Lesson #2: Exploring Data Transmission Media
Data transmission is the process of conveying data between two points
by way of a communication medium. A wide variety of
media are available, but they fall into two classes: bounded and
Bounded media confine the data to specific
physical pathways. Common examples of bounded media are wire and optical
fiber cables. Cable TV uses bounded media.
Unbounded media transmit the data-carrying
signal through space, independent of a cable. Broadcast radio and
television are examples of unbounded media.
By far the most common media employed for data transmission are defined
as bounded -- the data signal is confined in a specific transmission pathway.
When practical, cable represents a low-cost and reliable means of
transmitting data between computing devices.
Practicality is a relative thing. Certainly cables are likely to be
the logical choice within a building or even a building complex. It may
not be possible, however, to run a cable between two buildings on different
sides of a public road, and it is certainly a major undertaking when the
buildings are located on different continents. Such conditions may call
for use of unbounded media.
You should be alert to several characteristics when examining cables:
- Resistance to electromagnetic interference (EMI).
- Bandwidth, the range of frequencies that
the cable can accommodate. LANs generally carry data rates of
1 to 100 megabits per second and require moderately high bandwidth.
- Attenuation characteristics. Attenuation
describes how cables reduce the strength of a signal with distance.
Resistance is one factor that contributes to signal attenuation.
EMI (ElectroMagnetic Interference) can be
a major headache for LAN technicians. Many electrical devices generate
magnetic fields that produce unwanted electrical currents in data cables.
The noise that results from these currents can degrade
data signals, sometimes stopping communication altogether due to excessive error
rates. Electrical motors and fluorescent lights are common sources of EMI,
and it can be a genuine challenge to cable a network in environments such as
factories that contain many electrical devices.
Cables fall into two broad categories -- electrical
conductors and fiber optic -- with
various types of cables available in each category. Prior to an examination
of fiber optic cables, this section examines two types of
electrical cables: coaxial and twisted pair.
Electrical cable types are frequently referred to as
"copper" because that metal is the most frequently
used conductor. You may hear fiber optic cables
called simply "fiber" or "glass".
This type of
cable is called coaxial (or coax for short)
because two conductors share a COmmon AXis. A typical
coaxial cable has the following components:
This conductor usually consists of a fairly heavy, solid yet flexible
wire; stranded wires can also be used. Solid conductors are
preferred for permanent wiring, but stranded conductors make the
cable more flexible and easier to connect to equipment.
Also called a dielectric layer, this layer provides
electrical insulation and keeps the inner and outer conductors in
precise coaxial relationship.
Outer conductor or shield.
This layer shields the inner conductor from outside electrical
interference. The shield can consist of braided wires, metal foil,
or a combination of both. Because of this shield, coax is highly
resistant to electromagnetic interference (EMI).
Jacket or sheath.
A durable PVC plastic or Teflon jacket coats the cable to prevent
Coax has many desirable characteristics. It
is highly resistant to EMI and can support high bandwidths. Some types of coax have heavy shields and center conductors
to enhance these characteristics and to extend the distances that signals can
be transmitted reliably.
A wide variety of coax cable is available. You must use cable
that exactly matches the requirements of a particular type of network. Coax
cables vary in a measurement known as the impedance
(measured in a unit called the ohm), which is an indication of the cable's
resistance to current flow. The specifications of a given cabling standard
indicate the required impedance of the cable.
Here are some common examples of coaxial cables used
in LANs, along with their impedances, and the LAN standards with which they are
- RG-8 and RG-11 are
50 ohm coax cables required for thickwire Ethernet. (10Base5
- RG-58 is a smaller 50 ohm coax cable
required for use with thinwire Ethernet. (10Base2 - ThinNet)
- RG-59 is a 75 ohm coax cable most
familiar when used to wire cable TV (CATV) and is also used
to cable broadband Ethernet (10Broad36).
- RG-62 is a 93 ohm cable used for
ARCnet. It is also commonly employed to wire terminals in an
IBM SNA (minicomputers & mainframes) network.
Some advantages of coaxial cable are as follows:
- Highly insensitive to EMI
- Supports high bandwidths
- Heavier types of coax are sturdy and can withstand
- Represents a mature technology that is well understood
and consistently applied among vendors
Coax also has some disadvantages including the following:
- Although fairly insensitive to RF, coax remains vulnerable
to EMI in harsh conditions such as factories.
- Coax is among the most expensive types of wire cables.
- Coax can be bulky.
shows how two conductors are twisted together to form the cable type known as
twisted pair (TP/UTP). Cables can be
constructed of multiple twisted pairs of cables contained in a common jacket.
The twists in the conductor pairs are an important part of the
electrical characteristics of UTP cable. Twists reduce
the cable's sensitivity to outside EMI and the degree to which
the cables radiate radio frequency signals (RF). Remember
that the frequencies at which LANs operate fall into the range of radio signals.
If UTP cable is left insufficiently twisted when terminated (end connectors
put on), the cable can function as an antenna and radiate significant amounts of
radio signals (NEXT) that can interfere with local broadcast reception equipment.
Twisted pair cable
used in early networks was most frequently surrounded by a braided shield that served to reduce both EMI sensitivity and radio emissions.
An example of this shielded twisted pair (STP)
cable is IBM Type 1, Type 6, and Type 9 cable used in Token
Ring installations. In the past, shielded twisted pair
cable (STP) was required for all high-performance networks such as IBM Token Ring.
STP cable, however, is expensive and bulky, and manufacturers of network
equipment have devoted extensive research to enabling high-speed networks to work
with unshielded twisted pair (UTP).
UTP is the cost leader among network cables.
The 10Base-T, 100Base-TX, and Gigabit Ethernet standards define Ethernet
configurations that utilizes UTP. Work by IBM and other vendors have developed network
equipment that can use UTP even for the higher speed 16 megabit per second Token
Ring. In most cases, UTP cable is implemented using telephone-type modular
connectors such as the RJ-11 (2 pair) and RJ-45 (4 pair) connectors. Telephone type modular
connectors are inexpensive and easy to install, serving to further reduce the
cost of UTP cabling systems.
UTP looks much like the cable used to wire voice grade telephones. In newer telephone installations,
it may indeed be possible to use the wiring installed for the voice
grade telephone system as data grade cable in a network.
UTP cable comes in a variety of grades, ranging from
CAT3 (low quality) to CAT6 (high quality). When investigating
the use of UTP cabling, be sure to determine the cable quality required for your
When utilizing UTP cable, it is necessary to ensure that all components in
the data network are data grade. Voice grade components used in voice telephone
systems are not of sufficiently high quality for the bandwidth needed for
Shielded twisted pair cable (STP) types are the
standard cables specified for IBM's Token Ring networks and for Apple's LocalTalk.
Unshielded twisted pair cables (UTP) can be
utilized for some configurations of Token Ring, Ethernet, and ARCnet networks.
Here are some advantages of twisted pair wiring:
- Telephone cable standards are mature and well established.
Materials are plentiful, and a wide variety of cable installers
are familiar with the installation requirements.
- It may be possible to use in-place telephone wiring if
it is of sufficiently high quality.
- UTP represents the lowest cost cabling. The cost
for STP is higher and is comparable to the cost of coaxial cable.
Some disadvantages of twisted pair are as follows:
- STP can be expensive and difficult to work with.
- Compared to fiber optic cable, all UTP cable is
more sensitive to EMI.
- UTP especially may be unsuitable for use in
- Cable segment lengths are also more limited with UTP.
- UTP cables are regraded as being less suitable for
high-speed transmissions than coax or fiber optic. Technology
advances, however, are pushing upward the data rates possible with UTP.
Fiber optic cables utilize light waves to transmit data
through a thin glass or plastic fiber. The structure of
a typical fiber optic cable is shown in the graphic. The parts of the fiber
optic cable are as follows:
light conductor is a very fine fiber core.
Glass is the most common material, allowing
signals to be transmitted for several kilometers without being refreshed.
Plastic is used in some circumstances, but plastic cables allow only
short cable runs.
The cladding is a glass
layer that surrounds the optical fiber core. The optical characteristics
of the cladding reflect light back to the core, ensuring that very little of
the light signal is lost.
A sheath or jacket protects
the cable from damage. A single sheath can be used to bundle multiple
core/cladding fibers into a multi-fiber cable.
The light signals on fiber optic cables are generated either by
light emitting diodes (LEDs) or by injection
laser diodes (ILDs), which are similar to LEDs
but produce laser light. The purity of laser light is desirable,
increasing both data rates and transmission distance. Light Signals are
received by photodiodes, solid state devices that
detect variations in light intensity.
The interface devices required to operate with fiber optic
cable are more expensive than those required for copper cable.
The higher cost is the result of several factors, including cost of the
components and tighter design characteristics because fiber optic cables generally
are operated at high data rates. The cost of fiber optic
cable installation, however, is trending downward.
Fiber optic cables have many desirable characteristics. Because the
fibers are small in diameter, a cable of a given size can contain more
fibers than copper wire pairs. Because fiber optic cables use light
pulses instead of electrical signals, they offer very high bandwidth. Bandwiths of 100 megabits (million bits per
second) are commonplace, and bandwidths in the Gigabit (billion bits per second)
and 10 Gigibit (10 billion bits per second) range are available.
Because the signal in a fiber optic cable consists of light pulses, the
signal cannot be affected by electromagnetic interference.
Nor can the cables radiate radio frequency noise.
Optical fibers are, therefore, suitable for use in the noisiest
and most sensitive environments. Because these cables radiate no
electromagnetic energy, it is impossible to intercept the data signal with electronic
eavesdropping equipment. Fiber optic transmissions are extremely secure.
Installation of fiber optic cable requires greater skill
than is necessary to install most copper cables. Cables
must not be bent too sharply, and connectors must be installed by skilled technicians
using special tools. However, new connector technologies have simplified
installation and reduced cost.
Here are some advantages of fiber optic cable:
- Very high bandwidth.
- Immunity to EMI; fiber optic cables can be used
in environments that make copper wire cables unusable.
- No radio frequency (RF) emissions; signals on fiber optic
cables cannot interfere with nearby electronic devices and cannot be
detected by conventional electronic eavesdropping techniques.
Summary of Cable Characteristics|