Before delving into the mechanics of how information is transferred between computers, you must grow familiar with the terminology used to describe the transmitted data. Many of the layers of the OSI Reference Model use their own specific terms to describe data transferred back and forth. As this information is passed from higher to lower layers, each layer adds information to the original data typically a header and possibly a trailer. This process is called encapsulation. Generically speaking, the term protocol data unit (PDU) is used to describe data and its overhead. Table as below describes the terms used at the various layers of the OSI Reference Model. For instance, as data is passed from the session layer to the transport layer, the transport layer encapsulates the data PDU in a transport layer segment. For TCP and UDP in the TCP/IP protocol stack, the transport layer only adds a header.
As the PDU information is passed down, each layer adds its own header and, possibly, trailer.
Once the physical layer is reached, the bits of the data link layer frame are converted into a physical layer signal a voltage, light source, radio wave, or other source according to the type of physical medium that is employed. When the destination receives the information, it goes through a reverse process of de-encapsulating information basically stripping off the headers of the PDU information at each layer as the information is passed up from layer to layer of the OSI Reference Model.
Figure as below shows an example of the process used for encapsulating and deencapsulating PDUs as data is passed down and back up the OSI Reference Model. In this example, you can see how the application, presentation, and session layers create the data PDU. As this information is passed down from layer to layer, each layer adds its own header.
Going Down the Protocol Stack
In the figure, The first thing that occurs on PC-A is that the user, sitting in front of the computer, creates some type of information, called data, and then sends it to another location (PC-B). This includes the actual user input (application layer), as well as any formatting information (presentation layer). The application (or operating system), at the session layer, then determines whether or not the data’s intended destination is local to this computer (possibly a disk drive) or a remote location. In this instance, the user is sending the information to PC-B. We’ll assume that the user is executing a telnet connection.
The session layer determines that this location is remote and has the transport layer deliver the information. A telnet connection uses TCP/IP and reliable connections (TCP) at the transport layer, which encapsulates the data from the higher layers into a segment.With TCP only a header is added. The segment contains such information as the source and destination port numbers. As you may recall from the section Connection Multiplexing, the source port is a number above 1,023 that is currently not being used by PC-A. The destination port number is the well-known port number (23) that the destination will understand and forward to the correct application.
The transport layer passes the segment down to the network layer, which encapsulates the segment into a packet. The packet header contains layer-3 logical addressing information (source and destination address), as well as other information, such as the upper-layer protocol that created this information. In this example, TCP created this information, so this fact is noted in the packet, and PC-A places its IP address as the source address in the packet and PC-B’s as the destination. This helps the destination, at the network layer, to determine if the packet is for itself and which upper-layer process should handle the encapsulated segment. In the TCP/IP protocol stack, the terms packet and datagram are used interchangeably to describe this PDU.
The network layer then passes the packet down to the data link layer. The data link layer encapsulates the packet into a frame. If you are using IEEE for the data link layer, remember that two encapsulations take place here: one for LLC and one for MAC. This example uses Ethernet as the data link layer medium, and there are two versions of Ethernet: Ethernet II and IEEE 802.3. To make this more complex, assume the data link layer is based on IEEE’s Ethernet implementation. At the LLC sublayer, either an 802.2 SAP or SNAP frame is used. TCP/IP uses a SAP frame type. The important information placed in the SAP frame header is which network layer protocol created the packet: IP. The 802.2 SAP frame is then passed down to the MAC sublayer, where the 802.2 frame is encapsulated in an 802.3 frame. The important components placed in the 802.3 frame header are the source and destination MAC addresses. In this example, PC-A places its MAC address in the frame in the source field and PC-B’s MAC address as the destination.
The data link layer frame is then passed down to the physical layer. At this point, remember that the concept of PDUs is a human concept that we have placed on the data to make it more readable to us, as well as to help deliver the information to the destination. However, from a computer’s perspective, the data is just a bunch of 1’s and 0’s, called bits. The physical layer takes these bits and coverts them into a physical property based on the cable or connection type. In this example, the cable is a copper cable, so the physical layer will convert the bits into voltages: one voltage level for a bit value of 1 and a different voltage level for a 0.
Going Up the Protocol Stack
For sake of simplicity, assume PC-A and PC-B are on the same piece of copper. Once the destination receives the physical layer signals, the physical layer translates the voltage levels back to their binary representation and passes these bit values up to the data link layer.
The data link layer takes the bit values and reassembles the original 802.3 frame. The NIC, at the MAC layer, examines the FCS to make sure the frame is valid and examines the destination MAC address to ensure that the Ethernet frame is meant for itself. If the destination MAC address doesn’t match its own MAC address, or is not a multicast or broadcast address, the NIC drops the frame. Otherwise, the NIC processes the frame: it strips off the 802.3 frame and passes the 802.2 frame up to the LLC sublayer. The LLC sublayer examines the SAP value to determine which upperlayer protocol at the network layer should process the encapsulated packet. In this case, the LLC sees that the encapsulated packet is a TCP/IP packet, so it strips off (de-encapsulates) the LLC frame information and passes the packet up to the TCP/IP protocol stack at the network layer. If this were an encapsulated IPX packet, the LLC would pass the encapsulated IPX packet up to the IPX protocol stack at the network layer.
The network layer then examines the logical destination address in the packet header. If the destination logical address doesn’t match its own address or is not a multicast or broadcast address, the network layer drops the packet. If the logical address matches, then the destination examines the protocol information in the packet header to determine which protocol should handle the packet. In this example, the logical address matches and the protocol is defined as TCP. Therefore, the network layer strips off the packet information and passes the encapsulated segment up to the TCP protocol at the transport layer.
In example, the user from PC-A was using telnet to transmit information to PC-B, so the destination port number is 23. The transport layer examines this port number and realizes that the encapsulated data needs to be forwarded to a telnet application. If PC-B doesn’t support telnet, the transport layer drops the segment. If it does, the transport layer strips off the segment information and passes the encapsulated data to the telnet application. If this is a new connection, a new telnet process is started up by the operating system.
| Term | OSI Reference Model Layer |
| Data | Application, presentation, and session layers |
| Segment | Transport layer |
| Packet | Network layer (TCP/IP calls this a datagram) |
| Frame | Data link layer |
| Bits | Physical layer |
Once the physical layer is reached, the bits of the data link layer frame are converted into a physical layer signal a voltage, light source, radio wave, or other source according to the type of physical medium that is employed. When the destination receives the information, it goes through a reverse process of de-encapsulating information basically stripping off the headers of the PDU information at each layer as the information is passed up from layer to layer of the OSI Reference Model.
Figure as below shows an example of the process used for encapsulating and deencapsulating PDUs as data is passed down and back up the OSI Reference Model. In this example, you can see how the application, presentation, and session layers create the data PDU. As this information is passed down from layer to layer, each layer adds its own header.
Going Down the Protocol Stack
In the figure, The first thing that occurs on PC-A is that the user, sitting in front of the computer, creates some type of information, called data, and then sends it to another location (PC-B). This includes the actual user input (application layer), as well as any formatting information (presentation layer). The application (or operating system), at the session layer, then determines whether or not the data’s intended destination is local to this computer (possibly a disk drive) or a remote location. In this instance, the user is sending the information to PC-B. We’ll assume that the user is executing a telnet connection.
The session layer determines that this location is remote and has the transport layer deliver the information. A telnet connection uses TCP/IP and reliable connections (TCP) at the transport layer, which encapsulates the data from the higher layers into a segment.With TCP only a header is added. The segment contains such information as the source and destination port numbers. As you may recall from the section Connection Multiplexing, the source port is a number above 1,023 that is currently not being used by PC-A. The destination port number is the well-known port number (23) that the destination will understand and forward to the correct application.
The transport layer passes the segment down to the network layer, which encapsulates the segment into a packet. The packet header contains layer-3 logical addressing information (source and destination address), as well as other information, such as the upper-layer protocol that created this information. In this example, TCP created this information, so this fact is noted in the packet, and PC-A places its IP address as the source address in the packet and PC-B’s as the destination. This helps the destination, at the network layer, to determine if the packet is for itself and which upper-layer process should handle the encapsulated segment. In the TCP/IP protocol stack, the terms packet and datagram are used interchangeably to describe this PDU.
The network layer then passes the packet down to the data link layer. The data link layer encapsulates the packet into a frame. If you are using IEEE for the data link layer, remember that two encapsulations take place here: one for LLC and one for MAC. This example uses Ethernet as the data link layer medium, and there are two versions of Ethernet: Ethernet II and IEEE 802.3. To make this more complex, assume the data link layer is based on IEEE’s Ethernet implementation. At the LLC sublayer, either an 802.2 SAP or SNAP frame is used. TCP/IP uses a SAP frame type. The important information placed in the SAP frame header is which network layer protocol created the packet: IP. The 802.2 SAP frame is then passed down to the MAC sublayer, where the 802.2 frame is encapsulated in an 802.3 frame. The important components placed in the 802.3 frame header are the source and destination MAC addresses. In this example, PC-A places its MAC address in the frame in the source field and PC-B’s MAC address as the destination.
The data link layer frame is then passed down to the physical layer. At this point, remember that the concept of PDUs is a human concept that we have placed on the data to make it more readable to us, as well as to help deliver the information to the destination. However, from a computer’s perspective, the data is just a bunch of 1’s and 0’s, called bits. The physical layer takes these bits and coverts them into a physical property based on the cable or connection type. In this example, the cable is a copper cable, so the physical layer will convert the bits into voltages: one voltage level for a bit value of 1 and a different voltage level for a 0.
Going Up the Protocol Stack
For sake of simplicity, assume PC-A and PC-B are on the same piece of copper. Once the destination receives the physical layer signals, the physical layer translates the voltage levels back to their binary representation and passes these bit values up to the data link layer.
The data link layer takes the bit values and reassembles the original 802.3 frame. The NIC, at the MAC layer, examines the FCS to make sure the frame is valid and examines the destination MAC address to ensure that the Ethernet frame is meant for itself. If the destination MAC address doesn’t match its own MAC address, or is not a multicast or broadcast address, the NIC drops the frame. Otherwise, the NIC processes the frame: it strips off the 802.3 frame and passes the 802.2 frame up to the LLC sublayer. The LLC sublayer examines the SAP value to determine which upperlayer protocol at the network layer should process the encapsulated packet. In this case, the LLC sees that the encapsulated packet is a TCP/IP packet, so it strips off (de-encapsulates) the LLC frame information and passes the packet up to the TCP/IP protocol stack at the network layer. If this were an encapsulated IPX packet, the LLC would pass the encapsulated IPX packet up to the IPX protocol stack at the network layer.
The network layer then examines the logical destination address in the packet header. If the destination logical address doesn’t match its own address or is not a multicast or broadcast address, the network layer drops the packet. If the logical address matches, then the destination examines the protocol information in the packet header to determine which protocol should handle the packet. In this example, the logical address matches and the protocol is defined as TCP. Therefore, the network layer strips off the packet information and passes the encapsulated segment up to the TCP protocol at the transport layer.
In example, the user from PC-A was using telnet to transmit information to PC-B, so the destination port number is 23. The transport layer examines this port number and realizes that the encapsulated data needs to be forwarded to a telnet application. If PC-B doesn’t support telnet, the transport layer drops the segment. If it does, the transport layer strips off the segment information and passes the encapsulated data to the telnet application. If this is a new connection, a new telnet process is started up by the operating system.
