In this study, a novel algorithm based on the MUSIC algorithm is proposed for the localization of a broadband source in the time domain. The source is in the close range of the randomly distributed sensors. The emitter signal is arrived at the sensors with different amplitudes and phase delays, therefore the received signals at the sensors are the noisy variation of the one another. The data collected by the each sensor are provided from the observations. The measure of the similarity of two sensor data can be attained by calculating the cross correlation of that collected data in the time domain and then the time difference between these two signal can be determined. After this step, one of the sensors is chosen as the reference point and the observation time delays are compensated with reference to the chosen sensor. Finally, the emitter location is estimated efficiently by using the MUSIC algorithm. In addition, the simulation results that supported the performance of the algorithm are given in the paper.