Switch based industrial Ethernet networks

Abstract

When Ethernet was first invented by Dr Robert Metcalfe in the mid-1970s, communication at the physical layer was based on coaxial cables. Steadily increasing technical demands in terms of data rates and network topologies have fuelled a continuous evolution in the physics of data transfer. Today it is the shielded twisted pair (STP) which is the preferred transmission medium. However, the access procedure which regulates the exchange of data over these lines, i.e. the procedure which defines how data stimulates the transmission medium, remains the same: CSMA/CD (carrier sense multiple access with collision detection). When the need arises at an Ethernet node to exchange data - to send an e-mail, for example - the Ethernet controller checks whether there is already data traffic on the line (Carrier Sense, CS). If the check proves negative, the node begins the transmission sequence. During this sequence, the node continuously monitors whether another Ethernet station has begun a data transmission during this timeframe which might cause data to collide and be destroyed.