In wireless sensor networks (WSNs), locations of sensors play a critical role in many applications. For some applications, the nodes are constantly moving in the network. Therefore, the positions of nodes in mobile WSNs can not be determined by current static localization algorithms. To solve this problem, a new centralized localization algorithm called node cooperation based support vector machine localization (NCBSVML) algorithm for mobile wireless sensor networks is proposed. NCBSVML algorithm classified the network into many classes, and justifies which class is node in. Then the NCBSVML algorithm determines the node location by filtering the impossible locations based on new observations. Our approach does not require additional hardware and works even when the movement of anchors and nodes is uncontrollable. The properties of our technique are analyzed from simulations. Our scheme outperforms the best known mobile localization schemes in accuracy and precision under a wide range of conditions.© (2011) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.