|
Part 1: Frames, Collisions, and 10 Mbps LANs
by James Antonakos
Start ý Ethernet
Frame Format ý The Interframe Gap
ý Collision Or No Collision ý Detecting
Errors ý Random Waiting Period ý 10-Mbps
Ethernet ý 10BaseFý Errors
In Ethernet LANS ý More To Come ý
Sources and PDF
ETHERNET FRAME
FORMAT
Figure 1 shows
the format of an 802.3 frame. A separate, and practically identical,
frame format called Ethernet II has an 8-byte preamble of identical
10101010 patterns (no 10101011 SFD) and a two-byte Type field in place
of the Length field. Because the Data field of an 802.3 Ethernet frame
is limited to 1500 bytes, a value larger than 1500 in the Type/Length
field indicates an Ethernet II frame. Many companies, such as Xerox,
DEC, and Novell, use Ethernet II frames in their networks. All the
data is packaged into one or more frames for transmission over an
Ethernet LAN. Letýs look at each portion of the Ethernet frame, which
consists of the preamble (7 bytes), start frame delimiter (1 byte),
destination address (6 bytes), source address (6 bytes), length (2
bytes), data (46ý1500 bytes), and frame check sequence (4 bytes).
The preamble contains
seven bytes with the identical pattern 10101010. This alternating
sequence of ones and zeros are the first bytes transmitted. Theyýre
provided to help listening stations synchronize quickly on the new
datastream. The start frame delimiter (SFD) is the one-byte pattern
10101011. Note that the LSB is high, whereas all the bytes of the
preamble had their LSB low.
The destination
MAC address is the address of the station that is to receive the frame.
Every Ethernet device manufactured (NIC, hub or switch, router) contains
a unique 48-bit MAC address assigned by the manufacturer. An example
of a MAC address is 00-C0-F0-27-64-E2. The first bit is used to distinguish
individual addresses from group addresses, and the second bit separates
globally- and locally-administered addresses. These are followed by
the organizationally unique identifier (OUI, a 22-bit field assigned
by the IEEE) and the organizationally unique address (OUA, a 24-bit
number assigned by the manufacturer). The Windows WINIPCFG utility
can be used to determine the MAC address, called the adapter address.
A MAC address of FF-FF-FF-FF-FF-FF (all ones) is reserved for use
as a broadcast address.
The source MAC
address is the MAC address of the source station transmitting the
frame. The length field indicates the number of bytes in the data
field. And, the data area is where all the data from the upper networking
layers is carried. Upper layer networking protocols (such as TCP/IP
protocols) are encapsulated and stored in the data area.
Finally, the frame
check sequence (FCS) is a 32-bit cyclic redundancy check (CRC) value
used to determine the validity of the received Ethernet frame. The
FCS is found using a 33-bit generator polynomial defined mathematically
as:
G(x) = x32 + x26
+ x23 + x22 + x16 + x12 + x11 + x10 + x8 + x7 + x5 + x4 + x2 + x +
1
This is equivalent
to the binary value 100000100110000010001110110110111, or 104C11DB7
hexadecimal. The ones in the pattern indicate where exclusive OR gates
are used in a recirculating shift register circuit that is able to
generate/check a valid stream of data of variable length. If even
a single bit in the frame is received in error, the shift register
will not generate the proper output and the error will be discovered.
All bits in the frame, except those in the FCS, are transmitted from
LSB to MSB.
From the numbers
provided in Figure 1, the minimum and maximum sizes of an Ethernet
frame are 72 and 1526 bytes, respectively. Many individuals refer
to the smallest Ethernet frame as 64 bytes by not including the 8
bytes of Preamble and SFD. The largest frame would then be 1518 bytes.
For the purposes of the discussion here, the Preamble and SFD will
be included in the length of the frame.
PREVIOUS NEXT
Circuit Cellar provides up-to-date information for engineers. Visit
www.circuitcellar.com for
more information and additional articles.
For subscription information, call (860) 875-2199, subscribe@circuitcellar.com
or subscribe online.
ýCircuit Cellar, the Magazine for Computer Applications. Posted with
permission. |