Read (and Download) the U.S. Department of Transportation Microwave System Planning Guide, describing how microwave links for traffic control are to be correctly installed. This information is quite applicable to general telecommunications and Wi-Fi/WiMAX as well.

How do you determine the required height for an antenna tower or mast for and outdoor Wi-Fi link?

There are two factors that come into play: the curvature of the Earth and the need to maintain not only a clear line-of-sight, but a clear area surrounding the direct line-of-sight. This area surrounding the line-of-sight is called the First Fresnel Zone (pronounced "frah-nell"). A detailed engineering discussion of Fresnel diffraction and its significance in Wi-Fi networking, as well as a mathematical derivation of the formulae used in determining Fresnel Zone radius is presented in the downloadable paper, "I'm Going To Let My Chauffeur Answer That", a discussion of math and physics for the 802.11 wireless LAN engineer. A general discussion and a table of heights, is also presented as part of this tech note.

How do wireless network analysis tools measure RF signal strength?

Since software-based network analysis tools (like AirMagnet, Sniffer Wireless, and AiroPeek) use an off-the-shelf 802.11 PCMCIA NIC to capture packets they can not provide the accuracy of a hardware based spectrum analyzer (which may cost ten times more than the wireless analysis tool). The 802.11 chipset provides a value called the Receive Signal Strength Indicator (RSSI) which is used by the chipset to determine when the channel is clear (and transmission is allowed), when it's time to roam to a new access point, and what data rate to use for transmission. Vendors use an algorithm or a look-up table to convert RSSI values to dBm, mW, or signal strength percentage. This is part of the reason why two different analysis tools may report different values for signal power even though they are sitting next to each other on the desk in front of you. Plus, environmental influences cause the signal to fade. This causes the measurement value to fluctuate by as much as 10 dB or more. A white paper is available for download that provides a more complete discussion.

Can Cisco's LEAP or other Temporal Key rotation protocols (like WPA TKIP) be hacked?

Lightweight Extensible Authentication Protocol (LEAP), the mutual authentication algorithm created by Cisco prior to Wi-Fi Protected Access (WPA) is vulnerable to dictionary attacks. A tool called "Asleap" was made available to the Internet community on April 6, 2004 and it allows the discovery of the password/key used for LEAP encryption. Discovery is accomplished using a "dictionary attack" in which variations of passwords and keys are tried off-line in an attempt to discover one that unlocks the encrypted information. Users of LEAP are encouraged to select passwords and keys that are not easily guessed. The use of words, names, or phrases are discouraged. Cisco already has alternative forms of EAP that are not subject to the disclosed attack. In addition, the IEEE 802.1x authentication mechanisms (other then LEAP) also are not subject to attack using the Asleap tool. More information is available at Cisco's web site.

Is a 200 mW access point a good thing?

The answer is normally, "No." A notebook computer or PDA using a standard PCMCIA (or built-in Wi-Fi) adapter transmits with a power level of 100 mW (or less). If an access point puts out a 200 mW signal then the access point will be capable of transmitting to a distance from which the notebook computer can't transmit back. The Wi-Fi client (the notebook computer) "hears" the access point. Unfortunately, the access point has no way to "hear" the client. The client connection fails, even though the signal level seen from the access point may be stronger than other, lower power, access points. The advantage of a 200 mW access point is that they can be used with a 50-foot length of low-loss antenna cable to attach a remote external antenna. The cable attenuates the 200 mW signal down to a 100 mW level at the point where the signal enters the antenna. Now the access point antenna and the client are both operating with a 100 mW signal, and they can hear each other properly. The output power from an access point should never exceed the output power from the clients that are attaching to it.

What is a wireless Mesh Router?

The term "wireless Mesh router" refers to a device that resembles a standard access point but which forwards packets to other mesh routers to create an infrastructure, as opposed to providing client association services. The essence of the mesh router concept is that of a "wireless Ethernet". By this it's meant that the mesh routers connect together to form a "cloud", through which data packets can pass transparently. These packets originate on Ethernet segments attached to the mesh routers. Often, mesh routers operate at higher power levels than standard access points since they are designed only to communicate to each other, and they aren't constrained by the need to have a power output that's consistent with client devices.

What are the transmit power limitations for 802.11 devices?

In the United States, the Federal Communications Commission defines power limitations for wireless LANS in FCC Part 15.247. The wording in the FCC rules is somewhat complex and has led to some misunderstandings. There are two values to be considered in assessing transmit power: TPO (Transmitter Power Output) and EIRP (Equivalent Isotropically Radiated Power, sometimes defined as Effective Isotropically Radiated Power). TPO is a measure of the power being delivered to the transmitting antenna and EIRP is the result of adding any antenna gain. Although the details of the rules for the 2.4 GHz and 5.8 GHz band are slightly different, the essence is as follows:

• Maximum TPO:    1 watt (1000 mW, 30 dBm)
• Maximum EIRP for non-point-to-point links:
  • 802.11b
    • 100 mW (20 dBm)
  • 802.11g   
    • 50 mW (17 dBm) @24 Mbps and less, 40 mW (16 dBm) @ 36 Mbps,
      31.6 mW (15 dBm) @48 Mbps, 20 mW (13 dBm) @54 Mbps
  • 802.11a   
    • 40 mW (16 dBm) @24 Mbps and less, 25.1 mW (14 dBm) @36 Mbps, 20 mW (13 dBm) @48&54 Mbps
• Maximum EIRP for fixed, point-to-point links using an antenna with a minimum 6 dB gain:
  • EIRP Maximum = 4 watts (4000 mW, 36 dBm) where TPO Max = 30 dBm
  • For antennas with greater than 6 dB of gain:
    • Reduce TPO from 1 watt by 1 dB for each 3 dB of additional antenna gain beyond 6 dB
    • This allows E IRP to be greater than 36 dBm for antennas with greater than 6 dB gain).

Additional information is available at: http://www.cisco.com/univercd/cc/td/doc/product/wireless/cb21ag/acau01/auappb.htm#73700

How do the new wireless standards (802.16 WiMAX and 802.20) compare with each other, and with conventional 802.11 Wi-Fi?

802.16 is designed as a point-to-multipoint technology, similar to the cell network, in which base stations provide access to multiple users.  802.16 uses frequencies in the 10-66 MHz range, so line of sight between the client and the base station is required.  At those frequencies, it supports data rates up to 120 Mbps at ranges of up to 30 miles.

802.16a is an extension to 802.16 that loosens some of these restrictions and makes 802.16 behave a bit more like 802.11.  802.16a uses OFDM at the PHY, at frequencies from 2 to 11 MHz, both licensed and unlicensed depending on the user's need and capabilities.  At those frequencies, line of sight restrictions are similar to those we are used to with 802.11: line of sight is preferred, but not required.  802.16a supports data rates of up to 70 Mbps.  In addition, it has built-in support for a mesh architecture, as opposed to 802.16's hub-and-spoke architecture.  Neither 802.16 nor 802.16a supports stations moving at faster than walking speed, but the 802.16e working group is studying this.  802.16e is an extension to 802.16 that is designed to address 802.16's lack of support for mobile stations (defined as stations moving faster than approximately walking speed--cars, trains, and so forth).

802.20 is similar in scope to 802.16 in that it is designed to provide WLAN throughput over MAN areas, but it differs from 802.16 in several ways.  Whereas 802.16 is currently designed to support devices moving slower than walking speed, 802.20 is designed to support devices moving at up to 150 miles per hour relative to the antenna.  Although 802.16e is designed to address 802.16's lack of support for mobile stations, 802.20 may support higher speeds.  802.20 is specified up to 150 mph relative to the antenna, while 802.16e is only specified for "vehicular speeds", and claims speeds up to about 90 mph in simulations.  802.20 provides very low latency--20 ms or less--but at much lower data rates than 802.16--up to only 1 Mbps.  The approximate range of an 802.20 base station is 15 km.

At its core, 802.16 is intended to be used in more of a distribution role, such as a telco beaming phone service to many subscribers, each of which has an antenna on the top of his or her building.  Some have predicted that the phone companies intend 802.16 as an eventual replacement for their aging in-ground wire networks.  In this form, 802.16 has a range of approximately 30 miles.  Interestingly, the 802.16 MAC layer provides support for at least two key protocols: ATM and TCP/IP.  The ATM component supports the proposition that 802.16 is eventually intended to carry voice.  Because of its line-of-sight restriction, 802.16 would not be appropriate for "man on the street" wireless access.

802.16a, with its mesh architecture and relaxed line-of-sight requirements, would be particularly appropriate for "man on the street" wireless access, especially considering its much higher throughput compared to 802.20.  Where 802.16 stumbles and 802.20 shines is in support for mobile users.  In addition, 802.20's guaranteed low latency may also make it more appropriate for voice communications, and some have even suggested that it may be able to act as a replacement for the existing cellular network.