An Enhanced Impulsive Noise Suppression Method Based on Wavelet Denoising and ICA for Power Line Communication

Aiming at the problem of noise suppression in power lines, traditional noise suppression methods need to know prior knowledge and other defects. In this paper, blind source separation methods that do not need prior knowledge are selected. In the case of low signal-to-noise ratio, the basic independent component analysis algorithm has poor denoising effect. Therefore, this paper proposes a joint independent component analysis algorithm based on Wavelet denoising and Power independent component analysis (WD-PowerICA). In this work, firstly, the pseudo observation signal is constructed by weighted processing, and the blind separation model of single channel is transformed into a multi-channel determined model. Then, the proposed WD-PowerICA algorithm is used to separate noise and source signals. Finally, the simulation results demonstrate that the proposed algorithm in this paper can effectively separate noise and source signal under low SNR. At the same time, the stronger the α pulse noise is, the closer the WD-PowerICA separated signal is to the source signal. The proposed algorithm is better than the state of the art PowerICA algorithm.

Cooperation of CUDA and Intel multi-core architecture in the independent component analysis algorithm for EEG data

ObjectivesThe electroencephalographic signal is largely exposed to external disturbances. Therefore, an important element of its processing is its thorough cleaning.MethodsOne of the common methods of signal improvement is the independent component analysis (ICA). However, it is a computationally expensive algorithm, hence methods are needed to decrease its execution time. One of the ICA algorithms (fastICA) and parallel computing on the CPU and GPU was used to reduce the algorithm execution time.ResultsThis paper presents the results of study on the implementation of fastICA, which uses some multi-core architecture and the GPU computation capabilities.ConclusionsThe use of such a hybrid approach shortens the execution time of the algorithm.

Research of improved fast independent component analysis algorithm in rectal diagnosis signal preprocessing

With the maturity of artificial organ technology, the use of artificial anal sphincters was proposed to help patients who suffered anal incontinence for various causes reconstruct rectal perception, monitor rectal pressure and diagnose rectal lesion. Aimed at the lack of signal pretreatment in the artificial anal sphincter system, we find a way to solve it, that is, the multi-dimensional reconstruction of the intestinal one-dimensional pressure signal sequence by using phase space reconstruction, and the separation of the reconstructed signal by using the improved fast independent component analysis algorithm. We did some relevant experiments, further extracted the features of the isolated rectal signal, and used back propagation neural network to diagnose the rectal lesions. Experiments show that the method can pretreat the rectal signal, and further analyze the separated signal to diagnose of rectal function. The improved fast independent component analysis algorithm has few iterations, fast convergence, short run time, low requirements on initial weights and good diagnosis. This study lays a foundation for the diagnosis of rectal function by using artificial anal sphincters.