الفهرس | Only 14 pages are availabe for public view |
Abstract This thesis addresses and studies the problem of Blind Source Separation BSS. In recent years, the importance of BSS techniques has been grown due to limited knowledge of their properties and how they contribute to each sensor output and the explosive growth of using Multiple-Input Multiple-Output MIMO systems in different applications and cases. The term blind refers to the fact that there is no explicit information about the mixing process or about source signals. Applications of BSS are numerous such as biomedical signal analysis, speech separation, image processing and feature extraction, wireless communication and sensor array processing, geophysical data processing, data mining and financial applications. Depending on the application, multiple antennas, multiple microphones or multiple biomedical sensors are used for the data acquisition.This thesis addresses and studies the problem of Blind Source Separation BSS. In recent years, the importance of BSS techniques has been grown due to limited knowledge of their properties and how they contribute to each sensor output and the explosive growth of using Multiple-Input Multiple-Output MIMO systems in different applications and cases. The term blind refers to the fact that there is no explicit information about the mixing process or about source signals. Applications of BSS are numerous such as biomedical signal analysis, speech separation, image processing and feature extraction, wireless communication and sensor array processing, geophysical data processing, data mining and financial applications. Depending on the application, multiple antennas, multiple microphones or multiple biomedical sensors are used for the data acquisition.In this thesis, the de-nosing technique is applied to separate mixed signals or images. Preprocessing for mixed signals and images will be investigate and study its effect in the performance of de-noise process. Moreover, post processing of the extracted signals or images will be investigate and compared to the preprocessing technique for improvement the quality of separated signals or images. The curvelet de-nosing is applied as pre-processing and post processing. Also, BSS algorithm using different frequency transforms are applied and studied their performance Ridgelet Transform RT, Discrete Cosine Transform DCT, Discrete Sine Transform DST and Discrete Wavelet Transform DWT. These transforms are proposed and compared to the time domain using ICA performance the separation of mixed signals and images. Moreover, a wavelet based ICA approach using Kurtosis for BSS is proposed. In this approach, the observations are transformed into elementary forms using Wavelet Packets Decomposition WPD and choose the best band by Kurtosis criteria. |