Networking 101
Copyright Priscilla Oppenheimer
Study Topics 2
Unguided (Wireless) and Guided (Wired) Media
Wireless Communications
Radio frequency bands
- Most are licensed by government agencies such as the FCC
- Unlicensed bands exist for use by cordless phones, wireless LANs, and other industrial, scientific, and medical purposes
Spread spectrum technology
- Developed in the 1940s
- Spreads the transmission signal over a broad band of radio frequencies
- Power is reduced, but the frequency band usage is increased
- This means less interference and makes spread spectrum appropriate for unlicensed uses, such as wireless LANs, cordless phones, etc.
Frequency-hopping spread spectrum (FHSS): transmissions hop from one frequency to another
Direct-sequence spread spectrum (DSSS): uses a chipping sequence where each bit is represented by a string of 1s and 0s. Even if some of the string is lost, the original transmission can be reconstructed.
Orthogonal Frequency Division Multiplexing (OFDM): divides a channel into subchannels and encodes a portion of the signal across each subchannel in parallel
Spatial Division Multiplexing (SDM): transfers multiple independent data streams simultaneously within one spectral channel of bandwidth; each spatial stream requires a discrete antenna at both the transmitter and receiver
IEEE Wireless Standards
|
802.11 |
802.11a |
802.11b |
802.11g |
802.11n |
Bandwidth |
1-2 Mbps |
54 Mbps |
1, 2, 5.5, and 11 Mbps |
54 Mbps |
600 Mbps |
Frequency band |
2.4 GHz |
5 GHz |
2.4 GHz |
2.4 GHz |
2.4, 5 GHz |
Modulation |
DSSS and FHSS |
OFDM |
DSSS with better coding technique |
OFDM |
SDM |
Features |
Hmm, can't think of any |
Fast |
Widely implemented; products from many vendors tend to interoperate well |
Fast; Compatible with 802.11b |
Fast; multiple-input multiple-output (MIMO); frame aggregation |
Concerns |
Security; slow; frequency band is over-populated |
Security; not widely implemented |
Security; frequency band is over-populated |
Security; frequency band is over-populated |
Security; pre-standard implementations may not interoperate well |
Wireless topologies and devices
- Ad hoc network: shapeless topology created by wireless NICs on their own (no central control)
- Infrastructure mode: a star topology with an access point at the center
- Access Point (AP): a wireless hub that has an Ethernet connection for access to the cabled LAN (and presumably the Internet)
Active scanning versus passive scanning
- With active scanning, the client sends a Probe Request to join the network.
- The Probe Request contains the Service Set Identifier (SSID)
- The AP replies with a Probe Response.
- With passive scanning, the client listens for Beacon messages transmitted by the AP and learns the SSID.
- Because a user must know the SSID in order to use active scanning, it is considered more secure.
Wireless Security
|
Wired Equivalent Privacy (WEP) |
Extensible Authentication Protocol (EAP) |
Lightweight EAP (LEAP) |
Wi-Fi Protected Access (WPA) |
Virtual Private Network |
Features |
Encrypts data
Authenticates client device
Requires client to be configured with the same encryption key as the AP (shared key)
Simplicity |
Uses an authentication server
Authenticates users, not just devices (server can ask for username and password) |
Uses an authentication server
Authenticates users, not just devices
Also authenticates the AP to avoid naive users associating with a rogue AP |
Supports Temporal Key Integrity Protocol (TKIP)
Supports Advanced Encryption Standard (AES) with WPA2
Supports WPA Personal for home (no server required) & WPA Enterprise for offices |
Uses an authentication server
Authenticates users, not just devices
Most secure
Data is sent in a tunnel to a business's central HQ (usually)
Uses strong encryption |
Concerns |
Encryption is weak
A hacker can capture traffic and determine the key |
Too simple (doesn't change the key with each frame, for example) |
Requires Cisco-compatible NIC and AP |
Can be confusing to configure due to many options |
Lots of overhead
Difficult to configure
May not work with all applications |
Wired Communications
Copper Media
Advantages of copper media
- Conductivity
- Corrosion resistance
- Ductility
- Malleability
- Strength
Types of copper cabling
- Shielded twisted-pair (STP)
- Unshielded twisted-pair (UTP) - used by Ethernet
- Coaxial (coax)
Why is twisted-pair cabling twisted?
- Helps reduce and cancel noise
Advantages of UTP
- Small diameter, so can fit most places
- Does not require grounding
- Cables are relatively easy to make and install
- Inexpensive
- Has all the advantages of copper media (see above)
Disadvantages of UTP
- More susceptible to electrical noise and interference than STP or coax cable
- Can't go as long distances as coax cable (Max cable length is about 100 meters)
Ethernet connectors and cables
- RJ-45 most common connector for UTP Ethernet
- Cable can be straight-through or crossover
Straight-through cable
- Maintain the pin connection all the way through the cable
- Wire connected to pin 1 is the same on both ends
- Used to connect PCs or routers to hubs or switches
Crossover cable
- Cross the critical pair to properly align transmit and receive signals on devices with like connections
- Pin 1 connected to pin 3; pin 2 connected to pin 6
- Used to connect similar devices: switch to switch, switch to hub, hub to hub, router to router, PC to PC
UTP cable categories
- Categories 1 through 6 (higher number is better)
- Most new installations run a minimum of Category 5 Enhanced (CAT 5e) for voice and data.
- The older categories are becoming less relevant.
- In the future, CAT 6 will be common, although right now it's expensive.
Optical Media
Advantages of optical media
- Not susceptible to electromagnetic interference (EMI) or radio frequency interference (RFI)
- Greater bandwidth capabilities than other media
- Newer transmitter and receiver technologies can replace older ones without replacing the fiber itself to muster even more bandwidth out of the cable
- Longer distances
- More secure (harder to tap)
- Cheaper for long distances
- No grounding required
- Fiber cable is light (no pun intended) in weight, so it is easy to move around
Disadvantages of optical media
- You have to be careful when bending the cable
- The ends of the fiber must be properly polished
- Connectors must be carefully installed (requires an expert)
- Splices must be carefully made (requires an expert)
- For short distances, optical media is more expensive than UTP
- Network interface cards that have fiber-optic connectors are much more expensive than ones that have RJ-45 jacks for UTP
The ray model of light
When electromagnetic waves, including light, travel out from the source, they travel in straight lines called rays. When a light ray (called the incident ray) crosses the boundary from one material to another, the light can be:
- Reflected, which means the light returns to the material
- Refracted, which means the light crosses the boundary and changes direction and speed
The Index of Refraction (IR) is the speed of light in a vacuum divided by the speed of light in a material. It's essentially a measure of the density of a material and how quickly light can travel thought it. You don't have to memorize these, but FYI, to help you understand the concept:
- The IR for air is about 1.
- The IR for water is about 1.3.
- The IR for glass is about 1.5, though it depends on the type of glass.
- The IR for a diamond is about 2.4.
For fiber-optic cabling, the core (inside part of the cable) has a higher IR than the cladding (outside part of the cable). The cladding is less dense, very pure glass. This helps the light bounce back into the core of the cable. The important concept to understand is that we want the light to travel in the core (inside part) of the cable. Technically this happens because of the Law of Refraction and the critical angle, as described below.
The Law of Refraction (Snell's Law) in somewhat tekkie terms:
- If light travels from a substance whose IR is smaller into a substance where the IR is larger, the refracted ray is bent toward the vertical axis.
- If light travels from a substance whose IR is larger into a substance where the IR is smaller, the refracted ray is bent away from the vertical axis.
The Law of Refraction (Snell's Law) applied to fiber-optic cabling:
- When light travels form the core into the cladding, the light is bent away from the vertical axis and back toward the core. This is a good thing; it means the light doesn't escape or get absorbed. Since the light is carrying information, it's important that it stay inside the core and take a reasonably straight path to the destination.
The angle of incidence: The angle between the incident ray and a vertical line perpendicular to the surface.
The critical angle: The incident angle at which light is completely reflected back into the original material rather than being refracted.
Total internal reflection
The goal is that light will be reflected back into the core so it can continue on its merry way down the cable to the recipient. For total internal reflection to work:
- The cladding of the optical fiber has a smaller IR than the core.
- The cladding is less dense than the core.
- The rays are bent away from the vertical axis back into the core.
- The angle of incidence of the light ray must be greater than the critical angle for the core and its cladding.
Single-mode versus multimode cable
Single-mode
- Focused beam of laser light
- Small core
- Less dispersion
- Supports long distances (a few kilometers)
Multimode
- Many rays (modes) of light bouncing around
- LEDs provide light source
- Larger core
- More dispersion
- Shorter distances
Back to Networking 101