Network Cabling and Data Transmission

Basic Data Transmission Concepts

Networking media provides the physical foundation for data transmission. This module focuses on wired networks.

Key Concepts

• Frequency: The number of times an electrical signal changes states in a second, measured in MHz or GHz. It affects how quickly data can be transferred over a cable. Wireless signals must be contained within a specific range of the electromagnetic spectrum, wired signals don’t have to be so tightly contained because the cable itself mostly limits the dispersion of the signal.

  Example: A traffic engineer increasing the number of green light cycles per minute to release more cars onto a freeway.

• Bandwidth:

  • Frequency context: The possible range of frequencies up to a maximum, measured in MHz or GHz.

  • Data transfer context: The amount of data that could theoretically be transmitted during a given period, measured in Mbps or Gbps. It considers frequency, distance, and SNR (signal-to-noise ratio).

    Example: The number of lanes on a freeway. Adding more lanes increases bandwidth, allowing more cars per green light.

• Throughput: The number of data bits actually received across a connection each second, measured in Mbps or Gbps. It considers delays, noise, and errors.

  Example: The number of people who arrive at their destination per minute via a freeway, accounting for weather, traffic, and other real-world conditions.

$Bandwidth <br />\neq Throughput$

At one time, frequency, throughput, and bandwidth were directly related mathematically—a cable with a maximum frequency of $100 MHz$ could transmit a maximum of $100 Mbps: 100 MHz correlated with 100 Mbps$. Today, however, additional layers of technology complicate the mathematical relationship between these measurements. Cables with a maximum frequency of $250 MHz$ rated for a maximum throughput of $1000 Mbps$, which is a 400 percent increase in possible throughput.

Throughput Measures

Transmission Flaws

Factors that can degrade network performance include noise, attenuation, and latency.

• Noise: Interference that degrades or distorts a signal, measured in dB (decibels). A loss of 3 dB indicates the signal has lost half its power. Common sources include:
  • EMI (electromagnetic interference): Caused by motors, power lines, and other electrical devices or even severe thunderstorms. Contains RFI (radio frequency interference) caused by radio waves.
  • Crosstalk: When a signal on one wire infringes on a signal on an adjacent wire. Types include:
    • Alien crosstalk: Occurs between two cables.
    • NEXT (near end crosstalk): Occurs between wire pairs near the signal source.
    • FEXT (far end crosstalk): Measured at the far end of the cable from the signal source.
• Attenuation: Loss of a signal’s strength as it travels. Compensated for using a repeater, which regenerates the signal. A switch on an Ethernet network works as a multiport repeater, as the bits transmitted “start over” at each port on the switch.
• Latency: Delay between when data leaves the source and arrives at its destination. Factors include cable length, intervening devices, noise, traffic congestion, and conversions. Measured by RTT (round trip time) in milliseconds. Varying delays can cause jitter (PDV - packet delay variation), leading to problems with streaming video or voice.

Duplex, Half-Duplex, and Simplex

• Full-duplex: Signals travel in both directions simultaneously (e.g., telephone).
• Half-duplex: Signals travel in both directions, but only one direction at a time (e.g., an apartment building’s intercom system).
• Simplex: Signals travel in only one direction (e.g., broadcast radio).

In Windows, you can use Device Manager to configure a NIC, including speed and duplex settings. Choosing Full Duplex, Half Duplex, or Auto Negotiation. if you specify a particular speed and duplex that’s not supported by the neighboring device, the result is a speed and duplex mismatch and, therefore, slow or failed transmission.

Multiplexing

A form of transmission that allows multiple signals to travel simultaneously over one medium is known as multiplexing.

• TDM (time division multiplexing): Divides a channel into time slots, reserved for designated nodes regardless of data to transmit.
• STDM (statistical time division multiplexing): Assigns time slots based on priority and need, using all slots.
• FDM (frequency division multiplexing): Assigns different frequencies to create subchannels, allowing multiple signals at one time. Telephone companies once used FDM for all phone lines and still multiplex signals on resi dential phone lines for the last leg before entering a residence.
• WDM (wavelength division multiplexing): Carries multiple light signals over fiber-optic cable by dividing a light beam into different wavelengths (colors). Newer bidirectional WDM supports full-duplex light transmissions in both directions at the same time.
• DWDM (dense wavelength division multiplexing): Increases the number of channels in WDM, typically used on high-bandwidth WAN links.
• CWDM (coarse wavelength division multiplexing): Spacers frequency bands wider apart to lower cost, but has limited distance.

Copper Cable

Coaxial Cable and Twinaxial Cable

Coaxial Cable

Coaxial cable (“coax”) was the foundation for Ethernet networks in the 1980s and is still used for cable Internet, cable TV, and some multimedia connections. It has a central metal core, an insulator, a braided metal shielding, and an outer cover (sheath). RG stands for radio guide. A cable’s AWG (American Wire Gauge) refers inversely to the size of the conduct ing core. Lower impedance results in better power transfer, and higher impedance yields less attenuation of the data signal over a distance. An impedance of 50 ohms was determined to be a good compromise in these factors for computer networks and CB (citizens band) or ham radio connections.

• RG-59: Used for relatively short connections, distributing video signals. Less expensive but has greater attenuation.
• RG-6: Used for broadband cable Internet and cable TV.

RG-6 and RG-59 cables can terminate with:

• F-connector: Threaded connector screwed together like a nut-and-bolt assembly. Most often used with RG-6 cables.
• BNC connector: Connects via a turn-and-lock mechanism (bayonet coupling). Used with RG-59 cables and, less commonly, with RG-6.

Twinaxial Cable

Twinaxial cable (“twinax”) looks similar to coax but has two cores and support much higher throughput than coax. More recent twinax cables contain multiple pairs of these cores to carry even more data. Twinax can be a better choice than fiber to carry 10-Gigabit signals or higher over very short distances. Twinax is also called a DAC (direct attach copper) cable, which is a copper cable designed to handle very high-speed connections at very short distances.

• Passive Twinax: Suitable for distances less than 5-7 meters.
• Active Twinax: Contains internal electrical components for distances up to 10 meters.

Twinax is factory terminated with transceivers. Depending on the connector type, twinax can support throughput up to 100 Gbps. However, the higher data rates require even shorter distance limitations.

Twisted-Pair Cable

Twisted-pair cable consists of color-coded pairs of insulated copper wires twisted around each other and encased in a plastic sheath. Ethernet networks contain four wire pairs. Jill West describes on networks using Gigabit Ethernet and higher standards, with a data rate of 1000 Mbps or more, use all four pairs for both sending and receiving.

The TIA/EIA 568 standard divides twisted-pair wiring into several categories: Cat 3, 5, 5e, 6, 6a, 7, 7a, and 8.

• Cat 3: 10 Mbps, up to 16 MHz, used for wired telephone connections.
• Cat 5: 100 Mbps, 100 MHz, minimum for Fast Ethernet.
• Cat 5e: 1000 Mbps (1 Gbps), 350 MHz, higher-grade version of Cat 5.
• Cat 6: 1 Gbps (or 10 Gbps at shorter distances), 250 MHz, contains a plastic core to prevent crosstalk.
• Cat 6a: 10 Gbps, 500 MHz, reduces attenuation and crosstalk.
• Cat 7: Not included in TIA/EIA standards 10 Gbps (or up to 100 Gbps at shorter distances), 600 MHz, each wire pair is wrapped in shielding.
• Cat 7a: Not included in TIA/EIA standards 40–100 Gbps at very short distances, 1000 MHz, uses increased bandwidth to offer higher data rates than Cat 7, but still requires specialized connectors to reach full potential.
• Cat 8: Class I (Cat 8.1) and Class II (Cat 8.2) 25 Gbps and 40 Gbps at longer distances than Cat 7, 2 GHz. Optimized for short-distance backbone connections within the data center and supports up to 40 Gbps over 30 meters (98 feet).

NOTE 5-4 The more twists per foot in a pair of wires, the more resistant the pair will be to crosstalk or noise. Higher-quality, more expensive twisted-pair cable contains more twists per foot. The number of twists per meter or foot is known as the twist ratio. Because twisting the wire pairs more tightly requires more cable, however, a high twist ratio can result in greater attenuation.

STP (Shielded Twisted Pair)

STP cable consists of twisted-pair wires with metallic shielding (foil or braided copper) to protect against electromagnetic interference. The shielding must be grounded to enhance its protective effects and prevent reflection issues.

UTP (Unshielded Twisted Pair)

UTP cabling consists of insulated wire pairs encased in a plastic sheath without additional shielding. It is less expensive than STP and less resistant to noise. A plenum-grade cable’s jacket is flame-resistant and non-toxic when burned, while PVC cable’s coating is toxic when burned. A similar cable type, riser-rated cable, is also coated with a fire-retardant jacket and is a thicker cable to make it easier to push or pull through risers in buildings or between floors.

Comparing STP and UTP

Cable Pinouts

TIA/EIA has specified two methods of inserting twisted-pair wires into RJ-45 plugs: TIA/EIA-568A and TIA/EIA 568B (also known as T568A and T568B, respectively).

• T568B: more common and is likely what you’ll find on home and business networks.
• T568A: the federal government requires T568A on all federal contracts for backward- compatibility.

Straight-Through Cable (Patch Cable)

To create one, terminate the RJ-45 plugs at both ends of the cable identically, following one of the TIA/EIA-568 standards. These are designed for most connections. Computers and routers are intended to send and receive signals (MDI - medium dependent interface). In con trast, switches use an alternate port configuration called MDI-X (MDI crossover).

Crossover Cable

Connects like devices (MDI to MDI or MDI-X to MDI-X). TX (transmission) and RX (receive) wires are crossed. As you read earlier, modern devices have an autosense function that enables them to detect the way wires are terminated in a plug and then adapt their transmit and receive signal ing accordingly.

Rollover Cable (Console Cable)

Reverses all the wires without regard to pairing. Used to connect a computer to the console port of a router for configuration changes.

Terminating Twisted-Pair Cable

The tools needed for cable termination are a wire cutter or snips, cable stripper, and cable crimper.

1. Cut the cable to the desired length.
2. Remove the sheath off one end of the twisted-pair cable, beginning at approximately 1 inch from the end and remove the rip cord.
3. Carefully untwist each pair and straighten each wire.
4. Align all eight wires on a flat surface, one next to the other, ordered according to their colors and positions listed earlier in Figure 5-17
5. Measure 1/2 inch from the end of the wires and cleanly cut the wires straight across at this length.
6. Keeping the wires in line and in order, gently slide them into their positions in the RJ-45 plug
7. Place the RJ-45 plug in the crimping tool and press firmly to crimp the wires into place.
8. Remove and test the connection.

PoE (Power over Ethernet)

IEEE 802.3af standard specifies a method for supplying electrical power over twisted-pair Ethernet connections. It supports nodes needing power far from receptacles. It provides 15.4 watts for standard PoE devices and 25.5 watts for PoE+ devices (802.3at standard).

• PSE (power sourcing equipment): Supplies the power.
• PDs (powered devices): Receive power from the PSE.

PoE requires Cat 5 or better copper cable. Current may run over an unused wire pair or the pair used for data transmission.

Adding PoE to Non-PoE Networks

• Injector (midspan): Connects to a non-PoE switch or router to inject power onto the network.
• Splitter: Attaches to a non-PoE client to receive power over the Ethernet connection.

Ethernet Standards for Twisted-Pair Cable

Fiber-Optic Cable

Fiber-optic cable (“fiber”) contains glass or plastic fibers at its core. Data is transmitted via pulsing light from:

• Laser: For extremely long distances with very high throughput.
• LED (light-emitting diode): For shorter connections.

The fibers are surrounded by cladding (glass or plastic), a plastic buffer, strands of Kevlar for protection, and a plastic sheath. Because each strand transmits in one direction (simplex), two strands are needed for full-duplex communication. A zipcord cable combines two strands side by side in conjoined jackets.

A newer technology allows bidirectional transmission on both ports, which means each fiber cable carries data in both directions. It uses the newer bidirectional WDM technology to separate the data traveling in each direction on different wavelengths of light, or colors. To work, it requires special equipment on each end of the con nection called a BiDi (pronounced bye-dye) transceiver, also called a WDM transceiver.

Fiber Benefits

• Extremely high throughput
• Very high resistance to noise
• Excellent security
• Ability to carry signals for much longer distances before requiring repeaters

Fiber Characteristics

• Throughput: Up to 100 gigabits per second per channel.
• Cost: More expensive than copper cabling.
• Noise immunity: Unaffected by EMI.
• Size and scalability: Segment lengths vary from 2 to 40,000 meters due to optical loss (degradation of the light signal).

Types of Fiber Cable

SMF (Single Mode Fiber)

SMF consists of a narrow core (8 to 10 microns) where laser-generated light travels a single path with very little reflection. It accommodates the highest bandwidths and longest distances.Internet backbone depends on single mode fiber. However, because of its relatively high cost, SMF is rarely used for short connections, such as those between a server and switch.

MMF (Multimode Fiber)

MMF contains a larger core (50 or 62.5 microns) over which many pulses of light travel various angles. It experiences greater attenuation and is not suited to distances longer than a few kilometers. On the other hand, MMF is less expensive to install and, therefore, is typically used to connect routers, switches, and servers on the backbone of a network or to connect a desktop workstation to the net work.

FDP (fiber distribution panel)

The transition between SMF and MMF cabling might occur at an FDP (fiber distribution panel), which is usually a case on a rack where fiber cables converge, connect with each other, and connect with fiber-optic terminal equipment from the ISP.

Terminating Fiber-Optic Cable

A typical fiber termination kit might include a fiber stripper and fiber cleaver.

Fiber Connectors

SMF connectors are classified by the size and shape of the ferrule (extended tip). SMF connectors are designed to reduce back reflection (return of the light signal back into the fiber transmitting the signal). Back reflection is measured as optical loss in dB (decibels). Shapes and polishes cur rently used on SMF ferrules to reduce back reflection include the following:

• UPC (ultra-physical contact): Extensive polishing creates a rounded surface, which allows the two internal fibers to meet and increases efficiency over older types of connections.
• APC (angled physical contact): Uses a polished curved surface, but the end faces are placed at an angle to each other; the industry standard for this angle is 8 degrees. Notice how the APC connection reflects any signal loss in a different direction than the source of the signal.

The most common 1.25-mm ferrule connector is LC (local connector). Two 2.5-mm ferrules are SC (subscriber connector or standard connector) and ST (straight tip). MT-RJ (mechanical transfer registered jack). Older fiber networks might use ST or SC connectors.

The MT-RJ connector is unique because it contains two strands of fiber in a single ferrule. With two strands per ferrule, a single MT-RJ connector provides full-duplex signaling. SC and LC connectors are also available in full-duplex mode.

Media Converters

Media converters are hardware that enables networks or segments running on different media to interconnect and exchange signals. You must select the correct media converter for the type of fiber being connected, whether it’s SMF to copper or MMF to copper. Converters are also needed to connect networks using MMF with networks using SMF.

Fiber Transceivers

Fiber Tranceivers are modular interfaces that can be plugged in sockets and are hot-swappable.

• GBIC (Gigabit interface converter): A standard transceiver for Gigabit Ethernet connections.

  Newer transceivers that have made the GBIC obso lete include the following:

• SFP (small form-factor pluggable): Provides the same function as GBICs and is more compact, allowing more ports per linear inch. Also known as mini GBICs or SFP GBICs. Typically used for 1 Gbps connections, but theoretically capable of 5 Gbps.

• XFP (10 Gigabit small form-factor pluggable): Supports up to 10 Gbps and is slightly larger than SFP with lower power consumption than SFP+.

• SFP+: Developed later than XFP and is the same module size as SFP; theoretical maximum transmission speed is 16 Gbps. SFP+ transceivers are still widely used today.

• QSFP (quad small form-factor pluggable): Complies with the 802.3ba standard, squeezing four channels in a single transceiver and supporting data rates up to 40 Gbps (4 x 10 Gbps).

• QSFP+: Generally the same technology as QSFP while supporting data rates over 40 Gbps. Highest speed format at the time of this writing is QSFP56-DD, which doubles the data lanes to eight and supports a total theoretical maximum data rate of 400 Gbps (8 x 50 Gbps). The twinax cable you saw earlier in Figure 5-10 is terminated with QSFP+ transceivers.

• CFP (centum form-factor pluggable): Intended for 100-Gbps network connections, with each succeeding generation (CFP2, CFP4, CFP8) becoming smaller and more energy-efficient.

Ethernet Standards for Fiber-Optic Cable

Key Details

• 100BASE-SX: Low-cost Fast Ethernet solution using 850-nm wavelength light signal. Maximum segment length depends on fiber diameter and modal bandwidth.
• 100BASE-FX: Fast Ethernet at longer 1300 nm wavelength, rated up to 2 kilometers. Similar standards are 100BASE LX and 100BASE-BX.
• 1000BASE-SX: Gigabit Ethernet using 850 nm wavelengths, best for short network runs.
• 1000BASE-LX: Gigabit Ethernet using 1300 nm wavelengths, used for long backbones.
• 10GBASE-SR: “Short range” 10-Gigabit Ethernet using 850 nm wavelengths and MMF, distances vary.
• 10GBASE-LR: “Long range” 10-Gigabit Ethernet using 1310 nm wavelengths, extending to 10 kilometers.

Common Fiber-Cable Problems

• Fiber type mismatch: Connecting cables with different core widths (e.g., 50-micron and 62.5-micron cores).
• Wavelength mismatch: Using transmissions optimized for one cable type over a different cable type.
• Dirty connectors: Signal loss and errors due to dust on connectors.
• Link loss: The losses from distance along the cable, losses from multiplexing, and losses from imperfect connections patches, or splices.

Cable Troubleshooting Tools

Symptoms of cabling problems can range from lost packets to breaks in connectivity. Start by checking network connection LED status lights. If cabling is suspected, use tools to analyze and isolate problems.

• Toner and Probe Kit:
  • Tone generator (toner): Issues a signal on a wire.
  • Tone locator (probe): Emits a tone when detecting electrical activity. Used to determine where a wire terminates. Tone generators should never be used on a wire that’s connected to a device’s port .
• Multimeter: Measures resistance, voltage, and impedance to verify cable conduction, check for noise, and test for short or open circuits. You might use a multimeter to
  • Measure voltage to verify that a cable is properly conducting electricity
  • Check for the presence of noise on a wire (by detecting extraneous voltage).
  • Test for short or open circuits in the wire (by detecting unexpected resistance or loss of voltage).
• Cable Continuity Tester (Cable Checker/Tester): Tests whether a cable is carrying a signal to its destination. Some continuity testers will perform a wire map test, and it indicates that each pin on one end is paired with the appropriate pin on the other end.Common problems a wire map test can detect:
  • Reversed pair—The wires of one pair (for example, the orange pair) are reversed with each other when they shouldn’t be.
  • Crossed pair—Two pairs are reversed with each other when they shouldn’t be.
  • Split pair—One wire from each of two pairs are reversed with each other when they shouldn’t be
• Cable Performance Tester (Line Tester/Certifier/Network Tester): Sophisticated device that measures the overall performance of a cabling structure. Also perform continuity and fault tests.
  • Measure the distance to a connectivity device, termination point, or damage in a cable.
  • Measure attenuation along a cable.
  • Measure NEXT (near end crosstalk) between wires as well as alien crosstalk
  • Measure termination resistance and impedance
  • Issue pass/fail ratings for various categories of cabling standards