0% found this document useful (0 votes)
45 views27 pages

Ethernet

Uploaded by

Mr Lowkey
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
45 views27 pages

Ethernet

Uploaded by

Mr Lowkey
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Ethernet

Ethernet
• In itself Ethernet cannot make a network and needs some
other protocol such as TCP/IP to allow nodes to
communicate. Unfortunately, Ethernet in its standard form
does not cope well with heavy traffic, but this is offset by
the following:
• • Ethernet networks are easy to plan and cheap to install.
• • Ethernet network components, such as network cards
and connectors, are cheap and well supported.
• • It is a well-proven technology, which is fairly robust and
reliable.
• • It is simple to add and delete computers on the network.
• • It is supported by most software and hardware systems.
IEEE standards
IEEE 802.3 frame format
Ethernet – media access control
(MAC) layer
• It contains 2 or 6 bytes for the source and destination
addresses (16 or 48 bits each),
• 4 bytes for the CRC (32 bits) and 2 bytes for the LLC length
(16 bits).
• The LLC part may be up to 1500 bytes long.
• The preamble and delay components define the start and
end of the frame. The initial preamble and start delimiter
are, in total, 8 bytes long and the delay component is a
minimum of 96 bytes long.
• A seven-byte preamble precedes the Ethernet 802.3 frame.
Each byte of the preamble has a fixed binary pattern of
10101010 and each node on the network uses it to
synchronise their clock and transmission timings.
Ethernet – media access control
(MAC) layer
• It also informs nodes that a frame is to be sent and for them
to check the destination address in the frame.
• The end of the frame there is a 96-bit delay period, which
provides the minimum delay between two frames. This slot
time delay allows for the worst-case network propagation
delay.
• The start delimiter field (SDF) is a single byte (or octet) of
10101011.
• It follows the preamble and identifies that there is a valid
frame being transmitted.
• Most Ethernet systems use a 48-bit MAC address for the
sending and receiving node. Each Ethernet node has a
unique MAC address, which is normally defined as
hexadecimal digits, such as: 4C - 31 - 22 - 10 - F1 - 32 or
4C31 : 2210: F132.
• The LLC length field defines whether the frame
contains information or it can be used to define
the number of bytes in the logical link field. The
logical link field can contain up to 1500 bytes of
information and has a minimum of 46 bytes
• If the information is greater than this upper limit
then multiple frames are sent. Also, if the field is
less than the lower limit then it is padded with
extra redundant bits.
• The 32-bit frame check sequence (or FCS) is an
error detection scheme. It is used to determine
transmission errors and is often referred to as a
cyclic redundancy check (CRC) or simply as a
checksum.
Ethernet II or IEEE 802.3

• An Ethernet II frame is similar to the IEEE


802.3 frame; it consists of eight bytes of
preamble, six bytes of destination address, six
bytes of source address, two bytes of frame
type, between 46 and 1500 bytes of data, and
four bytes of the frame check sequence field.
When the protocol is IPX/SPX the frame type
field contains the bit pattern 1000 0001 0011
0111, but when the protocol is TCP/IP the type
field contains 0000 1000 0000 0000.
IEEE 802.2 and Ethernet SNAP
• The DSAP and SSAP fields define the types of
network protocol used.
• A SAP code of 1110 0000 identifies the network
operating system layer as NetWare, whereas 0000
0110 identifies the TCP/IP protocol.
• These SAP numbers are issued by the IEEE. The
control field is, among other things, for the
sequencing of frames. In some cases, it was
difficult to modify networks to be IEEE 802
compliant. Thus, an alternative method was to
identify the network protocol, known as Ethernet
SNAP
LLC Protocol
LLC Protocol
• There are three principal types of frame:
information, supervisory and unnumbered.
• An information frame contains data, a supervisory
frame is used for acknowledgement and flow
control, and an unnumbered frame is used for
control purposes.
• The first two bits of the control field determine
which type of frame it is. If they are 0X (where X is
a don’t care) then it is an information frame, 10
specifies a supervisory frame and 11 specifies an
unnumbered frame.
LLC Protocol
• An information frame contains a send sequence number in the control
field which ranges from 0 to 127. Each information frame has a
consecutive number, N(S) (note that there is a roll-over from frame 127 to
frame 0). The destination node acknowledges that it has received the
frames by sending a supervisory frame.
• The function of the supervisory frame is specified by the 2-bit S field. This
can either be set to receiver ready (RR), receiver not ready (RNR) or reject
(REJ).
• If an RNR function is set then the destination node acknowledges that all
frames up to the number stored in the receive sequence number N(R) field
were received correctly.
• An RNR function also acknowledges the frames up to the number N(R), but
informs the source node that the destination node wishes to stop
communicating.
• The REJ function specifies that frame N(R) has been rejected and all other
frames up to N(R) are acknowledged.
OSI and the IEEE 802.3 standard

The IEEE 802.3 standard splits into three


sublayers:
• MAC (media access control).
• Physical signalling (PLS).
• Physical media attachment (PMA).
MAC
• The interface between PLS and PMA is called
the attachment unit interface (AUI)
• The interface between PMA and the
transmission media is called the media
dependent interface(MDI).
• This grouping into modules allows Ethernet to
be very flexible and to support a number of bit
rates, signalling methods and media types.
MAC
• When sending frames – receive frames from LLC; control whether the
data fills the LLC data field, if not add redundant bits; make the number of
bytes an integer, and calculate the FCS; add the preamble, SFD and address
fields to the frame; send the frame to the PLS in a serial bit stream.
• When receiving frames – receive one frame at a time from the PLS in a
serial bit stream; check whether the destination address is the same as the
local node; ensure the frame contains an integer number of bytes and the
FCS is correct; remove the preamble, SFD, address fields, FCS and remove
redundant bits from the LLC data field; send the data to the LLC.
• Avoid collisions when transmitting frames and keep the right distance
between frames by not sending when another node is sending; when the
medium is free, wait a specified period before starting to transmit.
• Handle any collision that appears by sending a jam signal; generate a
random number and back off from sending during that random time.
Ethernet types

The six main types of standard Ethernet are:


• Standard, or thick-wire, Ethernet (10BASE5).
• Thinnet, or thin-wire Ethernet, or Cheapernet
(10BASE2).
• Twisted-pair Ethernet (10BASE-T).
• Optical fibre Ethernet (10BASE-FL).
• Fast Ethernet (100BASE-TX and 100VG-Any LAN).
• Gigabit Ethernet (1000BASE-SX, 1000BASE-T,
1000BASE-LX and 1000BASE-CX).
100Mbps Ethernet

• New standards relating to 100 Mbps Ethernet are now


becoming popular:
• • 100BASE-TX (twisted-pair) – which uses 100 Mbps over
two pairs of Cat-5 UTP cable or two pairs of Type 1 STP
cable.
• • 100BASE-T4 (twisted-pair) – which is the physical layer
standard for 100 Mbps over Cat-3, Cat-4 or Cat-5 UTP.
• • 100VG-AnyLAN (twisted-pair) – which uses 100 Mbps over
two pairs of Cat-5 UTP cable or two pairs of Type 1 STP cable
• • 100BASE-FX (fiber-optic cable) – which is the physical layer
standard for 100 Mbps over fiber-optic cables.

You might also like