Wireless Connectivity Update

Connect802 is a nationwide wireless data equipment reseller providing system design consulting, equipment configuration, and installation services.

Call today and ask about your 30% design discount when you buy a Wi-Fi wireless network from Connect802

FOURTH QUARTER 2010


Product Focus Exploring the Connect802 value proposition...	Elfiq Broadband Load Balancers
Essential Wi-Fi For those who are new to Wi-Fi networking...	Next Steps in 802.11 Evolution
Technology and Engineering For the engineer and Wi-Fi network administrator...	Identifying Multipath Fading
To Infinity... and Beyond! News from the wireless marketplace...	Cloud-based WPA Key Cracking

Product Focus

Elfiq Link Balancers

Link balancing and redundancy has often been difficult to implement, especially for small and medium businesses. Enterprise customers can handle the high price and complex configuration of load-balancing routers from vendors like Cisco, but what about a customer who doesn’t want to spend that much money, or who already has routing equipment and simply needs to add load balancing or redundancy?

Elfiq’s line of link balancing appliances make it easy and cost-effective for customers to add multiple Internet uplinks to an existing network. For example, a customer could purchase multiple DSL lines or a combo of DSL and cable-modem, in order to increase upstream bandwidth and/or provide redundancy in the case of a link failure. Elfiq’s appliances present two or more Internet uplinks to the internal network as a single link, and automatically take care of all load balancing, QoS, failover, and so forth. Additionally, they are easy to configure, with a web-based interface that can be maintained by the average network administrator, as opposed to a specialized, trained technician.

Because of their unique combination of cost-effectiveness, capability, and usability, Elfiq is Connect802’s choice for link balancing and redundancy appliances. Contact Connect802 sales at 925-552-0802 for more information.

Essential Wi-Fi

Next Steps in 802.11 Development

Development of the 802.11 standard never slows down. You can tell how long someone has been an 802.11 engineer by which standard was “just around the corner” when he or she entered the industry. (If you know anybody who pre-dates 802.11b, they’ve been doing it a while!) For a long time, 802.11a and 802.11g were established standards, and 802.11n was “just around the corner.” Now that 802.11n has been ratified, what’s next? This issue, we’ll highlight 802.11ac and 802.11ad, two standards that you’ll be hearing more about in the future. (Don’t let the name fool you. 802.11ac and 802.11ad are not sub –standards of 802.11a. The IEEE simply ran through the entire alphabet—802.11a through 802.11z—and then started over at the beginning with 802.11aa, 802.11ab, 802.11ac, and so on.)

Both 802.11ac and 802.11ad aim to provide gigabit-level speeds over wireless. Aside from the normal demand for more throughput, one specific application driving this goal is high-definition video. The ultimate outcome would be to get rid of the rat’s-nest of cables that connects the Blu-Ray player, the Xbox, and so forth to the receiver to the television. Another application driving this goal may be the connection of peripherals like hard drives to a PC. Standards such as Ultra-Wideband (UWB) and Wireless USB have attempted to address this role, but none of them have gained widespread acceptance. Perhaps an 802.11-backed version would be more palatable to the market?

802.11ac

One defining characteristic of 802.11ac is that it will operate in the same 5 GHz ISM frequency band that 802.11a and 802.11n use today. The 5 GHz band offers more channels than 2.4 GHz and a wider frequency bandwidth, so already its potential for throughput is higher. But 802.11n already wrings pretty much all of the throughput out of the air that can be expected, especially with backwards compatibility a requirement. So how will 802.11ac achieve its throughput targets? Increased frequency bandwidth will probably be one way. 802.11n offers either 20 or 40 MHz channels. 802.11ac may use as wide as 160 MHz channels (the equivalent of eight 802.11a channels) to achieve its goals.

An Intel study concluded that 160 MHz of bandwidth, combined with 4x4 MIMO was necessary to achieve 1 Gbps of MAC-level throughput. Connect802’s experience has been that MIMO radios often do not achieve the maximum number of spatial streams they’re capable of, so even if 802.11ac radios were designed to handle four spatial streams, their ability to reach 1 Gbps with 160 MHz of bandwidth would be questionable. Additionally, a 160 MHz channel would create channel allocation problems, where APs would overlap with each others’ channels, similar to what we see in the 2.4 GHz band now. Intel argues that one way to resolve this would be to use multiple separate 20 MHz channels, which would increase efficiency, allowing 1 Gbps of throughput to be reached with only 100 Mhz (five channels’ worth) of frequency bandwidth.

802.11ad

802.11ac’s re-use of the 5 GHz ISM band allows it to be backwards-compatible with 802.11n and 802.11a. This means that older clients could coexist in the same space as 802.11ac devices, but it also means that maximizing 802.11ac’s throughput will be problematic. Just as 802.11g and 802.11a devices can reduce the throughput of an 802.11n network, 802.11ac devices will probably be slowed down by the presence of, and the need to interoperate with, legacy devices. That brings us to 802.11ad.

802.11ad operates in an unlicensed frequency band near 60 GHz . All other things being equal, higher-frequency signals can carry more data than lower-frequency signals, because the base carrier frequency is higher. This means that 60 GHz signals are inherently more capable of carrying higher data rates compared to lower-frequency signals.

Although some transmitters currently operate in the 60 GHz band, the band is not widely used, so interference with other devices can be expected to be low. Interference is further reduced by the fact that these high-frequency signals do not pass through walls and other obstructions very well. From the perspective of interference, this is good. A device could be operating in one room, and a device in a nearby room could be operating on the same channel without interference, because the walls will block the signal. From the perspective of RF coverage, of course, this is terrible! 60 GHz devices will not be expected to cover, say, a whole house or office, like their 2.4 and 5 GHz cousins do.

So what good is a gigabit-speed wireless device that can’t pass through walls? As was mentioned before, this technology is perfect for carrying high-definition video and audio from A/V components like Blu-Ray players, Xboxes, and so forth, to their corresponding display device. Devices do exist now that carry high-definition video over wireless in 2.4 and 5 GHz, but these devices typically must contain an H.264 encoder and decoder, and they re-encode the video on the fly to reduce it to a level that can be carried by the non-gigabit wireless technology. Of course, this degrades the quality of the signal, which home-theater enthusiasts hate. A gigabit speed wireless technology like 802.11ad would allow un-compressed video and audio to stream over the network, straight off the disc.

A second use for 802.3ad would be wireless PC peripherals—not so much low-throughput devices like keyboards and mice, but more like hard drives and video cards. A 1 Gbps wireless technology could make a cordless external hard drive almost as fast as an internal hard drive. Imagine that!

Conclusion

The only thing we know for sure at this point is that these technologies are on the table and under development. Only a few hard details are known for sure: that the technologies target 1 Gbps and will use the 5 GHz and 60 GHz frequency bands. The exact technological details that will be used to achieve these goals will take years (or more, if 802.11n is any guide!) to sort out. Of course, the question that’s on your mind is probably, “When can I buy it?”

We’ll let you know as soon as we do.

Technology and Engineering

Recognizing and Troubleshooting Multipath Fading

You know that spot in the office where your wireless VoIP always drops? Or the bin in the warehouse that you can never scan? But as soon as you walk three feet in any direction, suddenly the wireless network is performing as expected! This type of issue is not usually indicative of a normal gap in coverage. Instead, multipath fading may be the cause.

Diagnosing Multipath

Signal from an access point’s antenna propagates in many directions. As it moves through the environment, it bounces off of obstructions, forming reflections. Multipath fading occurs when two or more reflections overlap in such a way that they cancel each other out. The specific locations in which multipath will occur are very difficult to predict, because multipath depends on very fine details in the environment, but the symptoms of multipath make it relatively easy to diagnose.

Multipath fading causes a dramatic, consistent drop in signal strength, that is highly localized to a relatively small area. The most important of these characteristics is that the fading is localized to a small area. With multipath fading, moving a few feet in any direction, or even a few inches in some cases, causes signal strength to return to normal. If the reduction in signal strength occurs over an area larger than a few feet, then multipath is not the most likely of causes.

Troubleshooting Multipath

Recognizing multipath is only the first step. Once you’ve diagnosed it, what can be done? Multipath is caused by the interaction between the signal and the environment, but changing the environment is not the ideal way of addressing it. Large, flat surfaces like walls are usually a factor, but moving them isn’t usually an option, nor is coating them with less-reflective material. Regardless, moving obstructions will simply move the fading-point, not eliminate it. Finally, the effect of moving obstructions will probably be unpredictable; you’ll know you’re moving the fading-point, but not where to.

The primary way of addressing multipath is through diversity antennas on the access point. Diversity was specifically invented to address multipath, so it’s not surprising that it’s the best solution. In order for diversity to work, the AP needs to have two antennas, and diversity needs to be turned on (which is nearly always the default setting). Some APs, mostly consumer-grade ones, don’t have dual antennas, and don’t support diversity.

When external antennas are used, it’s common to unintentionally disrupt the operation of antenna diversity. Many external antennas don’t support diversity. If your antenna has two connectors coming out of it, it’s probably a diversity antenna. If it’s only got one connector, it’s not.

With a diversity antenna, you simply plug the antenna connectors into your AP and you’re good to go. You can make a diversity pair out of two non-diversity external antennas by mounting them an odd multiple of one-half wavelength apart. For 2.4 GHz, one-half wavelength is about 2.35”, and for 5 GHz, one-half wavelength is about 1.175”. Multiply these distances by any odd number, while keeping total distance between the antennas below about 5’ to get an acceptable distance for diversity to work well. Keep in mind that even multiples of one-half wavelength are the worst for diversity, so accuracy is crucial here. You only have to be an inch or two off to take diversity from great to non-existent. For this reason, it’s recommended to stick to antennas with diversity built-in when multipath is expected to be an issue.

A more subtle way of dealing with diversity is to reduce the output power of the APs. It might sound counter-intuitive that reduced output power would increase signal strength at the receiver, but multipath fading is caused when two or more reflections add up out-of-phase. The stronger the output power of the AP, the more reflected signal energy there is. When you reduce the output power of the AP, reflections from faraway objects get weak enough that they don’t contribute to the fading. Of course, by reducing the AP’s output power, you’re also reducing the signal strength everywhere else in the network, so there is a balance to be had here. Increased AP density at reduced output power may be a way of addressing multipath, but it probably should be designed into the network from the beginning. Retro-fitting a network at a reduced output power is usually less efficient than designing that way from the beginning.

The final, and potentially the best, way of dealing with multipath is to migrate to 802.11n. Through its technological wizardry, 802.11n turns multipath from a liability into an asset. Whereas with 802.11a, b, and g, multipath reduces or eliminates throughput, with 802.11n, the AP takes advantage of the multipath reflections to create multiple spatial streams. This means that the more multipath you have, the more throughput you can get. While this provides some hope for the future, currently there are still lots of devices out there that only support 802.11a, b, or g. Fortunately, 802.11n APs, with their two to four antennas, can still do good old-fashioned diversity if they need to, and better in some cases than an 802.11a, b, or g AP would!

A "2X2 MIMO" 802.11 implementation showing two separately reflected spatial streams

To Infinity... and Beyond!

Cloud-based WPA and WPA-2 Key-Cracking

Most of the stories we bring you in this section of the newsletter are exciting developments in the wireless industry. Depending on your perspective, this one may not be seen as good news. The web site http://wpacracker.com will attempt to crack your WPA or WPA-2 pre-shared key using their custom-dictionary and server farm. Pricing runs at $17 or $35 depending on how fast you want the attack run. All that’s required from you is a pcap trace file containing an authentication to the network.

From the perspective of a network administrator trying to secure a network against attackers, this could be bad news. $17 is pocket change to an attacker who has a good reason to want to get into your network. On the other hand, the attack only works against passwords that are found in the site’s 135-million word dictionary. If you’re using PSK authentication (and not WPA-Enterprise with RADIUS), then you should already know to use passwords consisting of randomly generated letters and numbers, and you should be fairly immune to dictionary attacks. If you’re using WPA-Enterprise with RADIUS, you needn’t worry about this particular attack.

From the perspective of a security auditor, this is excellent news. The peach-of-mind of knowing whether your network’s pre-shared key will stand up against the most sophisticated attempt to crack it may well be worth the time and minimal cost.

Of course, none of this matters if users share the password amongst themselves or leave it stuck to the bulletin board, or anything like that. Ultimately, security is a multi-pronged fork, and the strength of your password may not be your weakest link.

Web Searching: The Connect802 Web Presence

At Connect802 we're your PAGE ONE resource for wireless networking!

Connect802 has the experience, expertise, and resources to help you with your wireless network system. Use your favorite search engine and see what Connect802 is doing. Each month we give you some suggested search terms for you to explore. Here's this month's list. As you look down the search engine results you'll find Connect802 either at the top, or on the first page (true for Google and Excite, unknown for the rest).

Ruckus Wireless Authorized Reseller
Hospital WiFi Consulting
802.11n Consulting
Medical Telemetry Consulting
Wireless Valley
RF CAD Design
WLAN Design and Product Sales

The Contents Of This Website Are Subject To Copyright - Unauthorized Reproduction Prohibited