9.4.4.2 Multi-Hop Synchronization.
Similar to TPSN, FTSP relies on an elected synchronization root to synchronize the network, where root election is based on unique node IDs (i.e., the node with the lowest ID is elected as the root node). The root node maintains the global time and all other nodes in the network synchronize their clocks to that of the root. Synchronization is triggered through a broadcast message by the root node containing its time stamp. All nodes within the communication range of the root can establish synchronization points directly from the broadcast message. Other nodes collect synchronization points from broadcasts of synchronized nodes that are closer to the root. Similar to TPSN, FTSP relies on a root election algorithm to ensure that there is exactly one synchronization root in the network. Every broadcast message contains the unique ID of the root (rootID) and a sequence number (besides the already discussed time stamp). Whenever a node does not receive a synchronization message for a certain amount of time, it declares itself to be the new root. Whenever a node receives a synchronization message with a rootID lower than its own ID, it gives up its root status. A new node joining a network with a lower ID than the rootID will not immediately declare itself as root, but instead wait for a certain period of time to collect synchronization messages and adjust its own clock to the current global time. These techniques ensure that TPSN can handle network topology changes, including mobile nodes.