基于模糊递归和最优硬阈值的局部投影降噪算法  

Local projective noise reduction algorithm based on fuzzy recurrence and optimal hard threshold

作　　者：王东 崔忠马 陈文东 舒勤[1] WANG Dong;CUI Zhongma;CHEN Wendong;SHU Qin(School of Electrical Engineering,Sichuan University,Chengdu 610065,China;Beijing Institute of Remote Sensing Equipment,Beijing 100084,China)

机构地区：[1]四川大学电气工程学院,四川成都610065 [2]北京遥感设备研究所,北京100084

出　　处：《系统工程与电子技术》2023年第3期621-628,共8页Systems Engineering and Electronics

摘　　要：针对原始局部投影降噪算法的局部邻域选取问题和子空间划分问题,提出了一种基于模糊递归图和最优硬阈值准则的局部投影降噪算法。首先,利用模糊递归图确定局部邻域范围;然后,再对邻域矩阵进行奇异值分解,并利用最优硬阈值准则对局部邻域的信号子空间和噪声子空间进行划分;最后,进行投影去噪。Lorenz信号的仿真结果表明,所提方法能够提高信噪比并降低均方误差,恢复原始吸引子的形态结构。对实测含噪心电图信号进行处理后,信噪比显著提高,验证了本方法的有效性。For the local neighborhood selection problem and subspace partition problem of the original local projective(LP)noise reduction algorithm,an improved LP noise reduction algorithm based on fuzzy recurrence plot and optimal hard threshold criterion is proposed.Firstly,the local neighborhood is determined by fuzzy recurrence plot.Then,the singular value decomposition(SVD)is implemented to local neighborhood matrix,and signal subspace and noise subspace in local neighborhood is partitioned by optimal hard threshold criterion.Finally,the projection for denoising is performed.The denoising results of Lorenz signal show that the proposed algorithm can improve the signal to noise ratio,reduce the mean square error,and recovery the morphological structure of the original attractor.After processing the measured noisy electrocardiogram(ECG)signal,the signal to noise ratio is significantly improved,which shows the effectiveness of the proposed algorithm.

关 键 词：局部投影 模糊递归图 最优硬阈值准则 降噪 混沌信号 

分 类 号：TN911.7[电子电信—通信与信息系统]