O GoConqr está de visual novo! Conte para nós o que você achou.

Notas

Physical Layer Cabling: Twisted-Pair Notes

Justin Bullock Network and Security: Foundations Inglês 1 página

Conteúdo das notas (1)

Página 1

Physical Layer

Cabling: Twisted Pair

Introduction

Examines the

twisted-pair media used to link computers together to form a local area

network (LAN). This is all a part of the Physical Layer of the OSI

model.

The main focus is on

unshielded twisted-pair (UTP) though an overview of shielded

twisted-pair (STP) is presented.

Chapter describes the six

subsystem of a structured cabling system.

Structured Cabling

The first

standard describing structured cabling system was TIA/EIA 568-A in

1995.

EIA is the

Electronic industries Alliance, a trade organization that lobbies for

the interests of manufactures of electronics-related equipment.

TIA stands for the

Telecommunications Industry Association, which is a trade organization

that represents the interests of the telecommunications industry.

Addendum 5 in 1999 was

the most important change to the standard because it defied the

transmission performance specifications for 4-pair 100-ohm category 5e

twisted-pair cabling.

They adopted the new

category 6 cable specifications in 2002, which is the recommended

cabling for use in today’s networks though CAT7 cabling may soon

become the recommended standard.

The EIA/TIA 568-A defines

the minimum requirements for the internal telecommunications wiring in

buildings and between structures in a campus network.

A campus network consists

of interconnected LANs within a limited geographic area such as a

college campus, military base, or group of commercial buildings.

In 2000 a new standard was

published called EIA/TIA 568-B with three parts:

EIA/TIA-568-B.1:

Commercial Cabling Standard, Master Document.

EIA/TIA-568-B.2:

Twisted-pair Media

EIA/TIA-568-B.3: Optical

Fiber Cabling Standard

With the EIA/TIA568-B

standard define the six subsystems of a structured cabling system:

Building

Entrance: The point where the external cabling and wireless services

interconnect with the internal building cabling in the equipment room.

Used for both public and private access. Also called the Entrance

Facilities (EF). Both public and private network cables enter the

building at this point, and typically each has separate facilities for

the different access providers.

Equipment Room (ER): A

room set aside for complex electronic equipment such as the network

servers and telephone equipment.

Telecommunications

Closet: The location of the cabling termination points that includes

the mechanical terminations and the distribution frames. The

connection of the horizontal cabling to the backbone wiring is made at

this point. This is also called the telecommunications room (TR) or

telecommunications enclose (TE).

Backbone Cabling: Cabling

that interconnects telecommunication closets, equipment rooms, and

cabling entrances in the same building and between buildings.

Horizontal Cabling:

Cabling that extends out from the telecommunications closet into the

LAN work area. Typically, the horizontal wiring is structured in a

star configuration running to each area telecommunications outlet

(TCO). This is the wall plate where the fiber or twisted-pair cable

terminates in the room. In some cases, the TCO terminates telephone,

fiber and video in addition to data into the same wall plate.

Work Area: The location

of the computers and printers, patch cables, jacks, computer adapter

cables, and fiber jumpers.

One room can

serve as the entrance facility, equipment room, and the

telecommunications closet.

Structuring

Main

Cross-Connect (MC): Usually connects two or more buildings and is

typically the central telecommunications connection point for a campus

or building. It is also called the main distribution frame (MDF) or

main equipment room. The MC connects to telco, an ISP, and so on.

Another term for MC is the campus distributor (CD).

Intermediate

Cross-Connect (IC): Also called the building distributor (BD), this is

the building’s connection point to the campus backbone. The IC links

the MC to the horizontal cross-connect (HC).

The Telecommunications

Cabling System Architecture

A cross-connect

is a space where you are going to take one or multiple cables and connect

them to one or more cables or equipment.

Typically connectors between

the MC and IC are single-mode and multi-mode fibers and possibly coax for

cable tv, although most installations are migrating to fiber.

Horizontal Cross-connect

(HC): The connection between the building distributors and the horizontal

cabling to the work area or workstation outlet: another term used for the

HC is the floor distributors (FD).

Workstation or Work Area

Outlets (WO): Also called the TO (telecommunications outlet). It’s used

to connect devices to the cable plant. The cable type typically used is

CAT3/5/5e/6/6A or various coaxial cables. Devices typically connected to

these outlets are PCs, printers, servers, phones, televisions, and

wireless access points.

The Campus Network

Hierarchical Topology

Horizontal

Cabling

Permanent network cabling within a building is

considered to be horizontal cabling, defined as the cabling that extends

out from the telecommunications closet into the LAN work area.

The Following components

are typically found in the telecommunications closet:

Backbone

Cabling interconnecting this closet with other closets

Switch or Hub

Patch Panels

Patch Cables

Cabling to the LAN

(horizontal cabling)

Wall Plate

Patch cable connecting the

computer to the wall plate

Unshielded

Twisted-Pair Cable

Category Cables

CAT6/5e/5

twisted-pair contains four twisted wire pairs for a total of eight

wires.

Twisted-pair

cables don’t connect to ground, the signals on the wires are set up

for a high (+) and low (-) signal line. Both signals are relative to a

virtual ground.

Balance Mode -- the balance of the two wire pairs helps

maintain the required level of performance in terms of crosstalk and

noise rejection.

Categories

Category 3

(CAT3)

Telephone

installations Class C

Up to 16 Mbps

Category 5 (CAT5)

Computer

Networks Class D

Up to 100 MHz / 100 Mbps

100-m length

Enhanced Category 5

(CAT5e)

Computer

Networks

100 MHz / 1000 Mbps

applications with improved noise performance in a full duplex mode.

Category 6 (CAT6)

Higher-speed

Computer

Up to 250 MHz networks

Class E / 1000 Mbps CAT6 supports 10 Gbps but at distances fewer than

100 meters.

Category 6a (CAT6a)

Increased

bandwidth

Up to 500 MHz networks

Class Ea / 10 Gbps

Category 7 (CAT7)

International

Organization for Standardization (ISO) standard, not an EIA/TIA

standard

Up to 600 MHz speed

computer networks Class F / 10 Gbps

Category 7a (CAT7a)

ISO standard,

not an EIA/TIA standard

Up to 1000 MHz speed

computer networks Class FA / 10 Gbps

CAT1 and CAT2 are not in

the EIA/TIA 568B standard, CAT3 is bing replaced with CAT5e or better

and CAT4 because it was removed for being outdated.

FastEthernet is used to

describe the 100 Mbps data rate for FastEthernet networks.

Network Congestion or

Bottlenecking is a slowdown on network data traffic movement.

Can be caused

by file size and limited bandwidth.

Full Duplex means that the

computer system can transmit and receive at the same time.

Terminating CAT6/5e/5 UTP

Cables

Wiring

Guidelines

Within the

EIA/TIA-568-B standard are the wiring guidelines T568A and T568B, they

specify the color of wire that connects to which pin on the connector,

called the color map.

There

is no difference between the two wiring guidelines, just different

manufacturer standards to be used, most use T568A.

Computer

Communication

Speeds of 10 Mbps / 100

Mbps use two of the twisted-pairs.

For communication to work

in a LAN, transmit and receive pairs must be properly aligned.

The transmit (TX) (+) and

(-) signals must connect to the receive (RX) (+) and (-).

In

a LAN, proper alignment is managed by a switch or hub.

When operating at 1 Gbps

and 10 Gbps speeds the signal names for a UTP cable are different and

all four of the wire pairs are used.

The

data is bidirectional, which means the same wire pairs are being used

for both transmitting and receiving data.

Straight-through

and Crossover Patch Cables

Straight-through

has all wires connected to the same corresponding pin on each end.

Crossover cable has the

transmit pair of device A connects to the receive pair of device B, and

transmit pair of device B connects to the receive pair of device A. If a

1 Gbps or 10 Gbps cable is set up all pairs are crossed.

Terminating a CAT6

Horizontal Link Cable

Inspect cable

for any damage.

At the work area outlet

end, add one foot extra and cut the wire. Coil the extra cable and

insert in the receptacle box. Cannot splice a CAT6 cable. At the

distribution end, route the cable and create a slack loop. Allow for 5

meters in the TC and 5 meters in the work area.

Place a bend limiting

strain relief boot on the cable. Strip 3 inches of cable jacket,

Remove the jacket from the

UTP cable and expose the four wire pairs.

Cut the plastic pull line

and the string. Place a lacing fixture on the cable and sort the wires

in either T568A or T568B.

Place the wires in the

slots of the lacing tool.

Align an RJ-45 (8P8C) jack

with the lacing fixture.

Insert the RJ-45 modular

jack into the AMP SL tool and then insert the RJ-45 jack into the AMP SL

tool.

Connect

the bend-limiting strain relief boot to the RJ-45 jack.

Assembling the

Straight-through CAT5e/5 Patch Cable

Inspect the

cable for damage.

Measure cable to length,

add about 6 inches and cut the wire. You can’t splice CAT5e/5

twisted-pair cable.

Strip about ¾ of an inch of

the cable jacket from the end of the cable using a cable stripper.

Sort the wire pairs so that

they fit into the connector and orient the wire in the proper order for

either T568A or T568B.

Clip the wires so that they

are even and insert the wires onto the RJ-45 modular plug.

Push the wires into the

connector until the ends of each wire can be seen through the clear end

of the connector.

Use a crimping tool to

crimp the wires onto the RJ-45 plug.

Repeat these steps for the

other end of the twisted-pair cable.

Cable Testing and

Certification

Link: Point from

one cable termination to another.

Full Channel: Consists of

all the link elements from the wall plate to the hub or switch.

Attenuation (Insertion

Loss): The amount of loss in the signal as it propagates down a wire or

fiber strand.

Near-end crosstalk (NEXT):

A measure of the level of crosstalk or signal coupling within the cable,

with a high NEXT (dB) value being desirable.

EIA/TIA 568B

Channel Specifications

Attenuation

(insertion loss): defines the amount of loss in signal strength as it

propagates down the wire. Test reports a margin, margin for attenuation

is defined as the difference between the measured value and the limit

for the test. If the margin shows a negative value the test failed. The

limit for CAT6 is 21.3dB, CAT6A is 20.9dB, CAT7 is 20.8 dB, and CAT7a is

20.3 dB. Bend Radius can also introduce attenuation loss.

NEXT:

When current travels in a wire, and electromagnetic field is created.

This can induce a voltage in adjacent wires resulting in crosstalk.

Crosstalk is what you occasionally hear on the telephone when you can

faintly hear another communication. NEXT is a measure of the level of

crosstalk, or signal coupling within the cable. More likely to happen

near the ends of the cable and is more problematic at higher data

speeds. A high NEXT (db) value is desirable.

Power Sum NEXT (PSNEXT):

This measures the total crosstalk of all cable pairs. This test ensures

that the cable can carry data traffic on all four pairs at the same time

with minimal interference. A higher value is desirable as it indicates

better cable performance.

Equal Level FEXT (ELFEXT):

Measures for the far end of the cable, and doesn’t depend on the length

of the cable. ELFEXT is obtained by subtracting the attenuation value

from the FEXT. A larger value is desirable.

PSELFEXT: Power sum ELFEXT

uses all four wire pairs to obtain a combined ELFEXT. A higher PSELFEXT

value indicates better performance.

ACR: Compares the signal

level from a transmitter at the far end to the crosstalk measured at the

near end. A larger ACR indicates greater data capacity and bandwidth.

PSACR: Power sum ACR uses

all four wire pairs to obtain the measure of the attenuation-crosstalk

ratio. Is the difference between PSNEXT and attenuation. Higher value

indicates better cable performance.

Return Loss: Provides a

measure of the ratio of power transmitted into a cable to the amount of

power returned or reflected. Reflection can be due to impedance changes

in the cable link which can contribute to cable loss. A low value is

desirable.

Propagation Delay: Measure

the amount of time it takes for a signal to go from end of the cable to

the other. Delays are affected by nominal velocity of propagation (NVP)

of the cable. NVP is some percentage of the velocity of light and is

dependent on the type cable being tested. Typical delay value is 5.7

nsec per meter. EIA/TIA allows for 548 nsec for 100 meter run.

Delay

Skew: Measures the difference in arrival time between the fastest and

slowest signal in a UTP wire pair. Data on the wire pair needs to arrive

at the other end at the same time. Different lengths of wires will cause

one to take longer than the other which can cause data and data packet

loss. Shortest length will typically have the least delay skew.

10 Gigabit

Ethernet over Copper

Overview

Ethernet over

copper is available for 10 Mbps (Ethernet), 100 Mbps (FastEthernet),

1000 Mbps (gigabit Ethernet), and 10 Gbps (ten gigabit Ethernet).

Three improvements required for higher

data bit over copper:

Improve the

cable so it can carry greater bandwidth

Improve the electronics

used to transmit and receive (recover) the data.

Utilize improvements in

both the cable and electronics to facilitate greater bandwidths and

distance.

IEEE 802.3an-2006

10GBASE-T is the standard developed to support running 10 Gbps data

over twisted-pair cabling. This requires the bandwidth to be increased

from 250MHz to 500MHz.

Alien Crosstalk

Alien Crosstalk

(AXT) is unwanted signal coupling from one permanent link to another.

This is an important issue at higher data rates, and the coupling of a

signal from one 4-pair cable to another 4-pair cable.

Alien Crosstalk is

considered to be the main electrical limiting parameter for 10G

Ethernet. It causes disturbances in the neighboring cable.

Cable manufactures are

starting to offer CAT6 and higher grades of twisted-pair cable with

foil over each of the four wire-pairs. The designation of this type is

F/UTP. Advantages are:

A shielded

cable offers better security because there is less chance that the

data will radiate outside the cable.

The foil shield helps

improve noise immunity from EMI, radio frequency interference (RFI),

and (most importantly) AXT.

Transmission over

twisted-pair cabling relies on the signals being “balanced” over the

wire pairs. The balance helps minimize unwanted leakage of signal.

TCL (Transverse Conversion

Loss) measurement is obtained by applying a common-mode signal to the

input and measuring the differential signal level on the output, also

sometimes called LCL (Longitudinal Conversion Loss).

TCTL (Transverse

Conversion Transfer Loss) is the loss from a balance signal at the

near-end to the unbalanced signal at the far end.

ELTCTL (Equal Level

Transverse Conversion Transfer Loss) is the difference between the TCTL

and the differential mode insertion loss of the pair being measured.

These tests have been

developed to help ensure compatibility with data transmission and

reception that requires the use of all four wire-pairs.

PSANEXT

(Power-Sum Alien Near-End Cross-Talk)

PSAACRF (Power-Sum Alien

Attenuation Cross-Talk Ratio Far-End)

Signal Transmission

The 10GBASE-T

system requires the use of all four wire pairs. It splits the 10 Gbps

of data into four 2.5 Gbps data channels. Requires the use of signal

conditioners and digital signal processing (DSP) circuits for both

transmission and reception. The data transmission for 10 gigabit uses a

multilevel encoding technique that reduces the required bandwidth to

transport the data.

10GBASE-T data

transmission also requires the use of DSP Compensation Techniques.

The DSP

circuitry provides many functions:

Signal

Conditioning

Echo Cancellation

In 10GBASE-T, the transmit

and receive signals are sharing the same wire pair. This is called full

duplex transmission and requires the use of a device called a hybrid

echo cancellation circuit which removes the transmitted signal from the

receive signal.

Cabling

systems that will support 10GBASE-T are CAT6 with the foil screen,

CAT6a, CAT7, and CAT7a.

Troubleshooting Cabling

Systems

Installation

If you obtain

bad Power-Sum or NEXT or FEXT, there might be a problem with the

installation.

A NEXT failure is most

likely due to a problem at the terminations. Most commonly the

installer has allowed too much untwisted cable at the termination point

(less than 3/8 inch). Inspect the terminations.

Cable Stretching

Avoid

stretching of the UTP cable because it changes the electrical

characteristics of the cable, increasing the attenuation and crosstalk.

Cable Failing to Meet

Manufacturer Specifications

Manufacturers

do experience problems with the cable failing to meet specifications

due to bad production run. Test failures with no apparent reason for

the failure could indicate that the problem is with the cable.

Informações

Descrição
Network and Security: Foundations (Physical Layer Cabling: Twisted-Pair) Notas sobre Physical Layer Cabling: Twisted-Pair Notes, criado por Justin Bullock em 11-10-2016.
Criado por
Justin Bullock
Visualizações
3
Avaliação

Fale Conosco

Sobre o que é sua mensagem?

O que você acha do novo visual?

Você gostou?