To reduce effect of measurement error of received signal strength indication（ RSSI） on localization precision of node and improve robustness of algorithm

propose a new wireless sensor networks（ WSNs）localization algorithm based on interval analysis. This algorithm uses bootstrap sampling method in ranging phase to eliminate measurement error of RSSI and build confidence intervals of ranging; in positioning phase

solve set of feasible solution sets of unknown node position coordinates

combined with B-box positioning method and set membership identification

then obtain estimated position of unknown nodes by grid scanning. Feasibility of algorithm is verified by simulation analysis

compared this algorithm with the traditional localization algorithm in error

the experimental results show that this algorithm has higher precision and better robustness.