Ethernet Standards and Protocols Explained

This tutorial explains Ethernet Standards and Protocols in detail. Learn how the most common Ethernet standards (such as 10Base5, 10BaseT, 100BaseFX, 802.5-Token ring, 802.11b-Wireless, CSMA CD, etc.) are defined in computer network with their functions and purpose.

IEEE shorthand identifiers, such as 10Base5, 10Base2, 10BaseT, and 10BaseF include three pieces of information:

  • The number 10: At the front of each identifier, 10 denotes the standard data transfer speed over these media - ten megabits per second (10Mbps).
  • The word Base: Short for Baseband, this part of the identifier signifies a type of network that uses only one carrier frequency for signaling and requires all network stations to share its use.
  • The segment type or segment length: This part of the identifier can be a digit or a letter:
  • Digit - shorthand for how long (in meters) a cable segment may be before attenuation sets in. For example, a 10Base5 segment can be no more than 500 meters long.
  • Letter - identifies a specific physical type of cable. For example, the
  • T at the end of 10BaseT stands for twisted-pair.

One of the most common types of Ethernet in use today is 10BaseT. This particular implementation uses four-pair UTP wiring (Cat3 or higher, but most commonly you will see Cat5) using RJ-45 connectors. Each cable is connected from each network device to a central hub in a physical star topology. Within the hub, the signals are repeated and forwarded to all other nodes on the network because it is a logical bus topology. Older network interface cards are configured with jumpers to set addresses and interrupts.

Today's network interface cards can be managed through a diagnostic program, or automatically configure themselves through plug and play technology. There is a limit of 1024 devices on an Ethernet segment, plus you can have a maximum of 1024 network segments. A UTP cable has a maximum distance of 100 meters, which is equivalent to 328 feet.


10BaseF is an implementation of Ethernet 802.3 over fiber optic cabling. 10BaseF offers only 10 Mbps, even though the fiber optic media has the capacity for much faster data rates. One of the implementations of 10BaseF is to connect two hubs as well as connecting hubs to workstations. The best time to use 10BaseF is in the rewiring of a network from copper to fiber optic, when you need an intermediate protocol using the new wiring. 10BaseF is not often a permanent solution because the data rate is so low and the cabling so expensive in comparison to using UTP.


10Base2, also called ThinNet, is one of the two Ethernet specifications that use coaxial cable. (One of the best ways to remember that 10Base2 is ThinNet, and 2 is smaller than 10Base5, which is ThickNet.) One of the most important issues to remember in an Ethernet coax wiring scheme is the 5-4-3 rule,

5-4-3 rule
which states that you can have up to five cable segments, connected by four repeaters, with no more than three of these segments being mixing segments. In the days of coaxial cable networks, this meant that you could have up to three mixing segments of 500 or 185 meters each (for 10Base5 and 10Base2, respectively) populated with multiple computers and connected by two repeaters. You could also add two additional repeaters to extend the network with another two cable segments of 500 or 185 meters each, as long as these were link segments connected directly to the next repeater in line, with no intervening computers,

A 10Base2 network could therefore span up to 925 meters and a 10Base5 network up to 2,500 meters which states that there can only be 5 segments in a series and 4 repeaters between these 5 segments, although only 3 of the segments can be populated with devices. 10Base2 uses BNC connectors and is implemented as both a physical and logical bus topology using RG-58 cabling.

The minimum distance for cables between workstations must be at least a half-meter. Drop cables should not be used to connect a BNC connector to the network interface card (NIC) because this will cause signaling problems unless the NIC is terminated. 10Base2 ThinNet segments cannot be longer than 185 meters, although it is often exaggerated to 200 meters, and you can't put more than 30 devices on each populated segment. The entire cabling scheme, including all five segments, can't be longer than 925 meters.


10Base5 is nearly identical to 10Base2, except that it uses a different type of cabling and media connector. 10Base5 is known as ThickNet because it uses the RG-8 coaxial cable. It requires an external transceiver to attach to the network interface card on each device. The transceiver is a device that translates the workstation's digital signal to a baseband cabling format. ThinNet and UTP network interface cards have built-in transceivers. Only 10Base5 ThickNet network interfaces use external transceivers. In the 10Base5 configuration, the NIC attaches to the external transceiver using an AUI connector. The transceiver then clamps into the ThickNet cabling, which is why it is usually called a vampire tap. 10Base5 can also use BNC connectors. For 10Base5, the following rules apply: First the 5-4-3 rule applies to ThickNet just as it did to ThinNet. In addition, the minimum cable distance between each transceiver is 2.5 meters. The maximum network segment length is 500 meters, which is where 10Base5 gets the "5" in its name. The entire set of five segments cannot exceed 2,500 meters. You can have 100 devices on a 10Base5 network segment.


100BaseFX is simply Fast Ethernet over fiber. Originally, the specification was known as 100Base-X over CDDI (Copper Data Digital Interface) or FDDI (Fiber Data Digital Interface). Because the signaling is so vastly different, these two technologies were split into 100BaseFX and 100BaseTX. 100BaseFX runs over multimode fiber. There are two types of fiber in use. Multimode fiber optic cables use LEDs to transmit data and are thick enough that the light signals bounce off the walls of the fiber. The dispersion of the signal limits the length of multimode fiber. Single mode fiber optic cables use injected lasers to transmit the data along fiber optic cable with an extremely small diameter. Because the laser signal can travel straight without bouncing and dispersing, the signal can travel much farther than multimode.


100BaseT4 was the specification created to upgrade 10BaseT networks over Cat3 wiring to 100 Mbps without having to replace the wiring. Using four pairs of twisted pair wiring, two of the four pairs are configured for half-duplex transmission (data can move in only one direction at a time). The other two pairs are configured as simplex transmission, which means data moves only in one direction on a pair all the time.


100BaseTX, Fast Ethernet, transmits data at 100 Mbps. Leveraging the existing IEEE 802.3u standard rules, Fast Ethernet works nearly identically to 10BaseT, including that it has a physical star topology using a logical bus. 100BaseTX requires Cat5 UTP.

Gigabit Ethernet

The fastest form of Ethernet is currently Gigabit Ethernet, also known as 1000BaseT over Cat5 or highergrade cable, using all four pairs of the cable. It uses a physical star topology with logical bus. There is also 1000BaseF, which runs over multimode fiber optic cabling. Data transmission is full-duplex, but half-duplex is also supported.

Specify the characteristics (For example: speed, length, topology, and cable type) of the following cable standards:
  • 10BASE-T and 10BASE-FL
  • 100BASE-TX and 100BASE-FX
  • 1000BASE-T, 1000BASE-CX, 1000BASE-SX and 1000BASE-LX
  • 10 GBASE-SR, 10 GBASE-LR and 10 GBASE-ER
DesignationSupported MediaMaximum Segment Length Transfer SpeedTopology
10Base-5 Coaxial 500m 10Mbps Bus
10Base-2 ThinCoaxial (RG-58 A/U) 185m 10Mbps Bus
10Base-T Category3 or above unshielded twisted-pair (UTP) 100m 10Mbps Star,using either simple repeater hubs or Ethernet switches
1Base-5Category3 UTP, or above 100m 1Mbps Star,using simple repeater hubs
10Broad-36 Coaxial(RG-58 A/U CATV type) 3600m 10Mbps Bus(often only point-to-point)
10Base-FL Fiber-optic- two strands of multimode 62.5/125 fiber 2000m (full-duplex) 10Mbps Star(often only point-to-point)
100Base-TX Category5 UTP 100m 100Mbps Star,using either simple repeater hubs or Ethernet switches
100Base-FX Fiber-optic- two strands of multimode 62.5/125 fiber 412 meters (Half-Duplex), 2000 m (full-duplex) 100 Mbps, (200 Mb/s full-duplex mode) Star(often only point-to-point)
1000Base-SX Fiber-optic- two strands of multimode 62.5/125 fiber 260m 1Gbps Star,using buffered distributor hub (or point-to-point)
1000Base-LX Fiber-optic- two strands of multimode 62.5/125 fiber or monomode fiber 440m (multimode) 5000 m (singlemode) 1Gbps Star,using buffered distributor hub (or point-to-point)
1000Base-CXTwinax,150-Ohm-balanced, shielded, specialty cable 25m 1Gbps Star(or point-to-point)
1000Base-T Category5 100m 1Gbps Star
802.5 (token ring)

The IEEE 802.5 Token Ring standards define services for the OSI physical layer and the MAC sublayer of the data link layer. Token Ring computers are situated on a continuous network loop. A Token Ring controls access to the network by passing a token, from one computer to the next. Before they can transmit data they must wait for a free token, thus token passing does not allow two or more computers to begin transmitting at the same time.

  • Token Ring has some major advantages over Ethernet:
  • The maximum frame size for Token Ring is 4k, which is much more efficient that the small Ethernet maximum.
  • Token Ring has long-distance capability.
  • Every station in the ring is guaranteed access to the token at some point; thus, every station can transmit data.
  • Error detection and recovery techniques are also enhanced in a Token Ring environment by using a monitor function normally controlled by a server. For example, if the token is lost or corrupted, the protocol provides a mechanism to generate a new token after a specified time interval has elapsed.
Media MAC Method Signal Propagation Method Speed Topologies Maximum Connections
Twisted-pair(various types) Token passing Forwarded from device to device (or port to port on a hub) in a closed loop 4Mbps

16 Mbps

Star-using Token Ring repeater hubs
255nodes per segment
802.11b (wireless)

802.11b is a wireless Ethernet technology operating at 11MB. 802.11b devices use Direct Sequence Spread Spectrum (DSSS) radio technology operating in the 2.4GHz frequency band. An 802.11b wireless network consists of wireless NICs and access points. Access points act as wireless hubs to link multiple wireless NICs into a single subnet. Access points also have at least one fixed Ethernet port to allow the wireless network to be bridged to a traditional wired Ethernet network.. Wireless and wired devices can coexist on the same network. 802.11b devices can communicate across a maximum range of 50-300 feet from each other.

FDDI networking technologies

Fiber Distributed Data Interface, shares many of the same features as token ring, such as a token passing, and the continuous network loop configuration. But FDDI has better fault tolerance because of its use of a dual, counter-rotating ring that enables the ring to reconfigure itself in case of a link failure. FDDI also has higher transfer speeds, 100 Mbps for FDDI, compared to 4 - 16 Mbps for Token Ring. Unlike Token Ring, which uses a star topology, FDDI uses a physical ring. Each device in the ring attaches to the adjacent device using a two stranded fiber optic cable. Data travels in one direction on the outer strand and in the other direction on the inner strand. When all devices attached to the dual ring are functioning properly, data travels on only one ring. FDDI transmits data on the second ring only in the event of a link failure.

Media MAC Method Signal Propagation Method Speed Topologies Maximum Connections
Fiber-optic Token passing Forwardedfrom device to device (or port to port on a hub) in a closed loop 100 Mbps Double ringStar 500 nodes

In this section we would discuss about media protocols, media standards. Later we would explore how system gets access over media and how topology works.

  • Access method
  • CSMA / CD (Carrier Sense Multiple Access / Collision Detection)
  • CSMA / CA (Carrier Sense Multiple Access/Collision Avoidance)
  • Topology
  • Media
  • Speed

Gaining Access to the Media

Media access methods are independent of the physical and logical topologies. You will find that there are usually just a few combinations that seem to work well, however. Media access methods are simply the rules that govern how a device can submit data to the network. Each access method will have a different effect on network traffic.

Contention as a Method of Media Access

Contention, often called random access, is the media access method that acts as an open door to anyone who wants to walk in. Two types of contention methods exist for media access; they are similar, but a single difference between them changes how efficiently they operate. They are:

  • CSMA/CD (Carrier Sense Multiple Access / Collision Detection)
  • CSMA/CA (Carrier Sense Multiple Access/Collision Avoidance)

In a traditional, or hub-based, Ethernet environment, only one NIC can successfully send a frame at a time. All NICs, however, can simultaneously listen to information on the wire. Before an Ethernet NIC puts a frame on the wire, it will first sense the wire to ensure that no other frame is currently on the wire. If the cable uses copper, the NIC can detect this by examining the voltage levels on the wire. If the cable is fiber, the NIC can detect this by examining the light frequencies on the wire. The NIC must go through this sensing process, since the Ethernet medium supports

csma cd

multiple access

another NIC might already have a frame on the wire. If the NIC doesn't sense a frame on the wire, it will transmit its own frame; otherwise, if a frame is found on the wire, the NIC will wait for the completion of the transmission of the frame and then transmit its own frame.

Collision Detection

If two or more devices simultaneously sense the wire and see no frame, and each places its frame on the wire, a collision will occur. In this situation, the voltage levels on a copper wire or the light frequencies on a piece of fiber get messed up. For example, if two NICs attempt to put the same voltage on an electrical piece of wire, the voltage level will be different from that of only one device. Basically, the two original frames become unintelligible (or indecipherable). The NICs, when they place a frame on the wire, examine the status of the wire to ensure that a collision does not occur: this is the collision detection mechanism of CSMA/CD.

If the NICs see a collision for their transmitted frames, they have to resend the frames. In this instance, each NIC that was transmitting a frame when a collision occurred creates a special signal, called a jam signal on the wire. It then waits a small random time period, and senses the wire again. If no frame is currently on the wire, the NIC will then retransmit its original frame. The time period that the NIC waits is measured in microseconds, a delay that can't be detected by a human. Likewise, the time period the NICs wait is random to help ensure a collision won't occur again when these NICs retransmit their frames. The more devices you place on an Ethernet segment, the more likely you will experience collisions. If you put too many devices on the segment, too many collisions will occur, seriously affecting your throughput. Therefore, you need to monitor the number of collisions on each of your network segments. The more collisions you experience, the less throughput you will get.


WLANs use a mechanism called Carrier Sense, Multiple Access/Collision Avoidance (CSMA/CA). Unlike Ethernet, it is impossible to detect collisions in a wireless medium. In a WLAN, a device cannot simultaneously send or receive and thus cannot detect a collision: it can only do one or the other. To avoid collisions, a device will use Ready-to-Send (RTS) and Clear-to-Send (CTS) signals. When a device is ready to transmit, it first senses the airwaves for a current signal. If there is none, it generates an RTS signal, indicating that data is about to send. It then sends its data and finishes by sending a CTS signal, indicating that another wireless device can now transmit.

Ethernet (802.3) and LLC (802.2)

There are two ways that specifications become standards. One is through standardized development, and the other is through common usage of a proprietary specification, where the usage becomes so prevalent that the specification is adopted as a standard. Ethernet is the latter. The IEEE was not the first to develop Ethernet. That honor goes to the research and development efforts of three companies in the 1970s: Digital, Intel, and Xerox, which were known collectively as DIX. Later on, the IEEE based its 802.3 standard on the DIX specification. In return, DIX updated its implementation to match the small changes made by the IEEE.

Nowadays, Ethernet is used for these and several other specifications. Ethernet 802.3 is generally implemented in conjunction with 802.2. The system uses the CSMA/CD media access method, with a logical bus topology. Physically, Ethernet can be either a star or a bus. It can use copper coaxial cabling, UTP, and fiber optics. Since Ethernet uses the broadcast system of a bus topology, each node receives every data message and examines the frame header to see whether the message is meant to be received by it. If not, the frames are discarded; if so, the frames are passed on to upper layer protocols so that the receiving application can act on them.

Data Link Layer Name IEEE Standard Description
Top part Logical Link Control (LLC) 802.2

Defines how to multiplex multiple network layer protocols in the data link layer frame, which doesn't have to be Ethernet. LLC is performed in software.

Bottom part Media Access Control (MAC) 802.3

Defines how information is transmitted in an Ethernet environment and defines the framing, MAC addressing, and mechanics as to how Ethernet works. MAC is performed in hardware.


ComputerNetworkingNotes Networking Tutorials Ethernet Standards and Protocols Explained