In this work, we propose a new approach to localize multiple concurrent sources using a distributed acoustic sensor network. Only two node-arrays are required in this sensor network, and each node-array consists of only two widely spaced sensors. Firstly, direction-of-arrivals (DOAs) of multiple sources are estimated at each node-array by utilizing a new pooled angular spectrum proposed in this paper, which can implement the spatial aliasing suppression effectively. Based on minimum variance distortionless response (MVDR) beamforming and the DOA estimates of the sources, the time-frequency spectra containing the corresponding energy distribution features associated with those sources are reconstructed in each node-array. Then, scale invariant feature transform (SIFT) is employed to solve the DOA association problem. Performance evaluation is conducted with field recordings and experimental results prove the effectivity and feasibility of the proposed method.