Bearings are the most important part of any machinery used for domestic purpose or in big industries. Accurate working of these machines and devices relies on resistance free running of the bearings. In industrial applications, these bearings act as the most crucial machine component because they have to work under high load and speed; and defects in bearing, until noticed in time, can even lead to failure of the machinery. Therefore, identification of these defects is important for condition monitoring and quality inspection of bearings. Different methods are used for detection and diagnosis of bearing defects; they are classified as vibration measurements, acoustic measurements, temperature measurements, and wear debris analysis. For monitoring the condition and diagnosing the fault that may occur in REB, the vibration analysis technique is widely used because of its reliability and accuracy. In this work, the vibration signal of the REB under the normal and faulty condition from the database of Case Western Reserve University is taken for detailed analysis. Useful data from the raw signal is mined using Time Synchronous Averaging (TSA) and then TSA signal is processed using Discrete Wavelet Transform (DWT). The features of the DWT signal (D3) are extracted and further processed using Feed forward Backpropagation Artificial Neural Network. Moreover, comparison between existing work and this work has been shown.