This paper proposes a passive islanding detection technique for microgrid. The proposed technique relies on capturing the underlying signatures of a wide variety of system events on critical system parameters through the utilization of pattern recognition tools for islanding detection in a microgrid. The proposed technique is tested on a microgrid model implemented on IEEE 13-node distribution feeder system under a wide variety of system operating states. Results from test case study have been analyzed to evaluate the effectiveness of the proposed method. Case study results indicate that the proposed method can detect islanding events with high accuracy and reliability.