WLANs operate in one of the unlicensed radio frequency bands under the regulation of the FCC (or its equivalent in other countries). These bands are called the Industrial, Scientific, Medical (ISM) and the Unlicensed National Information Infrastructure (U-NII) bands. These bands are reserved for use by low-power radio equipment that does not require a radio operator’s license to use (some WLANs can be operated in the ISM band at much higher power if the operator acquires a license). Table 8.1 lists the information of interest.

The core WLAN standard is IEEE 802.11, sometimes known as Wi-Fi because this was the first standard championed by the Wi-Fi Alliance. Wi-Fi is short for Wireless Fidelity, a retro/hip reference to Hi-Fi. The IEEE ratified 802.11 in 1997.

In much the same way that Ethernet was standardized by 802.3 with subsequent iterations of improved Ethernet getting extra letters to distinguish them from the original (such as 802.3u, 802.3z, 802.3ae, and so on), the subsequent variations and improvements to 802.11 are distinguished by a letter as well. The ones you want to remember are 802.11a, 802.11b, and 802.11g. There are a couple others in the works; you have probably seen 802.11n gear in stores. The CCENT and CCNA tests are not talking about 802.11n yet, but we’ve included a little of it here for interest and perspective, and because we think it will be included on the exam one of these days.

Table 8.2 lists the WLAN standards you should know, and some of the relevant info about each. (Some of the terms and acronyms will be explained in later sections.)

802.11, sometimes known as legacy, specifies an 11Mbps maximum speed, using 11 channels in the 2.4GHz band. The 11 channels available were used in a random, rapid sequence to statistically avoid interference from other devices using the same frequencies. This “skipping around” the channels is called Frequency Hopping Spread Spectrum (FHSS). In reality, a lot of the data was lost to interference anyway, and a more sophisticated system was engineered for 802.11a.

802.11a uses a much higher frequency (5GHz) and a fancy method of using the available channels, called Orthogonal Frequency Division Multiplexing (OFDM). The science of how OFDM works is well beyond the scope of this exam, but you should know the term and which standards it applies to. A big advantage of the 5GHz band is that it is immune to common 2.4GHz emissions, such as from cordless phones, baby monitors, microwave ovens, and many of the wireless conveniences we take for granted that can really interfere with WLAN transmissions in the 2.4GHz range. A disadvantage is that the higher frequencies are more easily absorbed by structures and furniture, reducing the effective range. However, the way OFDM works actually gives it a range advantage in these office-type surroundings; in addition, higher frequencies mean smaller antennas, which means we can increase the antenna gain (sort of like turning up the listening volume). These things improve the range and so balance out the range loss of the higher frequencies. It was mostly enterprise customers who liked (and often still like) 802.11a, in part because it never caught on with the general public so the risk of interference and security breaches was reduced.

802.11b, although later in the standards list, was actually in the market before 802.11a. 802.11b is back in the 2.4GHz range, so interference sources are a concern, but now we have another method of using the available channels called Direct Sequence Spread Spectrum (DSSS). Again, the complex science behind how it works is not of concern for the CCNA exams, but you need to understand a little about it. There are eleven channels within the frequency band assigned to this standard by the FCC. Typically, only three of these channels (channels 1, 6, and 11) are ever used because all 11 channels overlap each other. When two APs use channels that are adjacent or close together (say, 2 and 3) or the APs themselves are close enough together to “jam” each other, the signals from one channel get stepped on, interfered with, and generally disrupted by the other. Channels 1, 6, and 11 do not overlap each other, as shown in Figure 8.1.

Figure 8.1. 2.4GHz DSSS 11-channel overlap showing channels 1, 6, and 11 nonoverlapping.

By sticking to these three channels, the frequencies are far enough apart that it is very unlikely that they will interfere with each other, so less data is lost. In an environment where you have multiple APs (as in most business implementations), you take advantage of this by having your different access points using different channels, consequently overlapping their coverage area without overlapping the channels they use that could cause interference. We’ll mention this again when we talk about BSS and ESS a little later.

802.11g is currently the most popular standard, mostly because the home market has significantly driven demand. Consumers wanted more speed and backward compatibility with the wireless devices they already had, and they wanted it quickly and cheaply, well before the standard was ratified. The home market exploded, but the business market held back until that ratification happened to avoid deploying immature technology (another reason that you will find more 802.11a than 802.11g in business environments). 802.11g is backward compatible with 802.11b and uses the same 2.4GHz frequency band, but provides the same higher speeds of 54Mbps and slightly longer range. This was a difficult engineering feat to achieve, and if 802.11b and 802.11g do coexist in the same system, the 802.11b tends to impair the performance of the 802.11g system. 802.11g uses OFDM for most of its possible data speeds, but reverts to an 802.11b-compatible system for others. Many wireless devices now provide a, b, and g capability in a single unit.

The problem with the popularity of 802.11g goes back to overcrowding and interference. As I sit here in my home office, I can pick up no less than 12 wireless APs, and all of them are using the same channel. Add in all the other interfering devices in the 2.4GHz spectrum, and it’s a safe bet that these wireless networks are providing less than optimal performance. A business-class WLAN is expected to be of much higher quality than that, and the challenges of providing that kind of quality are significant.

Table 8.2 lists 802.11n, which at the time of this writing is not yet ratified by the IEEE but is expected to be so in 2008, possibly as late as 2009. Manufacturers are already selling “prestandard” 802.11n devices that comply with the draft standard, which will presumably be upgradeable if the standard should substantially change. The big draw is, of course, speed, with theoretical speeds of 600Mbps. This opens up the exciting possibility of wirelessly streaming HD-format movies from your media center to the TV in the bedroom without excessive delay. In addition to OFDM and the use of the 5GHz band, 802.11n uses a feature called Multiple Input Multiple Output (MIMO), which requires two sending antennae and two receiving antennae, and allows for up to four send and four receive. Effectively, you use multiple concurrent streams to achieve the high data rates. That’s why the 802.11n APs often have several antennae on the box.

802.11n is exciting because it is fast, has great range, uses the 5GHz band (reducing the interference and overcrowding problem) and is backward compatible with a, b, and g. The only caution is that all of this is still prestandard; if you are going to buy something, make sure the box at least says it is compliant with the 802.11n draft standard.

Ad Hoc is Latin for “this purpose”; in this case, it refers to a connection between wireless devices (such as two laptops, for example) for a temporary high-speed link, perhaps to share files or play a game. No AP is required; indeed, no network devices other than the two client machines are needed. Ad Hoc mode is not usually associated with business-class networks.

In Infrastructure mode, an AP is required because the client devices cannot send directly to each other; they must send through the AP. The AP, in turn, handles all the wireless client data and provides the connection to the wired LAN by way of its connection to the switch. Infrastructure mode is what you will find in business environments. With the use of an AP, many devices can connect to the network, and you can provide authentication and encryption for wireless connections to help secure the WLAN.

BSS is defined as a single AP that provides network connectivity for its associated clients. You could have several APs in your system, but they would each be offering a separate WLAN, and you could not “roam” between the APs; your laptop would need to associate itself with each new AP when you lost signal from the old one as you walked around the building.

ESS is more typical of a business environment. With ESS, each AP still defines a BSS, but the group of APs and their BSSs form the ESS. The main advantage is that clients can roam between the overlapping coverage areas of the individual APs’ BSSs, without losing the functionality of their connection to the ESS and thus to the network. To form an ESS, all the APs use the same Service Set Identifier (SSID). An SSID is just a name (technically, it is a string of up to 32 alphanumeric characters) that identifies the WLAN. You might have three APs for the office area, all configured with the same SSID, with overlapping coverage areas. Each wireless client changes its association to the AP with the strongest signal, which will change depending on where you are in the office and where you move to. It is typical to have the APs use different channels, to reduce interference and keep the throughput speed up. As you move, even though your association to the BSS of the AP changes to that of the next one, your association to the ESS provided by the three APs together does not, so your perception is that you never lose your connection to the network. Figure 8.2 shows what an ESS system might look like; there are two APs, each using the same SSID but on different channels. Mobile devices such as laptops, PDAs, or a wireless IP Phone can move between the two APs without losing connectivity to the network. In the diagram, the wireless IP Phone is roaming between the APs.

Figure 8.2. Two APs in Infrastructure mode using the same SSID form an ESS to allow devices to roam.

Any network is vulnerable to attack. A wired network requires that you physically connect to a cable to execute an attack. If there is no possibility of an outside connection to your wired system, an attack must come from an available wired connection and is much easier to find and stop.

A WLAN’s versatility and convenience makes it a huge security vulnerability. By adding an AP to your network, you are making it possible for people to connect to your network without being in physical contact with it. The following are just some of the more common WLAN threats that we need to guard against:

These scenarios make a business network admin freak out. For quite a while, businesses were not implementing WLANs because they knew that the lack of security was a major issue.

There are three main ways to apply security to WLANs:

As usual, different vendors were keen to get their product to market and start making money, so there were some nonstandardized security systems in place before the standards came out, but things have settled down somewhat into a fairly consistent and pretty secure standard that most vendors are supporting. It goes without saying that if you do not actually apply the security, it is not secure!

The next section describes the four WLAN security standards you need to know for the exam.