Abstract: Today, Cluster based routing protocols are the most useful schemes for extending Wireless Sensor Networks lifetime through dividing the nodes into several clusters and electing of a local cluster head for aggregating/fusing of cluster nodes data and transmitting a packet to Base Station. However, there are several energy efficient cluster-based methods in the literature; most of them used the topological neighborhood or adjacency as main parameter to form the clusters. This paper presents a new centralized adaptive Energy Based Clustering protocol through the application of Self-organizing map neural network (called EBCS) which can cluster sensor nodes, based on multi parameters, energy level and coordinates of sensor nodes. We applied some maximum energy nodes as weights of SOM map units, so that the nodes with higher energy attract the nearest nodes with lower energy levels. Thus, a cluster may not necessarily Today, Cluster based routing protocols are the most useful schemes for extending Wireless Sensor Networks lifetime through dividing the nodes into several clusters and electing of a local cluster head for aggregating/fusing of cluster nodes data and transmitting a packet to Base Station. However, there are several energy efficient cluster-based methods in the literature; most of them used the topological neighborhood or adjacency as main parameter to form the clusters. This paper presents a new centralized adaptive Energy Based Clustering protocol through the application of Self-organizing map neural network (called EBCS) which can cluster sensor nodes, based on multi parameters, energy level and coordinates of sensor nodes. We applied some maximum energy nodes as weights of SOM map units, so that the nodes with higher energy attract the nearest nodes with lower energy levels. Thus, a cluster may not necessarily contain adjacent nodes. The new algorithm enables us to form energy-balanced clusters and distribute energy consumption equally. Moreover, we proposed a new cost fuction to incorporate different useful criteria for election of Cluster head nodes with energy efficiency. Simulation results for two different scenes and comparison of them with previous similar protocols (LEACH and LEA2C) prove that the new algorithm is able to extend the lifetime of the network and preserve more network coverage with the same number of dead nodes. In addition, the effectiveness of new cost function is apparently shown.