Feature Extraction, Risky Classifications and Fault Diagnosis on Rolling Bearings of EEG Signals Denoised using Stationary Wavelet Transform of Patient Monitoring and IoT

Background: One of the primary causes of sleep disorders, depth of anesthesia, coma, encephalopathy’s and brain death in the world is epilepsy. EEG is most often used to diagnose epilepsy which causes the abnormalities in EEG readings. Different high-resolution anatomical imaging techniques are used to detect these abnormalities like MRI, PET, CT, etc. Methods: SWT method will be an enhanced system from wavelet transform. It may be fit for the signal for time-invariant on the break down also enhance those force of indicator denoising. SWT additionally employs upsampled technique at every level of decay for those signs. The decay of SWT produces the coefficients from claiming close estimation and points in the same length. The DWT will be actualized by a channel bank that decomposes those indicators over progressive coarser approximations. The output of the low pass and high pass filter coefficients is decomposed to the next level and further proceeds up to N levels. The yield of the wavelet decay may be that close estimation and the point of interest coefficients which would get to each level of decay. This system consists of five main processing steps: acquisition, pre-processing, feature extraction, feature selection and classification. Results: This paper overviews some of the current state-of-the-art IOT systems and presents the statistical-based algorithm used for each processing step. Conclusion: This paper also provides a comparison of the performance of the existing approaches.