介绍了一种在标准 1 2导联心电图中消除噪声信号的正交化方法。使用一种基于奇异值分解 (SVD)的算法 ,将信号分解为两个时正交的子空间即一个包含了 ECG信号 ,另一个则包含了人工噪声 ,诸如基线漂移 (BW)和肌电信号 (EMG) ,这种方法利用了存在于 1 2导联心电图中的冗余。在完整丢失信道的重建方面该方法也适用。最后 ,给出了这种方法的在线实现 An orthogonalization method to eliminate noise signals in a standard 12-lead electrocardiogram was introduced. Using an algorithm based on Singular Value Decomposition (SVD), the signal is decomposed into two orthogonal subspaces, one containing ECG signals and the other one containing artificial noise such as baseline wander (BW) and myoelectric signals (EMG), a method that takes advantage of the redundancy that exists in a 12-lead electrocardiogram. This method also applies in the reconstruction of completely lost channels. Finally, an online implementation of this method is given