In order to solve problems that existed in the underground positioning systems, e.g., low precision of node positioning, and potential conflicts during data transfer, this study proposed an improved received signal strength indicator (RSSI) underground positioning algorithm which is based on wireless sensor networks. This algorithm analyzed and improved the distance measuring principles of the received signal strength indicator. Iteration computation was applied to calculate data, resulting in the improvement of positioning precision and a considerable decrease of the positioning error of the node position. Accuracy and stability of the data communication among different nodes were also guaranteed and real-time positioning monitoring of the underground worker was realized. Therefore, highly efficient rescue and the safety coefficient of the underground operation were guaranteed. The simulation experiment showed that this algorithm was able to finish complex computation process with fewer parametric variables, thus effectively reducing the communication overhead among network nodes, extending the life cycle of the network, and strengthening the robustness and stability of the entire network system.