Brain activity produces electroencephalogram signals, which consists of some of vital signs of neurological disorders and very much helpful in Brain Computer Interfacing(BCI). These signals can be acquired by placing the electrodes on the scalp at specified positions and exists in few hundreds micro volts range with a frequency band of DC-100 Hz. Acquisition of these signals is mainly suffers from different unwanted signals (noise) results in less signal information for identification. In this paper modeling of EEG signals has been done with wavelets and Iterative Soft Thresholding (IST)algorithm. Results reveal that the enhancement of these signals gives the exact features without losing the signal information.