The past decade has observed the development of wireless sensor networks (WSNs) being used in a great number of applications. For the better routing and energy efficiency, WSNs are partitioned into clusters. Partitioning was done to minimize the distances between the source node and the sink. In order to achieve better partitioning, graph partitioning methods can be used. Such techniques are mostly used in distributed environment and applications, but they are not much efficient in wireless sensor networks due to the dynamic topology and multi-hop transmission. In this paper, few existing graph partitioning methods have been discussed and studied which may be implemented for WSN partitioning. A novel partitioning method is tested in this paper, which is better in terms of efficiency in WSN partitioning.