The location estimation is a fundamental and essential issue for wireless sensor networks(WSNs). In this paper, we assume that only a few sensor nodes (named as beacon nodes) get their locations by Global Positioning System (GPS) and the remaining nodes without GPS (named as normal nodes) need to estimate their own locations by gathering the nearby neighboring information. Existing works are either too costly or not accurate enough. To improve previous works, we propose a distributed location estimation algorithm for WSNs. In our algorithm, each node without location information only needs to collect the location information of neighboring nodes and use simple computation to estimate its location. Besides, we improve the accuracy of the normal node?s estimative region by discarding the communication area of the beacon node (named as the farther neighboring beacon node), which does not cover the normal node, from the original estimative region. We derive some rules to adjust the estimative region according to the relative location of the normal node and the farther neighboring beacon node. Simulation results show that the proposed algorithm achieves better accuracy of estimative locations.