Wireless LAN Design Principles

Electromagnetic Waves

In this lesson we’ll go through some of the fundamentals that 802.11 uses for wireless communication, and mostly as it pertains to just RF, or radio frequencies, and how they interact with the environment.

As you probably guessed or know, most wireless communication uses radio frequency signals. Now RF signals are electromagnetic waves. In the chart below we’re showing where the different electromagnetic frequency ranges are and what recognizable uses each area contains. The ranges that we’re using for the wireless communication we’ll be discussing are the 2.4GHz and 5GHz ranges, shown in blue.

You’ve probably seen that before, when you’re configuring a wireless router, the 2.4GHz and 5GHz is where our modern Wi-Fi communication standards are taking place. To give that a little bit of perspective, in the above chart we have TV and AM/FM radio down in the Megahertz range. We’ve got our visible light spectrum way up in the Terahertz and Petahertz range, with x-ray and gamma-ray way in the Exahertz range.

Frequencies

A frequency is a descriptor of something cyclical. Hertz is a unit of measure (named after its inventor, Heinrich Rudolf Hertz) describing how many cycles that a process goes through per second. One cycle per second 1 Hertz (abbreviated Hz). So if we have a radio frequency, or electromagnetic wave which oscillates 1 million times per second, that is 1 megahertz (MHz).

Free Path Loss

Have you ever noticed that even with a clear line of sight between your laptop and a wireless access point, that your signal strength is better when you’re physically closer to access point? The electromagnetic waves do have energy loss when there are no obstructions, due to free path loss. The further away you are from an access point, the weaker the signal will be. This is due to the fact that the electromagnetic waves are emitted with a finite amount of energy, and propagate in all directions. As it spreads further out, the same amount of energy is spread thinner and thinner.

Obstruction Attenuation

Now of course other things that can get in your way are things such as walls and other physical obstructions. These obstructions cause attenuation of the radio wave, meaning a decrease in the wave’s amplitude. The obstruction causes an energy loss, the energy gets transferred into heat in the actual obstruction, and that the wave comes out on the other end with a decreased amplitude. Your receiving device is not able to interpret that attenuated signal as well.

Reflections and Interference

Reflections are actually something very important that have bigger play here than you might realize. They do not inherently cause a change in amplitude, but they can cause multi-path interference. Multi-path interference is a phenomenon that occurs when the same signal takes two different paths to reach a device and ends up interfering with itself due to the different physical distance traveled.

Due to the multiple paths being slightly different in length, and the electromagnetic waves all traveling about the same speed, the same signal arrives to the destination at slightly different times. Depending on when that time is in relation to your frequency will determine how out of phase it is. Depending how out of phase it is it can either amplify the signal, cause an increase in the amplitude of the waves, or it could cause a significant decrease in the signal, possibly even cancel each other out entirely.

These charts here show the radio wave, and the middle line is to give a reference point for how these waves line up in time where the x-axis is time. So an in-phase signal doesn’t need to necessarily arrive at the same time, it just needs to arrive in phase. For example, if the wave arrives one wavelength then it’s just off by one full cycle then it can be in phase. As shown in the multi-path interference diagram above, when two signals arrive, we add them together. When they’re in phase these guys add together to a wave that is twice the amplitude, or both of those amplitudes added together, giving us a better signal, a better amplitude in our signal. This type of interference is call constructive interference.

On the other end of the interference spectrum, if the wave is 180 degrees out of phase, then they are completely out of phase and cancel each other out. Although the reflected signal may not be as strong and you may not have complete 0 resulting signal, this interference is called destructive interference.

Signal Strength

A signal’s strength measurement is called RSSI, or Relative Signal Strength Index. It is measured in decibel milliwatts, abbreviated as dbm. Now decibels are a logarithmic unit, meaning small changes equate to large differences. It is measured in relation to milliwatts for our RSSI. Now RSSI is a negative value, and closer to zero is better. For example, if I have an RSSI of -30dbm, that is better than -40dbm. Similarly, a -20dbm RSSI value is better than -30dbm. Since dbm is logarithmic, a change of approximately 3dbm equates to doubling or halving the signal strength.

Another measurement that’s often used when talking about wireless signal strength is the SNR, or Signal-to-Noise Ratio. Our signal-to-noise ratio is calculated by subtracting our noise floor from the RSSI. What is the ‘noise floor’? This is the other electromagnetic signals that are picked up by the antenna which we don’t care about. As you might imagine, at a loud concert, I wouldn’t be able to hear someone talk even if they were 1 foot from me, because the noise floor is so high. It doesn’t matter that the RSSI of that information is pretty good, of there’s too much noise you still won’t be able to make out the useful signal from the junk.

A good tool to use on your computer to view this type of information about the wireless signals you care about is a free tool called inSSIDer.

Wi-Fi standards

For the CCNA exam, we’ll want to be aware of fiver different Wi-Fi standards. These are 802.11a, b, g, n, and ac. 802.11a operates at a 5GHz frequency and was first released back in 1999. At the same time 802.11b was also released, operating at 2.4GHz. The primary difference here is that higher frequency has lower object penetration, it cannot get through objects as well as our lower frequency. So we get a longer amount of distance with our lower frequency 802.11b, but as you can see we’re sacrificing our max throughput. 802.11b provides a mere 11Mbps of max throughput, where 802.11a gives us a max throughput of 54Mbps. About 4 years later 802.11g was released, which brought 54Mbps throughput to the 2.4GHz band.

In 2009, 802.11n was released, which operates in both the 2.4GHz and 5GHz bands and provided speeds of up to 600Mbps. This was a huge improvement, since the fastest throughput we saw previous to this standard was 54Mbps. The last standard we’ll need to be aware of for the CCNA exam is 802.11ac. We saw significant speed boosts again, up to 2.3Gbps. 802.11ac operates only at the 5GHz band.

Channels and Bands

The radio frequency ranges are regulated, with more transmitting devices existing we need to ensure interference is kept to an acceptable level. The 2.4GHz band is part of the ISM bands. ISM is a collection of radio frequencies that were reserved for Industrial, Scientific, and Medical purposes, giving light to its name. The 2.4GHz Wi-Fi band provides three non-overlapping frequencies that are 22MHz wide.

In the chart above we see the frequency allocations of the 2.4GHz Wi-Fi channels. Although we see there are several sets of 3 channels which are non-overlapping, it is recommended to use channels 1, 6, and 11. This will prevent interference between your own access points, and help you avoid interference from your neighbor’s access points as well.

SSIDs

Most people have configured an SSID before. SSID stands for Service Set Identifier. These essentially equate to a VLAN in a wired network. Cisco does recommend that you have only 1 SSID per VLAN. It doesn’t make much sense to have multiple SSIDs corresponding to a single VLAN, as typically your SSIDs will have different security requirements, and why would you want different security requirements to get onto the same network segment? The name of an SSID can be between 2 and 32 characters long, as defined in the Wi-Fi standard.

Collision Avoidance

At layer 2, Wi-Fi, 802.11, uses CSMA-CA. That’s Carrier Sense Multiple Access Collision Avoidance. That means that before each station transmits, it listens and it waits to see if anything else is transmitting first. It avoids a collision, because there’s no way to transmit full-duplex, Wi-Fi is all half-duplex. Actually what the station ends up doing, is it sends a request to send over to the recipient, and then if the recipient thinks the frequency is clear it’ll send back a clear to send message. In that ‘clear to send’ message it will even say how much time that the transmitting station has available to transmit in. That’s how wi-fi fundamentally works, is that it allows for very small increments of time to be used by each station in the area that needs to transmit. Once finished, the station will wait again until it believes the air is once again clear.

Wireless Security

The de-facto encryption standard for Wi-Fi right now is WPA2. With WPA being defunct due to TKIP, the Temporal Key Integrity Protocol, which uses RC4 as the encryption algorithm that has been found to be insecure. So WPA2 is really the only way to go right now. Both WPA and WPA2 use the same authentication method, with different encryption schemes. Both use the 4-way handshake, described in the illustration above. WPA2 utilizes AES, the Advanced Encryption Standard, for data encryption. Standard WPA2 is configured for authentication using a Pre-Shared Key, or PSK. WPA2 Enterprise supports 802.1X, and allows for authentication to be provided from a RADIUS server, an external authentication server. We’ll get more into wireless configuration later in the CCNA course.

As a final parting note, WPA3 has been released as WPA2’s successor, though not many devices support it yet at the time of this writing. WPA3 uses a new authentication method called Simultaneous Authentication of Equals.