A Little Wireless Wackiness

In my other posts, I covered some ground on the types of material covered by the CCNA to give readers clarification on topics that may come off as confusing. Wireless tends to be an after thought however, even though it's a super fun topic filled with depth.  As with my other posts, I'll provide a recap of things I've learned and separate subjects by dividers, so you can skip around.  There are a lot of topologies and network types covered in wireless material, but I'll go over core ones real fast before elaborating on specifics.

WPAN's or Wireless Personal Area Networks, connect devices within intimate ranges, so try to think of things like Bluetooth when you hear about them. They use a FHSS (Frequency Hopping Spread Spectrum) and a 2.4GHz ISM (Industrial Scientific Medical) band, which I'll get into in a bit. People probably have familiarity with wireless LAN's, so I won't go into specifics here, but WMAN's (Wireless Metropolitan Area Networks), are large city spanning networks that fall under the WiMAX (Worldwide Interoperability for Microwave Access) category, and WWAN's (Wireless Wide Area Networks) consist of point-to-point and point-to-multipoint connections over huge distances.      

It's important to briefly bring up site surveys, because they're used to determine what areas have interference when deciding how many access points to install, and where to install them.  You want to have a good idea of what radio frequency coverage looks like, and there are quite a few things that cause some pretty gnarly interference.  In any case, wireless uses radio frequency to communicate, and the FCC (Federal Communications Commission) helps to oversee regulations involved with correct band usage.

Do you recall in my Introduction to Networks post where I brought up how CSMA/CD and CSMA/CA (Carrier Sense Multiple Access - Collision Detection and Collision Avoidance), help with issues in both wireless and copper on the data link layer? Let's rehash this a bit with a quote because it's good to know:

Collision Detection, used by Ethernet, has end devices checking for signals and transmits them in their absence.  If it finds a signal, or ends are "busy" everything stops. Collision Avoidance, used by wireless, checks for signals, and in their absence, sends a request notification with intent to use. Once it gets an "OK", it sends data out.  

Let's break this down a little: We use CSMA/CA, because the carrier sense (what we refer to as the device sensing the medium) is air based.  This is extremely important to grasp because as an air based medium, you can't really have a collision. Part of the reason for this is due to what's known as the hidden station problem. With a limited range it's entirely possible that stations may not recognize each other because of the nature of the medium. If transmissions did overlap, things would just get lost. To work around this problem, CA uses a "listen before talk" style strategy and implements RTS and CTS (Request to Send and Clear to Send) mechanisms for transmissions instead. In order to avoid collisions, devices also utilize a CCA (Clear Channel Assessment), which is a technique that detects radio frequency energy on devices transmitting, and announces how much time is needed for frame exchanges, but I'm not going to delve into this, or Distributed Coordination Functions. Just know the core issue is the one I listed above.

Spread-spectrum technology uses a low power over a wide range of frequency, which takes digital information in 0's and 1's and modulates it between devices that use it. In order for this to work, the same spread-spectrum and modulation types have to be used between devices, so liken this to understanding someone who speaks the same language(s) as you do (if you speak Spanish to someone unfamiliar with it, you're going to have a bad time right?). The two primary types here are FHSS: Frequency-Hopping (Spread Spectrum), which I mentioned earlier, and DSSS: Direct-Sequence. Both operate in the same 2.4 GHz range, but FHSS sends small amounts of information rapidly through many frequency channels (or a hop sequence), which both a transmitter and receiver are synced to communicate with.

DSSS on the other hand, is a technique where a message signal modulates a PN (pseudo-noise) code, which includes characteristics like phase, amplitude, and frequency. It then uses a spreading code, like Barker, to provide data redundancy and help make it more resilient to interference. First used by the military in the 1940's, DSSS allows a receiver to determine if a single bit is 0 or 1, through a chip that combines the information with this Barker code through a XOR process. The DSSS channel is 1 of 14 in the 2.4-2.5 GHz ISM band range and is 22 MHz wide. The country and location a device is from helps determine which channels are usable.

802.11b introduced HR/DSSS (High Rate), which uses a Complementary Code Keying for transmitting data and allows for a potential of up to 600 Mbps for n. However, it wasn't until 2003 and 802.11g that ERP (Extended Rate Physical) was introduced with an amendment that allowed for higher data rates on devices, and a need for support with ERP-DSSS, and ERP-OFDM was seen. With Orthogonal Frequency Division Multiplexing, interference is reduced by splitting a channel over many narrow bands at different frequencies.    

There is a lot of material on this subject, but I'm only going to quickly bring up MIMO (Multiple-Input Multiple-Output) technology before moving onward, because it introduced physical and data link enhancements. Prior to a, b, and g, devices used a single radio to transmit and receive signals with SISO (Single-Input Single-Output), but MIMO changed this by allowing multiple radios, or radio chains, for signal transmission, through the use of reflections to enhance performance. I won't get into TxBF (Transmit beamforming), MRC (Maximal Ratio Combining), SM (Spatial Multiplexing), or STBC (Space Time Block Coding), but they're all subjects you can look into in your spare time if you feel up for it.

A Little on Radio Frequency Basics

Have you read this entire thing but still aren't sure how radio frequency works? Let's go into this real fast before covering antenna types. Radio frequencies are waves used in communication. Alternating current is passed over copper connected to antennas, which transform received signals into these radio waves (or air). An AC signal is a sine wave, that results in electrical current varying in voltage over time. The frequency is the number of waves, or cycles, that pass through in a certain time amounts.

Radio transmissions need transceivers, or wireless stations that can transmit and receive these AC signals and antennas and transform them into waves that carry information.  A wavelength is the distance of one complete cycle of 1 oscillation of an AC signal (λ), measured in cm. When you have a low frequency, you have a longer wave, and when you have a higher frequency, you have shorter waves.

Amplitude refers to the amount of power a radio frequency signal contains and is calculated from the height of the Y axis of this sine wave (representing voltage and change over time). When we have an increase in power, it's known as a gain, while a decrease is known as a loss, or attenuation. The phase is the difference in degrees at particular points in the times of a cycle from certain angle measurements. Lastly, multipaths are when waves arrive at receivers out of phase with distortion, which can cause corruption. Radio frequencies are divided into bands, which are further split into channels. The capacity, or number of devices that can connect to an AP, affects performance.  

Fresnel Zones

With wireless networking, radio frequency communication and line of sight becomes an issue, along with varying forms of interference that play a role in signal strength. There are two primary types of line of sight: One visual, and the other a direct link with radio frequencies.  With visual line of sight, transmitters and receivers can see each other because there isn't any obstruction in their path. Fresnel zones, coined after the physicist Augustin-Jean Fresnel, are ellipsoidal volumes that surround the direct line of sight between two points, which can be used to determine the strength of waves (pictured below).

At the beginning of this post, I mention a number of things can cause interference and while I won't go into co-channel related stuff here, I will briefly list off some interference types you should be aware of:

  • Reflection happens when signals hit a smooth, non-absorbing surface and bounce, causing a direction change. Things like certain wall types, or tables might cause it.
  • Refraction happens when signals pass between items of different densities, which might bend or change speeds. The signal comes into contact with an obstacle like glass, gets refracted as it passes through and attenuation happens.
  • Diffraction happens when signals pass through obstacles, and the wave changes direction by bending around the obstacle. Think body of water here.  
  • Scattering happens when signals strike an uneven surface and waves reflect into multiple directions, causing degradation.
  • Absorption happens when material absorbs a signal, like a body of water. Large populations of people are actually problematic because of their high water content in this case.  
  • Diffusion happens when a signal widens as it leaves an antenna. Eventually, it decreases in amplitude and becomes less powerful. This is also known as FSPL, or Free Space Path Loss.

RF Measurements

There are some basic units of measurement for radio frequency that I'll go over a teeny bit, but the main thing to know is that 1 milliwatt (mW) = 1/1000th of a watt and that dB, dBi, dBd and dBm are different units of measurement. The main formula you can look further into elsewhere is:

P = E * I
Power (P) = Voltage (E) * Current (I)

With signal measurements, there are tools like spectrum analyzers that can view different statistics though (if you want to play around with some of this, check out Acrylic Home WiFi). I'm not going to go much into the math here.

Whenever people bring up the SNR, or Signal-to-Noise Ratio, they're referring to the difference between the amount of a received signal and the "noise floor".  If you imagine a crowded room where everyone is speaking at high volume, you can also imagine that it would be harder to hear your partner sitting next to you.  The same rule applies with signal strength, so the difference between the signal (what you hear) and the noise floor (what you're surrounded by noise-wise) becomes pretty important.

If a device receives signal of -85 dBm and the noise floor is -95: The SNR is 10dB
-85 dBm -(-95dBm) = 10dB (Not good!)

If a received signal is -65dBm and the noise floor is -95dBm: SNR would be 30dB
-65dBm -(-95dBm) = 30dB (Excellent!)

Antenna Basics

Lastly, I'll cover a little on some antennas, so you can get a better grasp on the fun projects you might see people embark on in this area. We learned by now that antennas takes energy from transmissions and transform them into radio waves, so they can propagate through the air, but there are a lot of different antenna types and each have pros and cons.

When we talk about radio frequency lobes, we're referring to the shape of radiation patterns, or beam that contains the strongest point. Beamwidth is the horizontal and vertical angles of an antenna and there are charts that measure azimuth (spherical coordinates) and elevation with polarization, or the horizontal and vertical orientation of gains.

When it comes to wireless LAN antenna types, there are three main ones in use: Omnidirectional (dipole), semidirectional, and highly directional. With omnidirectional antennas, radio frequencies are propagated in every direction horizontally. Rubber ducks are a popular type (shown below) because they can be adjusted regardless of AP position.

With semidirectional antennas, radio frequencies focus transmissions into a more specified pattern, and things like panels, sectors and yagi are used (pictured below is a yagi). Radomes are covers that help to protect antennas and highly directional antennas, like parabolic dishes, are more or less for long-range connections.

I won't go into the specifics of each, but now that you know a little bit about this subject, you can do further research and know what to look for. Heck, maybe you can start by building your own cantenna in the process!

Happy Learning!