This document summarizes research on reference broadcast time synchronization in wireless sensor networks. It discusses how previous protocols like flooding time synchronization and gradient time synchronization have drawbacks like slow propagation speed and inability to maintain synchronization when nodes crash. It then introduces the reference broadcast synchronization protocol which chooses a reference node using an agreement algorithm and broadcasts time information to synchronize the network. It presents the system architecture and algorithm for how reference broadcast synchronization works to flood time information, perform synchronization based on messages from the parent node, and timestamp events in the network. Evaluation results showing the protocol implemented on line and distributed topologies are also included.