DUAL's packet format is designed to be minimal, flexible, and to
meet the needs of the both underlying MAC and Physical layers and the
Network layers above DUAL.
In order to minimise header sizes, we have removed the reliable virtual circuit capabilities found in AX.25. This dramatically decreases the size of DUAL's headers. At the same time we enabled DUAL to tailor its Link layer addresses so they were more useful to the Network layers above DUAL. This again helps to reduce the size of DUAL's headers.
Note in particular that there is no field to describe the size of the Data field. We believe this is unnecessary. Both the MAC layers and higher layers present DUAL with a packet of a fixed number of octets. Therefore DUAL can determine where the Checksum field is located, and hence the size of the Data field.
The packet format is shown in figure 5:
Figure : DUAL's Packet Format
DUAL encapsulates some higher layer data in a packet which contains the link addresses of both the source and destination, a higher layer protocol field, and a checksum.
The protocol field is divided into a Protocol Id and an Address Type,
allowing higher layer protocols, and
packet formats
per protocol.
The rationale behind permitting eight types of link addresses for each higher level protocol is to allow the link addresses to be tailored to the characteristics of the physical link (speed, quality) and to the addressing needs of the link (if there are only 100 stations on the link, a link address of 1-2 octets is sufficient). Section 8 gives as an example the link addressing scheme used when IP packets are transported over DUAL.
Figure : DUAL Field Dependencies
The size and format/content of the Source, Destination and Checksum fields are determined entirely by the Protocol-id/Address-type field. This allows the size and format of these fields to be tailored to suit the protocol being transmitted over the link.