The OSI (Open Systems Interconnection) Reference Model is a computer networking model developed by ISO (the International Organisation for Standardisation), which describes the flow of data within a network beginning with the physical connections and continuing to end user applications. There are seven layers in The OSI Reference Model, Physical, Data Link, Network, Transport, Session, Presentation and Application (Figure A-3).

Figure A-3 Layers in OSI model

The Physical Layer: This is the lowest layer that transfers and regulates the data bits to the physical cable. It specifies electrical (voltages, timing), mechanical (connectors, media), functional (restrictions, framing, multiplexing), procedural (handshaking, flow control), and quality of service (info to Link layer) features. It defines the connecting type between cable and the network adapter card and the transmission method as well as the way of bit synchronisation and checking. T1, T3, ATM, and SONET are examples of. this protocols.

The Data Link Layer: This packages blocks of raw bits from the Physical Layer into frames that create a logical, structured packet to hold data. It transfers frames from one computer to others with the necessary synchronisation, error control, and flow control. It has the features of addressing, error detection and recovery, message ordering, back pressure to control flow, alarms and diagnostics and quality of service to the Network layer. Ethernet, 802.3, 802.5 and ARCnet are examples of this layer’s protocols. The disadvantage of these protocols is that while the address is included, as the geographic location of the receiving station is not known, the messages must be broadcast across the entire network. This layer contains two pieces:

Logical Link Control (LLC) and Media Access Control (MAC).

The LLC is related to flow control while the MAC determines the input /output of the wire and creates the frames.

The Network Layer: It addresses messages and translates logical addresses and names into physical addresses. It also determines the network transmission path and deal with traffic problems like switching, routing and data packet congestion controlling. In addition, it segments messages to be transmitted on various services, security and determines priorities. The example of this protocol is Internet Protocol (IP). Routers also work at this layer. However it can not insure data integrity for an end-to-end transmission and this is the main problem of this layer.

The Transport Layer: This layer recognises and recovers errors by ACK (positive acknowledgement) and NAK (negative acknowledgement) between two communicating ends. It also re-packages long messages into smaller packets for transmission and rebuilds them into the original message at the receiving end. It multiplexes and controls flow to the Session layer as well. Good examples of these protocols are TCP (Transmission Control Protocol), UDP (User Diagram Protocol). The disadvantage of this layer is that if the connection is broken, the transfer stops.

The Session Layer: The session layer manages the data flow between computers or devices. When in a multi-user environment, it manages timing of the data transmission and time multiplexing. If an unexpected abort occurs, it sets up the synchronisation, and re-synchronisation of two applications. The understandable formats of their communication are not guaranteed in this layer.

The Presentation Layer: This layer supplies a common language for the application layer. Compression and decompression often on this layer. It also provides data encryption for the security issues. A drawback of this layer might be the incompatibility of presentation layers between vendors.

The Application Layer: This layer contains users' applications to access network services. The applications of file transfers, database access and electronic mail are supported on this layer