The Parallel Redundancy Protocol for industrial IP networks

The “Parallel Redundancy Protocol” (PRP) according to IEC 62439-3 realizes active network redundancy by packet duplication over two independent networks that operate in parallel. It has been specifically designed for industrial networks to meet highest availability requirements. In case of a single network failure, seamless redundancy is provided for data communication between PRP nodes that are connected to both networks. However, PRP was designed as a layer 2 Ethernet protocol that is transparent to higher protocol layers. In its current version, PRP is not able to support IP routing. This is because an IP router would change the source MAC address field of the Ethernet header which is used by a PRP receiver node for duplicate detection. In this work we propose and discuss a novel approach that keeps the PRP duplicate identification information across IP router boundaries.