The invention relates to a method for time synchronization in a communication network having a plurality of nodes (MA, SL1, SL2, SLn), the nodes (MA, SL1, SL2, SLn) comprising a first node (MA) and one or more second nodes (SL1, SL2, SLn), the first node (MA) generating first cycle counting states according to a reference cycle frequency and the one or more second nodes (SL1, SL2, SLn) each generating second cycle counting states according to an internal cycle frequency, wherein time synchronization is carried out in consecutive synchronization cycles, in which synchronization messages (SY (k)) are consecutively transferred from a node (MA, SL1, SL2, SLn) to a further node (MA, SL1, SL2, SLn) starting from the first node (MA) and a synchronization message (SY (k)) transmitted by a node (MA, SL1, SL2, SLn) contains a piece of information (SI, SI'), which is used for time synchronization in the node (SL1, SL2, SLn) receiving the synchronization message (SY (k)). The method is characterized in that, in a particular second node (SLn), a time synchronization is performed based on an estimate of a first cycle counting state in combination with a linear-quadratic regulator (LQR), the estimate preferably being based on a Kalman filter (KF). In this manner, a synchronized time is obtained in the form of a controlled first cycle counting state (CF). The method is suitable in particular for time synchronization in communication networks for industrial automation systems, wherein the processes carried out in the automation system must be exactly coordinated.