Abstract: In view of the characteristic of the energy constraint of the wireless sensor networks, a method is proposed for the estimation of the number of signal sources in wireless sensor networks, which is based on clustering and decision-level data fusion. And a new cluster head selection algorithm based upon nodes’ residual energy is proposed. In the method proposed in this paper, the first is to divide the sensor networks into clusters with each cluster independently working on the estimation of the number of signal sources so as to generate the final results. Secondly, the results from each cluster at the fusion center are fused to obtain the estimates of the number of signal sources. The simulation results demonstrate that the methods of estimating the number of signal sources and cluster head selection algorithm proposed in this paper can efficiently reduce energy consumption for network transmission, and increase network life on the premise that the reliability of the estimation is ensured.