB. higher than that of K-SVDN RLS-DLS and K-SVDD by sparse decomposition.
【作者单位】：重庆大学飞行器测控与通信教育部重点实验室;重庆电子工程职业学院;
【基金】：国家自然科学基金(51377179) 中央高校基本科研业务费(CDJZR12160020) 重庆教委项目(KJ120510)资助课题
【分类号】：TN911.4

【共引文献】