Abstract: In a cognitive radio (CR) context, spectrum detection is a key task for searching white space and avoiding harmful interference to primary users. A spectrum sensing scheme of time domain energy detector based on discrete wavelet transform (DWT) is proposed for sensing a narrow band PU signal in the channel accessed by SU. Firstly DWT is used to decomposed the received signals, and one of the detail spaces may show the irregularities in channel signals, then calculate the energy of the detail space in time domain. The proposed scheme which has lower complexity can be performed for multi-resolution spectrum sensing and exhibits a good detection performance. Because of easy to implement and no requirement of any priori information about signals, it can adapt to detect various unknown signals. By simulations, wireless microphone signals and terrestrial digital TV channels are detected for verifying our scheme.