Techniques for synchronization of clocks in nodes in a network are described. In one example, a node times or measures a synchronization timeout period. During the synchronization timeout period, the node may hear a beacon. In that event, the node may reset its clock using a time indicator found within the beacon. If the node does not hear a beacon before the end of the synchronization timeout period, the node may send a beacon request to one of its parents. In response, the parent node will broadcast a beacon, which may be heard by other nodes in the vicinity of the parent node. Upon receipt of the beacon and an included time indicator, the node will update its clock. Upon clock update, another synchronization timeout period is then started and the cycle is repeated, thereby maintaining synchronization of the clock with clocks of other nodes.