Time to Live: This eight-bit field is strictly used by the routers to prevent a datagram from a faulty transmission sequence to endlessly circulate around an internet. Originally the unit of measure was seconds because it was believed that it would take a router one or more seconds to process a datagram from its queue. Once the datagram was processed, the router would decrement this field by the amount of time that occurred. However, in practice modern routers are much faster then early routers and usually process the datagram within a second but only decrement the field by one (the minimum amount). Therefore, the field has come to be treated as a hop counter. A hop being an instance of a datagram being processed by a router. The originating host sets the Time to Live field and each router decreases it by one. If a router decrements the count to zero it will discard the datagram and inform the originating host that the datagram failed to reach its destination.

Protocol: The eight-bit protocol field informs the upper layer protocol that the received datagram is for its use. Usually the upper layer protocol is TCP or UDP but there are other protocols as well that could be sending and receiving data. The protocol field provides this distinction.

Header Checksum: The complete IP header is checked for integrity with the aid of the 16-bit header checksum. The originating host applies the checksum and all routers check the header for integrity and regenerate a new checksum when the datagram is resent. A new checksum is required since the Time to Live field would have been changed by the router. Finally, the checksum is again reconfirmed by the receiving host.

Source/Destination Address: The 32-bit source and destination addresses are included in the header. These are the IP addresses and not MAC addresses.

IP Options: There may be no options in which case this field is null or there can be options usually intended for router use only. The option fields must be at least 32-bits in length and must be padded to that amount if shorter.

ARP

As mentioned before, the IP routes datagrams between source and destination addresses in the form of packets over a data link layer. The data link does not understand datagrams nor does it understand IP addresses. It does know, however, its own MAC address and knows how to communicate to other MAC addresses when told to do so. Somehow we need to inform each host what IP address its MAC address or physical address has been assigned and we need to inform the same host all the other physical address assignments on the local network in order to have communication.

Usually the host IP address-physical address assignment is stored in non-volatile memory or in a file. Using a 32-bit DIP switch for assignment is not practical. Sometimes a serial port on the device is used for programming the IP address but once programmed all other hosts on the local network must still need to learn the assignment.

The Address Resolution Protocol (ARP) is used for learning physical address assignments. ARP has its own structure and does not use that of IP. ARP directly communicates to the data link layer and, therefore, must be aware of the various types of network adapters that are available.

When a host needs to send a datagram to another host on a local network, it first checks its ARP table to determine the physical address for that IP destination address. If one is found, the datagram transmission proceeds. If none is found, an ARP request is made. An ARP request consists of a broadcast message to all hosts on the local network. Within the ARP request is the originator's IP and physical addresses as well as the requested IP address. Since it is a broadcast message, all hosts have the opportunity to learn the IP address and physical address pairing of the requestor which can be appended to that host's ARP table. Only the host with the requested IP address responds to the ARP request by providing its IP address and physical address pairing. This message is sent as a unicast message back to the requestor. Once the physical address is known by the requestor, the datagram can be sent.

SUMMARY

The IP is responsible for the end-to-end delivery of datagrams over an internet. It also provides host and network addressing and the means for fragmenting datagrams into manageable packets. IP is a routable protocol and much of its complexity is due to its ability to route packets directly within a local network or indirectly through routers. Routers are not ideal for a control network since they reduce determinism and increase data latency. Still to accept TCP as a transport layer for an Ethernet control network requires acceptance of IP as well. By understanding the limitations of IP, a control network can still be designed using the TCP/IP family of protocols. This is especially true if the control network is restricted to that of a private or local network.

References

Illustrated TCP/IP, Matthew Naugle, 1998, Wiley Computer Publishing

Practical Networking With Ethernet, Charles E. Spurgeon, 2000, International Thomsom Computer Press

International Standard ISO/IEC 8802-3 ANSI/IEEE Std 802.3, 2000, The Institute of Electrical and Electronic Engineers, Inc.

TCP/IP Clearly Explain, Pete Loshin, 1997, Academic Press

TCP/IP Illustrated, Volume 1,The Protocols, W. Richard Stevens, 1994, Addison-Wesley Publishing Company