In the present study, the use of the multi-layer decomposition using wavelet for denoising of the non-stationary signal, i.e., the well log signal based on three wavelets such as Symlet wavelet (Sym8), biorthogonal wavelet (bior6.8) and Daubechies wavelet (db8) are used to identify lithology in the Jharia coalfield region. This work mainly focuses on the development of unconventional methods for signal denoising using wavelets. The well log data of the Jharia coalfield region includes gamma ray log (GR), resistivity log (shallow resistivity log, medium resistivity log), density log, sonic log, and neutron log (NPHI) which are used as the signal to which de-noising has been applied. This study describes the lithology of the Jharia coalfield region using the wavelet denoising effect, power spectrum analysis of the denoised signal, and lithology identification using denoised data. Lithology identification using de-noised signals is used to delineate three lithologies such as sand, shale, and coal to understand the performance of each wavelet decomposition method. Different parameters like the ‘heursure’ soft thresholding and 8-level decomposition are used for denoising the signal. Lithology obtained from the denoised signal using the 'sym8' wavelet gives lithofacies differences compared to other wavelets, information obtained from the 'sym8' waveform is more applicable to identify the reservoir properties, stratigraphic sequence, and sedimentary facies.