Wireless sensor network (WSN) always comes up with the need of deploying either mobile or immobile sensor nodes or both. Wireless communication among these nodes is crucial and it requires identifying the location of these nodes within a specific region. Global positioning system (GPS) is widely used for location tracking. However, when it comes to WSN, GPS has its limitations, due to its high power consumption and the overhead of additional hardware cost. The research challenge here lies in the efficient location tracking of wireless sensor nodes, especially in closed indoor and outdoor environments. This paper comes up with a simple and easy-to-implement technique using artificial neural networks (ANNs) to manipulate the location of the sensor nodes. In this paper, the back-propagation network training algorithm for providing supervised learning to multilayer perceptron is generalized to synthesize the WSN and gives out 2D Cartesian coordinates of the nodes. The technique is both cost-efficient and achieves 98% accuracy. © 2018 IEEE.