Chapter

12

WLAN Troubleshooting
IN ThIs chApTer, you WILL LeArN AbouT The foLLoWINg:
ÛÛ Layer 2 retransmissions
NÛ

RF interference Multipath Adjacent cell interference Low SNR Mismatched power settings Near/far Hidden node

NÛ

NÛ

NÛ

NÛ

NÛ

NÛ

ÛÛ 802.11 coverage considerations
NÛ

Dynamic rate switching Roaming Layer 3 roaming Co-channel interference Channel reuse/multiple channel architecture Single channel architecture Capacity vs. coverage Oversized coverage cells Physical environment

NÛ

NÛ

NÛ

NÛ

NÛ

NÛ

NÛ

NÛ

ÛÛ Voice vs. data ÛÛ Performance ÛÛ Weather

Copyright © 2009 John Wiley & Sons, Inc.

Diagnostic methods that are used to troubleshoot wired 802.3 networks should also be applied when troubleshooting a wireless local area network (WLAN). A bottoms-up approach to analyzing the OSI reference model layers also applies to wireless networking. A wireless networking administrator should always try to first determine whether problems exist at layer 1 and layer 2. As with most networking technologies, most problems usually exist at the Physical layer. Simple layer 1 problems such as nonpowered access points or client card driver problems are often the root cause of connectivity or performance issues. Because WLANs use radio frequencies to deliver data, troubleshooting a WLAN offers many unique layer 1 challenges not found in a typical wired environment. The bulk of this chapter discusses the numerous potential problems that can occur at layer 1 and the solutions that might be implemented to prevent or rectify the layer 1 problems. A spectrum analyzer is often a useful tool when diagnosing layer 1 issues. After eliminating layer 1 as a source of possible troubles, a WLAN administrator should try to determine whether the problem exists at the Data-Link layer. Authentication and association problems often occur because of improperly configured security and administrative settings on access points, WLAN controllers, and client utility software. A WLAN protocol analyzer is often an invaluable tool for troubleshooting layer 2 problems. In this chapter, we discuss many coverage considerations and troubleshooting issues that may develop when deploying an 802.11 wireless network. RF propagation behaviors and RF interference will affect both the performance and coverage of your WLAN. Because mobility is usually required in a WLAN environment, many roaming problems often occur and must be addressed. The half-duplex nature of the medium also brings unique challenges typically not seen in a full-duplex environment. Different considerations also need to be given to outdoor 802.11 deployments due to weather conditions. In this chapter, we discuss how to identify, troubleshoot, prevent, and fix instances of potential WLAN problems.

Layer 2 Retransmissions
The mortal enemy of WLAN performance is layer 2 retransmissions that occur at the MAC sublayer. As you have learned, all unicast 802.11 frames must be acknowledged. If a collision occurs or any portion of a unicast frame is corrupted, the cyclic redundancy check (CRC) will fail and the receiving 802.11 radio will not return an ACK frame to the transmitting 802.11 radio. If an ACK frame is not received by the original transmitting radio, the unicast frame is not acknowledged and will have to be retransmitted.

Copyright © 2009 John Wiley & Sons, Inc.

Layer 2 Retransmissions

391

Excessive layer 2 retransmissions adversely affect the WLAN in two ways. First, layer 2 retransmissions increase overhead and therefore decrease throughput. Many different factors can affect throughput, including a WLAN environment with abundant layer 2 retransmissions. Second, if application data has to be retransmitted at layer 2, the timely delivery of application traffic becomes delayed or inconsistent. Applications such as VoIP depend on the timely and consistent delivery of the IP packet. Excessive layer 2 retransmissions usually result in latency and jitter problems for time-sensitive applications such as voice and video. When discussing VoIP, latency and jitter often get confused. Latency is the time it takes to deliver a VoIP packet from the source device to the destination device. A delay in the delivery (increased latency) of a VoIP packet due to layer 2 retransmissions can result in echo problems. Jitter is a variation of latency. Jitter measures how much the latency of each packet varies from the average. If all packets travel at exactly the same speed through the network, jitter will be zero. A high variance in the latency (jitter) is the more common result of 802.11 layer 2 retransmissions. Jitter will result in choppy audio communications and reduced battery life for VoWiFi phones. Most data applications in a Wi-Fi network can handle a layer 2 retransmission rate of up to 10 percent without any noticeable degradation in performance. However, time-sensitive applications such as VoIP require that higher-layer IP packet loss be no greater than 2 percent. Therefore, Voice over Wi-Fi (VoWiFi) networks need to limit layer 2 retransmissions to 5 percent or less to guarantee the timely and consistent delivery of VoIP packets. How can you measure layer 2 retransmissions? As shown in Figure 12.1, any good 802.11 protocol analyzer can track layer 2 retry statistics for the entire WLAN. 802.11 protocol analyzers can also track retry statistics for each individual WLAN access point and client station. Unfortunately, layer 2 retransmissions are a result of many possible problems. Multipath, RF interference, and low SNR are problems that exist at layer 1 yet result in layer 2 retransmissions. Other causes of layer 2 retransmissions include hidden node, near/far, mismatched power settings, and adjacent cell interference, which are all usually a symptom of improper WLAN design.

RF Interference
Various types of RF interference can greatly affect the performance of an 802.11 WLAN. Interfering devices may prevent an 802.11 radio from transmitting, thereby causing a denial of service. If another RF source is transmitting with strong amplitude, 802.11 radios can sense the energy during the clear channel assessment (CCA) and defer transmission entirely. The other typical result of RF interference is that 802.11 frame transmissions become corrupted. If frames are corrupted due to RF interference, excessive retransmissions will occur and therefore throughput will be reduced significantly. There are several different types of interference: Narrowband interference A narrowband RF signal occupies a smaller and finite frequency space and will not cause a denial of service (DoS) for an entire band such as the 2.4 GHz ISM band. A narrowband signal is usually very high amplitude and will absolutely disrupt communications in the frequency space in which it is being transmitted.

Copyright © 2009 John Wiley & Sons, Inc.

392

Chapter 12

N

WLAN Troubleshooting

Narrowband signals can disrupt one or several 802.11 channels. Narrowband RF interference can also result in corrupted frames and layer 2 retransmissions. The only way to eliminate narrowband interference is to locate the source of the interfering device with a spectrum analyzer. To work around interference, use a spectrum analyzer to determine the affected channels and then design the channel reuse plan around the interfering narrowband signal. Figure 12.2 shows a spectrum analyzer capture of a narrowband signal close to channel 11 in the 2.4 GHz ISM band. f I g u r e 1 2 .1 Layer 2 retransmission statistics

Wideband interference A source of interference is typically considered wideband if the transmitting signal has the capability to disrupt the communications of an entire frequency band. Wideband jammers exist that can create a complete DoS for the 2.4 GHz ISM band. The only way to eliminate wideband interference is to locate the source of the interfering device with a spectrum analyzer and remove the interfering device. Figure 12.3 shows a spectrum analyzer capture of a wideband signal in the 2.4 GHz ISM band with average amplitude of –60 dBm. All-band interference The term all-band interference is typically associated with frequency hopping spread spectrum (FHSS) communications that usually disrupt HR-DSSS and/or ERP-OFDM channel communications at 2.4 GHz. As you learned in earlier chapters, FHSS constantly hops across an entire band, intermittingly transmitting on very small subcarriers of frequency space. A legacy 802.11 FHSS radio, for example, transmits on hops that are 1 MHz wide. While hopping and dwelling, an FHSS device will transmit in sections of the frequency space occupied by an HR-DSSS or ERP-OFDM channel. Although an FHSS device will not typically cause a denial of service, the frame transmissions from the HR-DSSS and ERP-OFDM devices can be corrupted from the all-band transmissions of the FHSS interfering radio.

Copyright © 2009 John Wiley & Sons, Inc.

Layer 2 Retransmissions

393

f I g u r e 12 . 2

Narrowband RF interference

f I g u r e 12 . 3

Wideband RF interference

Copyright © 2009 John Wiley & Sons, Inc.

394

Chapter 12

N

WLAN Troubleshooting

What Devices cause rf Interference?
Numerous devices, including cordless phones, microwave ovens, and video cameras, can cause RF interference and degrade the performance of an 802.11 WLAN. The 2.4 GHz ISM band is extremely crowded, with many known interfering devices. Interfering devices also transmit in the 5 GHz UNII bands, but the 2.4 GHz frequency space is much more crowded. Often the biggest source of interference is signals from nearby WLANs. If RF interference cannot be eliminated at 2.4 GHz, special consideration should be given to deploying the WLAN in the less-crowded 5 GHz frequency bands. The tool that is necessary to locate sources of interference is a spectrum analyzer. In Chapter 16, “Site Survey System and Devices,” we discuss proper spectrum analysis techniques that should be part of every wireless site survey. Chapter 16 also lists the many interfering devices that can cause problems in both the 2.4 GHz and 5 GHz frequency ranges.

Bluetooth (BT) is a short-distance RF technology defined by the 802.15 standard. Bluetooth uses FHSS and hops across the 2.4 GHz ISM band at 1,600 hops per second. Older Bluetooth devices were known to cause severe all-band interference. Newer Bluetooth devices utilize adaptive mechanisms to avoid interfering with 802.11 WLANs. Digital Enhanced Cordless Telecommunications (DECT) cordless telephones also use frequency hopping transmissions. A now-defunct WLAN technology known as HomeRF also used FHSS; therefore, HomeRF devices can potentially cause all-band interference. The existence of a high number of frequency-hopping transmitters in a finite space will result in some 802.11 data corruption and layer 2 retransmissions. The only way to eliminate all-band interference is to locate the source of the interfering device with a spectrum analyzer and remove the interfering device. Figure 12.4 shows a spectrum analyzer capture of a frequency hopping transmission in the 2.4 GHz ISM band.

Multipath
As discussed in Chapter 2, “Radio Frequency Fundamentals,” multipath can cause intersymbol interference (ISI), which causes data corruption. Because of the difference in time between the primary signal and the reflected signals, known as the delay spread, the receiver can have problems demodulating the RF signal’s information. The delay spread time differential results in corrupted data. If the data is corrupted because of multipath, layer 2 retransmissions will result. In Chapter 16, we discuss active and passive site survey techniques. The main purpose of the active site survey is to look at the percentage of layer 2 retries. If it is determined during the spectrum analysis portion of the site survey that no RF interference

Copyright © 2009 John Wiley & Sons, Inc.

Layer 2 Retransmissions

395

occurred, the most likely cause of the layer 2 retransmissions will be multipath. As pictured in Figure 12.5, WLAN vendor Berkeley Varitronics Systems makes a line of WLAN troubleshooting tools that can detect and then visualize occurrences of multipath and the delay spread into a useful graphical display. f I g u r e 12 . 4 All-band RF interference

f I g u r e 12 . 5

Multipath analysis troubleshooting tool

PHOTOS COURTESy OF BERKELEy VARITRONICS SySTEMS.

There is no way to “fix” multipath indoors because some reflection will always occur, and thus there will always be multiple paths of the same signal. However, many of the negative effects of multipath, including intersymbol interference, can be compensated for with the use of antenna diversity, which is covered in Chapter 4, “Radio Frequency Signal and Antenna Concepts.” High-multipath environments exist indoors in areas such as long corridors and anywhere metal is located (for example, warehouses with metal shelving or

Copyright © 2009 John Wiley & Sons, Inc.

396

Chapter 12

n

WLAN Troubleshooting

metal racks). The use of indoor diversity patch antennas is highly recommended in highmultipath environments. Using a unidirectional antenna will cut down on reflections and thereby decrease data corruption and layer 2 retransmissions. Some RF technologies also compensate for multipath better than other RF technologies. Any 802.11 radio that uses OFDM technology will be more resilient to multipath than any radio using DSSS technology. Therefore, 802.11a and 802.11g radios that both use OFDM will handle multipath better than older legacy 802.11b radios. In Chapter 18, “High Throughput (HT) and 802.11n,” we discuss the 802.11n amendment, which defines the use of High Throughput (HT) clause 20 radios. 802.11n radios use multiple-input multiple-output (MIMO) technology, which actually takes advantage of multipath. It should be noted that 802.11n clause 20 radios (HT) are required to be backward compatible with older clause 18 radios (HR-DSSS), clause 17 radios (OFDM), and clause 19 radios (ERP). 802.11n access points will not solve the problems that multipath creates for WLANs because older client devices will still be negatively affected by multipath.

Adjacent Cell Interference
Most Wi-Fi vendors use the term adjacent channel interference to refer to degradation of performance resulting from overlapping frequency space that occurs due to an improper channel reuse design. In the WLAN industry, an adjacent channel is considered to be the next or previous numbered channel. For example, channel 3 is adjacent to channel 2. As you learned in Chapter 6, “Wireless Networks and Spread Spectrum Technologies,” the 802.11-2007 standard requires 25 MHz of separation between the center frequencies of HR-DSSS and ERP-OFDM channels in order for them to be considered nonoverlapping. As pictured in Figure 12.6, only channels 1, 6, and 11 can meet these IEEE requirements in the 2.4 GHz ISM band in the United States if three channels are needed. Channels 2 and 7 are nonoverlapping, as well as 3 and 8, 4 and 9, and 5 and 10. The important thing to remember is that there must be five channels of separation in adjacent coverage cells. Some countries allow the use of all fourteen IEEE 802.11-defined channels in the 2.4 GHz ISM band, but because of the positioning of the center frequencies, no more than three channels can be used while avoiding frequency overlap. Even if all fourteen channels are available, most vendors and end users still choose to use channels 1, 6, and 11. When designing a wireless LAN, you need overlapping coverage cells in order to provide for roaming. However, the overlapping cells should not have overlapping frequencies, and in the United States only channels 1, 6, and 11 should be used in the 2.4 GHz ISM band to get the most available, nonoverlapping channels. Overlapping coverage cells with overlapping frequencies cause what is known as adjacent cell interference. Although 802.11 radios might have overlapping frequency space, they do have different center frequencies and can therefore transmit at the same time. If overlapping coverage cells also have frequency overlap, the transmitted frames will become corrupted, the receivers will not send ACKs, and layer 2 retransmissions will significantly increase. Later in this chapter, we discuss channel reuse patterns that are used to mitigate adjacent cell interference.

Copyright © 2009 John Wiley & Sons, Inc.

Layer 2 Retransmissions

397

f I g u r e 12 .6

2.4 GHz nonoverlapping channels
25 MHz 3 4 5 25 MHz 8 9 10 11 12 13 14

1

2

6

7

1

6

11

Only three channels are nonoverlapping.

As defined by the IEEE, 23 channels are currently available in the 5 GHz UNII bands, as pictured in Figure 12.7. These 23 channels are technically considered nonoverlapping channels because there is 20 MHz of separation between the center frequencies. In reality, there will be some frequency overlap of the sidebands of each OFDM channel. The good news is that you are not limited to only three channels, and all twenty-three channels of the 5 GHz UNII bands can be used in a channel reuse pattern, which is discussed later in this chapter. f I g u r e 12 .7
20 MHz 36 40 44 48 52 56 60 64 149 153 157 161

5 GHz nonoverlapping channels

5 GHz UNII-1 and UNII-2 bands

5 GHz UNII-3 band

100 104 108 116 120 112 124 128 132 136 140

5 GHz UNII-2 Extended band Twenty-three nonoverlapping channels possibly available at 5 GHz

Low SNR
The signal-to-noise ratio (SNR) is an important value, because if the background noise is too close to the received signal or the received signal level is too low, data can get corrupted

Copyright © 2009 John Wiley & Sons, Inc.

398

Chapter 12

N

WLAN Troubleshooting

and retransmissions will increase. The SNR is not actually a ratio. It is simply the difference in decibels between the received signal and the background noise (noise floor), as depicted in Figure 12.8. If an 802.11 radio receives a signal of –85 dBm and the noise floor is measured at –100 dBm, the difference between the received signal and the background noise is 15 dB. The SNR is therefore 15 dB. Data transmissions can become corrupted with a very low SNR. If the amplitude of the noise floor is too close to the amplitude of the received signal, data corruption will occur and result in layer 2 retransmissions. An SNR of 25 dB or greater is considered good signal quality, and an SNR of 10 dB or lower is considered poor signal quality. To ensure that frames do not get corrupted, many vendors recommend a minimum SNR of 18 dB for data WLANs and a minimum SNR of 25 dB for voice WLANs. f I g u r e 12 . 8 Signal-to-noise ratio
–25 –30 –35 –40 –45 –50 –55 –60 –65 –70 –75 –80 –85 –90 –95

Received signal level

Fade margin Approx. 28 dB Signal-to-noise ratio (SNR) Approx. 53 dB Unusable signals

Received signal threshold

Ambient noise floor

Mismatched Power Settings
An often overlooked cause of layer 2 retransmissions is mismatched transmit power settings between an access point and a client radio. Communications can break down if a client station’s transmit power level is less than the transmit power level of the access point. As a client moves to the outer edges of the coverage cell, the client can “hear” the AP; however, the AP cannot “hear” the client. As depicted in Figure 12.9, if an access point has a transmit power of 100 mW and a client has a transmit power of 20 mW, the client will hear a unicast frame from the AP because the received signal is within the client station’s receive sensitivity capabilities. However, when the client sends an ACK frame back to the AP, the amplitude of the client’s transmitted signal has dropped well below the receive sensitivity threshold of the AP’s radio. The ACK frame is not “heard” by the AP, which then must retransmit the unicast frame. All of the client’s transmissions are effectively seen as noise by the AP, and layer 2 retransmissions are the result.

Copyright © 2009 John Wiley & Sons, Inc.

Layer 2 Retransmissions

399

How do you prevent layer 2 retries that are caused by mismatched power settings between the AP and clients? The best solution is to ensure that all of the client transmit power settings match the access point’s transmit power. If this is not possible, the access point’s power should never be set to more than the lowest-powered client station. One way to test whether the mismatched AP/client power problem exists is to listen with a protocol analyzer. An AP/client power problem exists if the frame transmissions of the client station are corrupted when you listen near the access point, but are not corrupted when you listen near the client station. AP/client power problems usually occur because APs are often deployed at full power to increase range. Increasing the power of an access point is the wrong way to increase range. If you want to increase the range for the clients, the best solution is to increase the antenna gain of the access point. Most people do not understand the simple concept of antenna reciprocity, which means that antennas amplify received signals just as they amplify transmitted signals. A high-gain antenna on an access point will amplify the AP’s transmitted signal and extend the range at which the client is capable of hearing the signal. The AP’s high-gain antenna will also amplify the received signal from a distant client station. f I g u r e 12 . 9 Mismatched AP and client power

Unicast frame

AP: 20 mW Client: 20 mW ACK frame is heard Unicast frame

ov

era

ge ce l l

ACK frame is not heard by the AP

AP: 100 mW Client: 20 mW

It should be noted that dynamic RF capabilities used by many WLAN controller vendors are notorious for causing mismatched power settings between the lightweight APs and client stations. A WLAN controller might dynamically increase a lightweight AP’s power to a level above the client’s transmit power. Dynamic changes of AP transmit power are well known to cause problems with VoWiFi phones. If the AP cannot hear the phone because of mismatched power, choppy audio may occur or phone conversations may drop entirely. The ratified 802.11k amendment does make it possible for an AP to inform clients to use transmit power control (TPC) capabilities to change their transmit amplitude dynamically to match the AP’s power.

Copyright © 2009 John Wiley & Sons, Inc.

W 20 m

10 0

c

mW c o v e r a g e c e l

l

400

Chapter 12

N

WLAN Troubleshooting

Near/Far
Disproportionate transmit power settings between multiple clients may also cause communication problems within a basic service set (BSS). A low-powered client station that is at a great distance from the access point could become an unheard client if other high-powered stations are very close to that access point. The transmissions of the high-powered stations could raise the noise floor near the AP to a higher level. The higher noise floor would corrupt the far station’s incoming frame transmissions and would prevent this lower-powered station from being heard, as seen in Figure 12.10. This scenario is referred to as the near/far problem. The half-duplex nature of the medium usually prevents most near/far occurrences. you can troubleshoot near/far problems with a protocol analyzer the same way you would troubleshoot the mismatched AP/client power problem. f I g u r e 1 2 .1 0 The near/far problem
Access point Unheard signal

Station B 1 mW

Station A 100 mW

10 ft.

100 ft.

Please understand that the medium access methods employed by Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) usually averts the near/far problem, and that it is not as common a problem as, say, mismatched power between an AP and a client.

Hidden Node
In Chapter 8, “802.11 Medium Access,” you learned about physical carrier-sense and clear channel assessment (CCA). CCA involves listening for 802.11 RF transmissions at the Physical layer, and the medium must be clear before a station can transmit. The problem with physical carrier-sense is that all stations may not be able to hear each other. Remember that the medium is half-duplex and, at any given time, only one radio card can be transmitting. What would happen, however, if one client station that was about to transmit performed a CCA but did not hear another station that was already transmitting? If the station that was about to transmit did not detect any RF energy during its CCA, it would transmit. The problem is that you then have two stations transmitting at the same time. The end result is a collision, and the frames will become corrupted. The frames will have to be retransmitted. The hidden node problem occurs when one client station’s transmissions are heard by the access point, but are not heard by any or all of the other client stations in the basic service set (BSS). The clients would not hear each other and therefore could transmit at the

Copyright © 2009 John Wiley & Sons, Inc.

Layer 2 Retransmissions

401

same time. The access point would hear both transmissions, which would be interfering with each other, and the incoming transmissions would be corrupted. Figure 12.11 shows the coverage area of an access point. Note that a thick block wall resides between one client station and all of the other client stations that are associated to the access point. The RF transmissions of the lone station on the other side of the wall cannot be heard by all of the other 802.11 client stations, even though all the stations can hear the AP. That unheard station is the hidden node. What keeps occurring is that every time the hidden node transmits, another station is also transmitting, and a collision occurs. The hidden node continues to have collisions with the transmissions from all the other stations that cannot hear it during the clear channel assessment. The collisions continue on a regular basis and so do the layer 2 retransmissions, with the final result being a decrease in throughput. A hidden node can drive retransmission rates above 15 to 20 percent or even higher. Retransmissions, of course, will affect throughput, latency, and jitter. f I g u r e 1 2 .11 Hidden node—obstruction

Collisions!

Hidden node

Thick wall obstacle

The hidden node problem may exist for several reasons—for example, poor WLAN design or obstructions such as a newly constructed wall or a newly installed bookcase. A user moving behind some sort of obstruction can cause a hidden node problem. VoWiFi phones often become hidden nodes because users take the phone into quiet corners or areas where the RF signal of the phone cannot be heard by other client stations. Users with wireless desktops often

Copyright © 2009 John Wiley & Sons, Inc.

402

Chapter 12

N

WLAN Troubleshooting

place their radio card underneath a metal desk and effectively transform that radio card into an unheard hidden node. The hidden node problem can also occur when two client stations are at opposite ends of an RF coverage cell and they cannot hear each other, as seen in Figure 12.12. This often happens when coverage cells are too large as a result of the access point’s radio transmitting at too much power. Later in this chapter, you will learn that it is a recommended practice to disable the data rates of 1 and 2 Mbps on an 802.11b/g access point for capacity purposes. Another reason for disabling those data rates is that a 1 and 2 Mbps coverage cell at 2.4 GHz can be quite large and often results in hidden nodes. If hidden node problems occur in a network planned for coverage, then RTS/CTS may be needed. This is discussed in detail later in this chapter. Another cause of the hidden node problem is distributed antenna systems. Some manufacturers design distributed systems, which are basically made up of a long coaxial cable with multiple antenna elements. Each antenna in the distributed system has its own coverage area. Many companies purchase distributed antenna systems for cost-saving purposes, but a hidden node problem as pictured in Figure 12.13 will almost always occur. Distributed antenna systems and leaky cable systems should always be avoided. f I g u r e 1 2 .1 2 Hidden node—large coverage cell

Station A and access point hear each other.

Station B and access point hear each other.

Station A and Station B cannot hear each other.

f I g u r e 1 2 .1 3

Hidden node—distributed antenna system

300 ft. + Heard signal

100 ft. + Access point

Unheard transmissions

Copyright © 2009 John Wiley & Sons, Inc.

Layer 2 Retransmissions

403

So how do you troubleshoot a hidden node problem? If your end users complain of a degradation of throughput, one possible cause is a hidden node. A protocol analyzer is a useful tool in determining hidden node issues. If the protocol analyzer indicates a higher retransmission rate for the MAC address of one station when compared to the other client stations, chances are a hidden node has been found. Some protocol analyzers even have hidden node alarms based on retransmission thresholds. Another way is to use request to send/clear to send (RTS/CTS) to diagnose the problem. Try lowering the RTS/CTS threshold on a suspected hidden node to about 500 bytes. This level may need to be adjusted depending on the type of traffic being used. For instance, let’s say you have deployed a terminal emulation application in a warehouse environment and a hidden node problem exists. In this case, the RTS/CTS threshold should be set for a much lower size, such as 30 bytes. Use a protocol analyzer to determine the appropriate size. As you learned in Chapter 9, “802.11 MAC Architecture,” RTS/CTS is a method in which client stations can reserve the medium. In Figure 12.14, you see a hidden node initiating an RTS/CTS exchange. f I g u r e 1 2 .1 4 Hidden node and RTS/CTS

No more collisions!

RT S

2. CT

1.

S

S

2.

CT 3. Da ta 4. AC
Thick wall obstacle

Hidden node

The stations on the other side of the obstacle may not hear the RTS frame from the hidden node, but they will hear the CTS frame sent by the access point. The stations that hear the CTS frame will reset their NAV for the period of time necessary for the hidden node to transmit the data frame and receive its ACK frame. Implementing RTS/CTS on a hidden

Copyright © 2009 John Wiley & Sons, Inc.

K

404

Chapter 12

N

WLAN Troubleshooting

node will reserve the medium and force all other stations to pause; thus the collisions and retransmissions will stop. Collisions and retransmissions as a result of a hidden node will cause throughput to decrease. RTS/CTS usually decreases throughput as well. However, if RTS/CTS is implemented on a suspected hidden node, throughput will probably increase due to the stoppage of the collisions and retransmissions. If you implement RTS/CTS on a suspected hidden node and throughput increases, you have confirmed the existence of a hidden node. RTS/CTS typically should not be viewed as a mechanism to fix the hidden node problem. RTS/CTS can be a temporary fix for the hidden node problem but should usually be used for only diagnostic purposes. One exception to that rule is point-to-multipoint (PtMP) bridging. The nonroot bridges in a PtMP scenario will not be able to hear each other because they are miles apart. RTS/CTS should be implemented on nonroot PtMP bridges to eliminate collisions caused by hidden node bridges that cannot hear each other. The following methods can be used to fix a hidden node problem: Use RTS/CTS to diagnose. Use either a protocol analyzer or RTS/CTS to diagnose the hidden node problem. RTS/CTS can also be used as a temporary fix to the hidden node problem. Increase power to all stations. Most client stations have a fixed transmission power output. However, if power output is adjustable on the client side, increasing the transmission power of client stations will increase the transmission range of each station. If the transmission range of all stations is increased, the likelihood of the stations hearing each other also increases. This is not a recommended fix because, as you learned earlier, best practice dictates that client stations use the same transmit power used by all other radios in the BSS. Remove the obstacles. If it is determined that some sort of obstacle is preventing client stations from hearing each other, simply removing the obstacle will solve the problem. Obviously, you cannot remove a wall, but if a metal desk or file cabinet is the obstacle, it can be moved to resolve the problem. Move the hidden node station. If one or two stations are in an area where they become unheard, simply moving them within transmission range of the other stations will solve the problem. Add another access point. The best fix for a continuous hidden problem is to add another AP. If moving the hidden nodes is not an option, adding another access point in the hidden area to provide coverage will also rectify the problem.

802.11 Coverage Considerations
Providing for both coverage and capacity in a WLAN design solves many problems. Roaming problems and interference issues will often be mitigated in advance if proper WLAN design techniques are performed as well as a thorough site survey. In the following sections, we discuss many considerations that should be addressed to provide proper coverage, capacity, and performance within an 802.11 coverage zone.

Copyright © 2009 John Wiley & Sons, Inc.

802.11 Coverage Considerations

405

Dynamic Rate Switching
As client station radios move away from an access point, they will shift down to lowerbandwidth capabilities by using a process known as dynamic rate switching (DRS). Access points can support multiple data rates depending on the spread spectrum technology used by the AP’s radio card. For example, an HR-DSSS (802.11b) radio supports data rates of 11, 5.5, 2, and 1 Mbps. Data rate transmissions between the access point and the client stations will shift down or up depending on the quality of the signal between the two radio cards, as pictured in Figure 12.15. There is a correlation between signal quality and distance from the AP. As a result, transmissions between two 802.11b radio cards may be at 11 Mbps at 30 feet, but 2 Mbps at 150 feet. f I g u r e 1 2 .1 5 Dynamic rate switching

1 Mbps 2 Mbps 5.5 Mbps 11 Mbps

DRS is also referred to as dynamic rate shifting, adaptive rate selection, and automatic rate selection. All these terms refer to a method of speed fallback on a wireless LAN client as signal quality from the access point decreases. The objective of DRS is upshifting and downshifting for rate optimization and improved performance. Effectively, the lower data rates will have larger concentric zones of coverage than the higher data rates, as pictured in Figure 12.16.

Copyright © 2009 John Wiley & Sons, Inc.

406

Chapter 12

N

WLAN Troubleshooting

f I g u r e 1 2 .1 6

Data rate coverage zones

2 Mbps

2 Mbps

2 Mbps

2 Mbps

2 Mbps

5.5 Mbps

5.5 Mbps

5.5 Mbps

5.5 Mbps

5.5 Mbps

5.5 Mbps

5.5 Mbps

11 Mbps 11 Mbps 11 Mbps 11 Mbps 11 Mbps 11 Mbps 11 Mbps 11 Mbps 11 Mbps 11 Mbps 11 Mbps 11 Mbps

Coverage area

The algorithms used for dynamic rate switching are proprietary and are defined by radio card manufacturers. Most vendors base DRS on receive signal strength indicator (RSSI) thresholds, packet error rates, and retransmissions. RSSI metrics are usually based on signal strength and signal quality. In other words, a station might shift up or down between data rates based on both received signal strength in dBm and possibly on a signal-to-noise ratio (SNR) value. Because vendors implement DRS differently, you may have two different vendor client cards at the same location, while one is communicating with the access point at 5.5 Mbps and the other is communicating at 1 Mbps. For example, one vendor might shift down from data rate 11 Mbps to 5 Mbps at –70 dBm while another vendor might shift between the same two rates at –75 dBm. Keep in mind that DRS works with all 802.11 PHys. For example, the same shifting of rates will also occur with ERP-OFDM radios shifting between 54, 48, 36, 24, 18, 12, 9, and 6 Mbps data rates. As a result, there is a correlation between signal quality and distance from the AP. It is often a recommend practice to turn off the two lowest data rates of 1 and 2 Mbps when designing an 802.11b/g network. A WLAN network administrator might want to consider disabling the two lowest rates on an 802.11b/g access point for two reasons: medium contention and the hidden node problem. In Figure 12.17, you will see that multiple client stations are in the 1 Mbps zone, and only one lone client is in the 11 Mbps zone. Remember that wireless is a half-duplex medium and only one radio card can transmit on the medium at a time. All radio cards access the medium in a pseudorandom fashion as defined by CSMA/CA. A radio transmitting a 1,500-byte data frame at 11 Mbps might occupy the medium for 300 microseconds. Another radio transmitting at 1 Mbps per second may take 3,300 microseconds to deliver that same 1,500 bytes. Radio cards transmitting at slower data rates will

Copyright © 2009 John Wiley & Sons, Inc.

802.11 Coverage Considerations

407

occupy the medium much longer, while faster radios have to wait. If multiple radio cards get on the outer cell edges and transmit at slower rates consistently, the perceived throughput for the cards transmitting at higher rates is much slower because of having to wait for slower transmissions to finish. For this reason, too many radios on outer 1 and 2 Mbps cells can adversely affect throughput. Another reason to consider turning off the lower data rates is the hidden node problem, which was explained earlier in this chapter. f I g u r e 1 2 .17 Frame transmission time

1,500-byte frame 300 microseconds 11 Mbps 5.5 Mbps 2 Mbps 1 Mbps

Roaming
As you have learned throughout this book, roaming is the method by which client stations move between RF coverage cells in a seamless manner. Client stations switch communications through different access points. Seamless communications for stations moving between the coverage zones within an extended service set (ESS) is vital for uninterrupted mobility. One of the most common issues you’ll need to troubleshoot is problems with roaming. Roaming problems are usually caused by poor network design. Because of the proprietary nature of roaming, problems can also occur when radio cards from multiple vendors are deployed. Changes in the WLAN environment can also cause roaming hiccups.

1,

50

Copyright © 2009 John Wiley & Sons, Inc.

0-

by

te

fra

m

e-

3,

30

0

m

icr

os

ec

on

ds

408

Chapter 12

N

WLAN Troubleshooting

Client stations, and not the access point, make the decision on whether or not to roam between access points. Some vendors may involve the access point or WLAN controller in the roaming decision, but ultimately, the client station initiates the roaming process with a reassociation request frame. The method in which client stations decide how to roam is entirely proprietary. All vendor client stations use roaming algorithms that can be based on multiple variables. The variable of most importance will always be received signal strength: As the received signal from the original AP grows weaker and a station hears a stronger signal from another known access point, the station will initiate the roaming process. However, other variables such as SNR, error rates, and retransmissions may also have a part in the roaming decision. Because roaming is proprietary, a specific vendor client station may roam sooner than a second vendor client station as they move through various coverage cells. Some vendors like to encourage roaming, whereas others use algorithms that roam at lower received signal thresholds. In an environment where a WLAN administrator must support multiple vendor radios, different roaming behaviors will most assuredly be seen. For the time being, a WLAN administrator will always face unique challenges because of the proprietary nature of roaming. As discussed in Chapter 5, “IEEE 802.11 Standards,” the 802.11k amendment has defined the use of radio resource measurement (RRM) and neighbor reports to enhance roaming performance. The 802.11r amendment also defines faster handoffs when roaming occurs between cells in a wireless LAN using the strong security defined in a robust security network (RSN). The best way to ensure that seamless roaming will commence is proper design and a thorough site survey. When designing an 802.11 WLAN, most vendors recommend 15 to 25 percent overlap in coverage cells at the lowest desired signal level. The only way to determine whether proper cell overlap is in place is by conducting a coverage analysis site survey. Proper site survey procedures are discussed in detail in Chapter 16. Roaming problems will occur if there is not enough overlap in cell coverage. Too little overlap will effectively create a roaming dead zone, and connectivity may even temporarily be lost. On the flip side, too much cell overlap will also cause roaming problems. For example, if two cells have 60 percent overlap, a station may stay associated with its original AP and not connect to a second access point even though the station is directly underneath the second access point. This can also create a situation in which the client device is constantly switching back and forth between the two or more APs. This often presents itself when a client device is directly under an AP and there are constant dropped frames. Another design issue of great importance is latency. The 802.11-2007 standard defines the use of an 802.1X/EAP security solution in the enterprise. The average time involved during the authentication process can be 700 milliseconds or longer. Every time a client station roams to a new access point, reauthentication is required when an 802.1X/EAP security solution has been deployed. The time delay that is a result of the authentication process can cause serious interruptions with time-sensitive applications. VoWiFi requires a roaming handoff of 150 milliseconds or less when roaming. A fast secure roaming (FSR) solution is needed if 802.1X/EAP security and time-sensitive applications are used together in a wireless network. Currently, FSR solutions are proprietary, although the 802.11i amendment defined optional FSR. The fast secure roaming mechanisms defined by the ratified 802.11r amendment will hopefully standardize fast secure roaming in the enterprise.

Copyright © 2009 John Wiley & Sons, Inc.

802.11 Coverage Considerations

409

The 802.1X standard and Extensible Authentication Protocol (EAP) are discussed in detail in Chapter 13, “802.11 Network Security Architecture.” Included on the CD of this book is a white paper titled “Robust Security Network (RSN) Fast BSS Transition (FT)” by Devin Akin. The CWNA exam will not test you on the details of this paper. However, it discusses FSR solutions and is recommended extra reading for the CWSP exam.

Changes in the WLAN environment can also cause roaming headaches. RF interference will always affect the performance of a wireless network and can make roaming problematic as well. Very often new construction in a building will affect the coverage of a WLAN and create new dead zones. If the physical environment where the WLAN is deployed changes, the coverage design may have to change as well. It is always a good idea to periodically conduct a coverage survey to monitor changes in coverage patterns. Troubleshooting roaming by using a protocol analyzer is tricky because the reassociation roaming exchanges occur on multiple channels. To troubleshoot a client roaming between channels 1, 6, and 11, you would need three separate protocol analyzers on three separate laptops that would produce three separate frame captures. CACE Technologies offers a product called AirPcap that is a USB 802.11 radio. As pictured in Figure 12.18, three AirPcap USB radios can be configured to capture frames on channels 1, 6, and 11 simultaneously. All three radios are connected to a USB hub and save the frame captures of all three channels into a single time-stamped capture file. The AirPcap solution allows for multichannel monitoring with a single protocol analyzer. f I g u r e 1 2 .1 8 AirPcap provides multichannel monitoring and roaming analysis.

Layer 3 Roaming
One major consideration when designing a WLAN is what happens when client stations roam across layer 3 boundaries. As pictured in Figure 12.19, the client station is roaming between

Copyright © 2009 John Wiley & Sons, Inc.

410

Chapter 12

N

WLAN Troubleshooting

two access points. The roam is seamless at layer 2, but a router sits between the two access points, and each access point resides in a separate subnet. In other words, the client station will lose layer 3 connectivity and must acquire a new IP address. Any connection-oriented applications that are running when the client reestablishes layer 3 connectivity will have to be restarted. For example, a VoIP phone conversation would disconnect in this scenario, and the call would have to be reestablished. f I g u r e 1 2 .1 9 Layer 3 roaming boundaries
192.168.100.1 10.0.0.1

192.168.100.10

10.0.0.10

Client roams seamlessly at layer 2.

192.168.100.17

10.0.0.17 Client must obtain new IP address.

The preferred method when designing a WLAN is to have overlapping Wi-Fi cells that exist in only the same layer 3 domains through the use of VLANs. However, because 802.11 wireless networks are usually integrated into preexisting wired topologies, crossing layer 3 boundaries is often a necessity, especially in large deployments. The only way to maintain upper-layer communications when crossing layer 3 subnets is to provide either a Mobile IP solution or a proprietary layer 3 roaming solution. Mobile IP is an Internet Engineering Task Force (IETF) standard protocol that allows mobile device users to move from one layer 3 network to another while maintaining their original IP address. Mobile IP is defined in IETF request for comment (RFC) 3344. Mobile IP and proprietary solutions both use some type of tunneling method and IP header encapsulation to allow packets to traverse between separate layer 3 domains with the goal of maintaining upper-layer communications. It is beyond the scope of this book to explain either the standards-based Mobile IP or proprietary layer 3 roaming solutions; however, most WLAN controllers now support some type of layer 3 roaming solution. Although maintaining upper-layer connectivity is possible with these layer 3 roaming solutions, increased latency is often an issue. Additionally, layer 3 roaming may not be a requirement for your network. Even if there are layer 3 boundaries, your users may not need to seamlessly roam between subnets. Before you go to all the hassle of building a roaming solution, be sure to properly define your requirements.

Copyright © 2009 John Wiley & Sons, Inc.

802.11 Coverage Considerations

411

Co-channel Interference
One of the most common mistakes many businesses make when first deploying a WLAN is to configure multiple access points all on the same channel. If all of the APs are on the same channel, unnecessary medium contention overhead occurs. As you have learned, CSMA/CA dictates half-duplex communications, and only one radio can transmit on the same channel at any given time. As pictured in Figure 12.20, if an AP on channel 1 is transmitting, all nearby access points and clients on the same channel will defer transmissions. The result is that throughput is adversely affected: Nearby APs and clients have to wait much longer to transmit because they have to take their turn. The unnecessary medium contention overhead that occurs because all the APs are on the same channel is called co-channel interference (CCI). In reality, the 802.11 radios are operating exactly as defined by the CSMA/CA mechanisms, and this behavior should really be called co-channel cooperation. The unnecessary medium contention overhead caused by co-channel interference is a result of improper channel reuse design, which is discussed in the next section of this chapter. f I g u r e 12 . 2 0 Co-channel interference
The AP on channel 1 transmits.

2.4 GHz

Channel 1 All nearby APs and clients on channel 1 defer transmissions. Channel 1

Channel 1

Channel 1

Copyright © 2009 John Wiley & Sons, Inc.

412

Chapter 12

N

WLAN Troubleshooting

Please do not confuse adjacent cell interference with co-channel interference. However, adjacent cell interference is also a result of improper channel reuse design. As pictured in Figure 12.21, overlapping coverage cells that also have overlapping frequency space from adjacent cells will result in corrupted data and layer 2 retransmissions. Please refer back to Figure 12.6 and you will see that channels 1 and 4, channels 4 and 7, and channels 7 and 11 all have overlapping frequency space. Adjacent cell interference is a much more serious problem than co-channel interference because of the corrupted data and layer 2 retries. Proper channel reuse design is the answer to both co-channel and adjacent cell interference. f I g u r e 1 2 . 2 1 Adjacent cell interference: Access points with overlapping coverage cells transmit at the same time. The overlapping frequencies cause data corruption.

Channel 1

Channel 4

Channel 7

Channel 7

Channel 11

Channel Reuse/Multiple Channel Architecture
To avoid co-channel and adjacent cell interference, a channel reuse design is necessary. Once again, overlapping RF coverage cells are needed for roaming, but overlap frequencies must be avoided. The only three channels that meet these criteria in the 2.4 GHz ISM band are channels 1, 6, and 11 in the United States. Overlapping coverage cells therefore should be placed in a channel reuse pattern similar to the one pictured in Figure 12.22. A WLAN channel reuse pattern also goes by the name of multiple channel architecture (MCA). WLAN architecture with overlapping coverage cells that utilizes three channels at 2.4 GHz, or numerous channels at 5 GHz, would be considered a multiple channel architecture. It should be noted that it is impossible to avoid all instances of co-channel interference when using a three-channel reuse pattern at 2.4 GHz, because clients also cause co-channel interference. As pictured in Figure 12.23, if a client is at the outer edges of a coverage cell, the client’s transmissions may propagate into another cell using the same channel. All of the radios in the other cell will defer if they hear the original client’s transmissions. Channel reuse patterns should also be used in the 5 GHz UNII bands. If all UNII bands are legally available for transmissions, a total of 23 channels may be used in a channel

Copyright © 2009 John Wiley & Sons, Inc.

802.11 Coverage Considerations

413

reuse pattern at 5 GHz. It may not be necessary to use all 23 channels; however, the more channels, the better. f I g u r e 12 . 2 2 2.4 GHz multiple channel architecture

2.4 GHz

6 11 1 6 11 1 6 11

6 11 1 6 11 1 6 11

6

1

11

6

Figure 12.24 depicts a 5 GHz channel reuse pattern using the 12 channels available in UNII-1, UNII-2, and UNII-3. Although by the IEEE’s definition, all 5 GHz OFDM channels are considered nonoverlapping, in reality there is some frequency sideband overlap from adjacent channels. It is a recommend practice that any adjacent coverage cells use a frequency that is at least two channels apart and not use an adjacent frequency. Following this simple rule will prevent adjacent cell interference from the sideband overlap. As shown in Figure 12.24, the second recommended practice for 5 GHz channel reuse design dictates that there are always at least two cells of coverage space distance between any two access points transmitting on the same channel. Following this rule will prevent co-channel interference from APs and most likely also from clients. The client’s signal will have to propagate a greater distance and should attenuate to an amplitude below the noise floor before the signal reaches another coverage cell using the same channel. It is necessary to always think three-dimensionally when designing a multiple channel architecture reuse pattern. If access points are deployed on multiple floors in the same building, a reuse pattern will be necessary, such as the one pictured in Figure 12.25. A common mistake is to deploy a cookie-cutter design by performing a site survey on only one floor and then placing the access points on the same channels and same locations on each floor. A site survey must be performed on all floors, and the access points often need to be staggered to allow for a three-dimensional reuse pattern. Also, the coverage cells of each access point should not extend beyond more than one floor above and below the floor on which the access point is mounted. It is inappropriate to always assume that the coverage bleed over to other floors will provide sufficient signal strength and quality. In some cases, the floors are concrete or steel and allow very little, if any, signal coverage through. As a result, a survey is absolutely required.

Copyright © 2009 John Wiley & Sons, Inc.

414

Chapter 12

N

WLAN Troubleshooting

f I g u r e 12 . 2 3

Clients and co-channel interference

2.4 GHz

Channel 1 Client transmits on channel 1 on the outer edge of the coverage cell. Channel 11

Channel 6

Client’s signal propagates into another cell on channel 1. Radios that hear the client’s transmissions will defer.

Channel 1

f I g u r e 12 . 2 4

5 GHz multiple channel architecture

5 GHz

161 48 36 149 64 44 153

40 161 60 36 52 64 157

153

48

149

44

Distance to cell with same channel is at least two cells.

Copyright © 2009 John Wiley & Sons, Inc.

802.11 Coverage Considerations

415

Many enterprise access points currently have dual radio card capabilities, allowing for both 2.4 GHz and 5 GHz wireless networks to be deployed in the same area. The 802.11a radio in an access point transmits at 5 GHz, and the signal will attenuate faster than the signal that is being transmitted at 2.4 GHz from the 802.11b/g radio card. Therefore, when performing a site survey for deploying dual-frequency WLANs, it is a recommended practice to perform the 5 GHz site survey first and determine the placement of the access points. After those locations are identified, channel reuse patterns will have to be used for each respective frequency. f I g u r e 12 . 2 5 Three-dimensional channel reuse

Floor 5 1 6 11

Floor 4 11 1

Floor 3 1 6 11

Floor 2 11 1

Floor 1 1 6 11

Single Channel Architecture
At the time of writing this book, two vendors, Meru Networks and Extricom, are offering an alternative WLAN channel design solution known as the single channel architecture (SCA). Imagine a WLAN network with multiple access points all transmitting on the same channel

Copyright © 2009 John Wiley & Sons, Inc.

416

Chapter 12

N

WLAN Troubleshooting

and all sharing the same BSSID. A single channel architecture is exactly what you have just imagined! The client stations see transmissions on only a single channel with one SSID (logical WLAN identifier) and one BSSID (layer 2 identifier). From the perspective of the client station, only one access point exists. In this type of WLAN architecture, all access points in the network can be deployed on one channel in 2.4 GHz or 5 GHz frequency bands. Uplink and downlink transmissions are coordinated by a WLAN controller on a single 802.11 channel in such a manner that the effects of co-channel interference are minimized. Let us first discuss the single BSSID. Single channel architecture consists of a WLAN controller and multiple lightweight access points. As shown in Figure 12.26, each AP has its own radio card with its own MAC address; however, they all share a virtual BSSID that is broadcast from all of the access points. f I g u r e 12 . 2 6 Single channel architecture
SCA WLAN controller

Channel 6

Channel 6

Channel 6

Lightweight AP MAC: 00:12:17:09:84:B1 Virtual BSSID: 00:12:17:AA:BB:CC

Lightweight AP MAC: 00:12:17:09:83:22 Virtual BSSID: 00:12:17:AA:BB:CC

Lightweight AP MAC: 00:12:17:09:48:34 Virtual BSSID: 00:12:17:AA:BB:CC

The client sees only one “virtual” AP.

Because the multiple access points advertise only one single virtual MAC address (BSSID), client stations believe they are connected to only a single access point, although they may be roaming across multiple physical APs. you have learned that clients make the roaming

Copyright © 2009 John Wiley & Sons, Inc.

802.11 Coverage Considerations

417

decisions. In an single channel architecture (SCA) system, the clients think they are associated to only one AP, so they never initiate a layer 2 roaming exchange. All of the roaming handoffs are handled by a central WLAN controller. As pictured in Figure 12.27, the main advantage is that clients experience a zero handoff time, and the latency issues associated with roaming times are resolved. The virtual AP used by SCA solutions is potentially an excellent marriage for VoWiFi phones and 802.1X/EAP solutions. As we discussed earlier, the average time involved during the EAP authentication process can be 700 milliseconds or longer. Every time a client station roams to a new access point, reauthentication is required when an 802.1X/EAP security solution has been deployed. VoWiFi requires a roaming handoff of 150 ms or less. The virtual BSSID eliminates the need for reauthentication while physically roaming within a single channel architecture and thus a zero handoff time. f I g u r e 12 . 2 7 Zero handoff time
Channel 6 BSSID: 00:12:34:BB:CC:DD

Channel 6 BSSID: 00:12:34:BB:CC:DD

Zero handoff time

00:00

Channel 6

you have learned that client stations make the roaming decision in an MCA environment. However, client stations do not know that they roam in a SCA environment. The clients must still be mobile and transfer layer 2 communications between physical access points. All the client-roaming mechanisms are now handled back on the WLAN controller, and client-sideroaming decisions have been eliminated. All station associations are maintained at the SCA WLAN controller, and the SCA controller manages all the lightweight APs. The SCA controller assigns a unique lightweight access point the responsibility of handling downlink transmissions for an individual client station. When the controller receives the incoming transmissions of a client, the SCA controller evaluates the RSSI values of the client’s transmissions. Based on incoming RSSI measurements, the SCA controller can allocate a specific AP for downlink transmissions. The client believes that it is associated to a single AP. However, the client moves between different physical APs based on RSSI measurements evaluated by the controller.

Copyright © 2009 John Wiley & Sons, Inc.

418

Chapter 12

N

WLAN Troubleshooting

One big advantage of the single channel architecture is that adjacent cell interference is no longer an issue. If all the access points are on the same channel, there can be no adjacent cell interference, which is caused by frequency overlap. However, a legitimate question about a SCA WLAN solution is, Why doesn’t co-channel interference occur if all of the channels are on the same channel? If all of the APs are on the same channel in an MCA wireless network, unnecessary medium contention overhead occurs. In a typical MCA environment, each access point has a unique BSSID and a separate channel, and each AP’s coverage cell is a single-collision domain. In an SCA wireless environment, the collision domains are managed dynamically by the SCA controller based on RSSI algorithms. The controller ensures that nearby devices on the same channel are not transmitting at the same time. Most of the mechanisms used by SCA vendors to centrally manage co-channel interference is proprietary and beyond the scope of this book. SCA solutions also use processes to limit medium contention overhead created by client stations transmitting at very low data rates. As we discussed earlier in this chapter, client stations transmitting at slower data rates will occupy the medium much longer, while faster clients have to wait. Medium contention overhead will be significant if multiple radio cards get on the outer cell edges transmitting at 1 or 2 Mbps. That is why it is often a recommend practice to turn off the two lowest data rates of 1 and 2 Mbps when designing an 802.11b/g network. An SCA controller has airtime fairness (ATF) capabilities that use layer 1 and layer 4–7 mechanisms back on the WLAN controller to prioritize transmissions from stations with higher data rates over the stations using lower data rates. One component of ATF is to load-balance the clients between access points at higher data rates. A full explanation of airtime fairness processes are beyond the scope of this book. It should be noted that airtime fairness is not just used by SCA wireless LAN vendors. Several multiple channel architecture vendors have also begun to implement airtime fairness processes. The SCA controller vendors may also have the option to configure and deploy all of the lightweight APS in the more common MCA architecture by using a multiple channel reuse pattern and multiple BSSIDs. In the future, do not be surprised to see some of the MCA architecture vendors also offering single channel architecture capabilities.

Capacity vs. Coverage
When a wireless network is designed, two concepts that typically compete with each other are capacity and coverage. In the early days of wireless networks, it was common to install an access point with the power set to the maximum level to provide the largest coverage area possible. This was typically acceptable because there were very few wireless devices. Because the access points were also very expensive, companies tried to provide the most coverage while using the fewest access points. Figure 12.28 shows the outline of a building along with the coverage area that is provided by three APs in a multiple channel architecture. If there are just a few client stations, this type of wireless design is quite acceptable. With the proliferation of wireless devices, network design has changed drastically from the early days. Proper network design now entails providing necessary coverage while trying

Copyright © 2009 John Wiley & Sons, Inc.

Capacity vs. Coverage

419

to limit the number of devices connected to any single access point at the same time. This is what is meant by capacity vs. coverage. As you know, all of the client stations that connect to a single access point share the throughput capabilities of that access point. Therefore, it is important to design the network to try to limit the number of stations that are simultaneously connected to a single access point. This is performed by first determining the maximum number of stations that you want connected to an access point at the same time (this will vary from company to company depending on network usage). In an MCA environment, you need to determine how big the cell size needs to be to provide the proper capacity, and then you need to adjust the power level of the access point in order to create a cell of the desired size. f I g u r e 12 . 2 8 RF coverage of a building using three APs with few wireless stations

Ch 1

Ch 6

Ch 11

Figure 12.29 shows the outline of the same building, but because there are many more wireless stations, the cell sizes have been decreased while the number of cells has been increased. Adjusting the transmit power to limit the coverage area is known as cell sizing and is the most common method of meeting capacity needs in an MCA environment. f I g u r e 12 . 2 9 Cell sizing—multiple channel architecture

Ch 11

Ch 1

Ch 11

Ch 11

Ch 1

Ch 11

Ch 11

Ch 6

Ch 11

Ch 1

Ch 6

Ch 1

Ch 1

Ch 6

Ch 1

Ch 1

Ch 1

Ch 1

Ch 6

Ch 11

Ch 6

Ch 6

Ch 11

Ch 6

Ch 6

Ch 11

Ch 6

Another way of providing wireless support for a large capacity of users is by access point colocation. Colocation refers to placing multiple access points in the same physical space to provide for greater capacity.

Copyright © 2009 John Wiley & Sons, Inc.

420

Chapter 12

N

WLAN Troubleshooting

802.11b and 802.11g APs are capable of having only three access points in the same area without causing interference. The three APs would need at least a five-channel separation to prevent RF interference. By colocating three APs, theoretically the potential cumulative speed is three times the speed of a single AP (assuming the three APs are equal). For example, three colocated 802.11g APs would provide a cumulative maximum speed of 162 Mbps (remember that actual throughput will be significantly less). When colocating APs, it is important to make sure that the APs are physically separated from each other by at least 15 feet (5 meters) to prevent possible interference by the sideband signals that the APs generate. In an MCA environment, access point colocation is recommended only when the concentration of users is so dense that even when the cell size is at its smallest, there are still more stations per cell than desired. This often will occur in large meeting halls or university lecture halls. When colocated, end users can be load-balanced and segmented by MAC filters or by separate SSIDs. Cell sizing is almost always the preferable method for meeting capacity needs in an MCA environment. Colocation can provide for capacity with MCA wireless LANs in a single predefined area. However, colocation does not scale in a WLAN that uses multiple channel architecture. Colocation does, however, scale within a single channel architecture design. As pictured in Figure 12.30, in a 2.4 GHz SCA deployment, multiple APs can be colocated by using three channels and three virtual BSSIDs. Colocation design in a single channel architecture is often referred to as channel stacking. Each layer of multiple APs on a single channel and using the same virtual BSSID is known as a channel blanket or channel span. f I g u r e 12 . 3 0 2.4 GHz channel stacking—single channel architecture
Channel 1 Virtual BSSID 00:11:22:AA:BB:A1

Channel 6 Virtual BSSID 00:11:22:AA:BB:B2

Channel 11 Virtual BSSID 00:11:22:AA:BB:C3

In Chapter 18, we discuss the 40 MHz channels that can be used by 802.11n. you will learn that using 40 MHz channels at 2.4 GHz is not possible in an MCA design due to adjacent cell interference caused by the frequency overlap of the 40 MHz channels. However, a single 40 MHz channel blanket could be deployed at 2.4 GHz with a single channel architecture. 40 MHz 802.11n channels do not scale in the 2.4 GHz ISM band within an MCA design, but a single 40 MHz channel blanket can scale at 2.4 GHz with a SCA design.

Copyright © 2009 John Wiley & Sons, Inc.

Voice vs. Data

421

Oversized Coverage Cells
A mistake often made when deploying access points is to have the APs transmitting at full power. Effectively, this extends the range of the access point but causes many problems that have been discussed throughout this chapter. Oversized coverage usually will not meet your capacity needs. Oversized coverage cells can cause hidden node problems. Access points at full power may not be able to hear the transmissions of client stations with lower transmit power. Access points at full power will most likely also increase the odds of co-channel interference due to bleed-over transmissions. If the access point’s coverage and range is a concern, the best method of extending range is to increase the AP’s antenna gain instead of increasing transmit power.

Physical Environment
Although physical environment does not cause RF interference, physical obstructions can indeed disrupt and corrupt an 802.11 signal. An example of this is the scattering effect caused by a chain-link fence or safety glass with wire mesh. The signal is scattered and rendered useless. The only way to eliminate physical interference is to remove the obstruction or add more APs. Keep in mind that the physical environment of every building and floor is different, and the shape and size of coverage cells will widely vary. The best method of dealing with the physical environment is to perform a proper site survey as described in detail in Chapter 16.

Voice vs. Data
As you have already learned, most data applications in a Wi-Fi network can handle a layer 2 retransmission rate of up to 10 percent without any noticeable degradation in performance. However, time-sensitive applications such as VoIP require that higher-layer IP packet loss be no greater than 2 percent. Therefore, Voice over Wi-Fi (VoWiFi) networks need to limit layer 2 retransmissions to 5 percent or less to guarantee the timely and consistent delivery of VoIP packets. When layer 2 retransmissions exceed 5 percent, latency problems may develop and jitter problems will most likely surface. The canary-and-coal-mine analogy is often used to describe the difference between voice traffic and other data application traffic within a WLAN environment. Early coal mines did not have ventilation systems installed in them, so miners would bring a caged canary into new coal shafts. Canaries are more sensitive to methane and carbon monoxide than humans, which made them ideal for detecting dangerous gas buildups. As long as the canary (voice traffic) was singing, the miners knew their air supply was safe. A dead canary signaled the presence of deadly gases, and the miners would evacuate or put on respirator masks. However, some species, such as a cockroach (data traffic) can still survive within the coal mine despite the existence of the deadly gases.

Copyright © 2009 John Wiley & Sons, Inc.

422

Chapter 12

N

WLAN Troubleshooting

All too often, WLANs are deployed in the enterprise without any type of site survey. Also, many WLANs are initially designed to provide coverage only for data applications and not for voice. Most enterprise data applications will operate within a poorly designed WLAN, but are not running optimally due to the lack of a site survey or improper survey. Many companies decide to add a VoWiFi solution to their WLAN at a later date and quickly discover that the WLAN has many problems. The VoWiFi phones may have choppy audio or echo problems. The VoWiFi phones may disconnect or “die” like a canary. Adding voice to the WLAN often exposes existing problems: Voice is the canary, and the WLAN is the coal mine. Because data applications can withstand a much higher layer 2 retransmission rate, problems that existed within the WLAN may have gone unnoticed. The data applications are analogous to the cockroach that can still live in the coal mine but probably would have had a better life if the poor conditions did not exist. As shown in Table 12.1, IP voice traffic is more susceptible to late or inconsistent packet delivery due to layer 2 retransmissions.
TA b L e 1 2 .1 IP Voice Small, uniform-size packets Even, predictable delivery Highly affected by late or inconsistent packet delivery “Better never than late” IP Voice and IP Data Comparison IP Data Variable-size packets Bursty delivery Minimally affected by late or inconsistent packet delivery “Better late than never”

Optimizing the WLAN to support voice traffic will optimize the network for all wireless clients, including the clients running data applications other than voice. A proper site survey will reduce lower layer 2 retransmissions and provide an environment with seamless coverage that is required for VoWiFi networks.

Performance
When designing and deploying a WLAN, you will always be concerned about both coverage and capacity. Various factors can affect the coverage range of a wireless cell, and just as many factors can affect the aggregate throughput in an 802.11 WLAN. The following variables can affect the range of a WLAN: Transmission power rates The original transmission amplitude (power) will have an impact on the range of an RF cell. An access point transmitting at 30 mW will have a larger coverage zone than an access point transmitting at 1 mW, assuming that the same antenna

Copyright © 2009 John Wiley & Sons, Inc.

Performance

423

is used. APs with too much transmission amplitude can cause many problems already discussed in this chapter. Antenna gain Antennas are passive-gain devices that focus the original signal. An access point transmitting at 30 mW with a 6 dBi antenna will have greater range than it would if it used only a 3 dBi antenna. If you want to increase the range for the clients, the best solution is to increase the antenna gain of the access point. Antenna type Antennas have different coverage patterns. Using the right antenna will give the proper coverage and reduce multipath and nearby interference. Wavelength Higher frequency signals have a smaller wavelength property and will attenuate faster than a lower-frequency signal with a larger wavelength. All things being equal, 2.4 GHz access points have a greater range than 5 GHz access points, due to the difference in the length of their waves. Free space path loss In any RF environment, free space path loss (FSPL) attenuates the signal as a function of distance and frequency. Physical environment Walls and other obstacles will attenuate an RF signal because of absorption and other RF propagation behaviors. A building with concrete walls will require more access points than a building with drywall because concrete is denser and attenuates the signal faster than drywall. As you have learned in earlier chapters, proper WLAN design must take into account both coverage and capacity. The variables just mentioned all affect coverage and range. Capacity performance considerations are equally as important as range considerations. Please remember that 802.11 data rates are considered data bandwidth and not throughput. The following are among the many variables that can affect the throughput of a WLAN: Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) The medium access method that uses interframe spacing, physical carrier-sense, virtual carrier-sense, and the random back-off timer creates overhead and consumes bandwidth. The overhead due to medium contention usually is 50 percent or greater. Encryption Extra overhead is added to the body of an 802.11 data frame whenever encryption is implemented. WEP/RC4 encryption adds an extra 8 bytes of overhead per frame, TKIP/RC4 encryption adds an extra 20 bytes of overhead per frame, and CCMP/AES encryption adds an extra 16 bytes of overhead per frame. Layer 3 VPNs often use DES or 3DES encryption, both of which also consume significant bandwidth. Application use Different types of applications will have variable affects on bandwidth consumption. VoWiFi and data collection scanning typically do not require a lot of bandwidth. Other applications that require file transfers or database access are often more bandwidth intensive. Number of clients Remember that the WLAN is a shared medium. All throughput is aggregate, and all available bandwidth is shared.

Copyright © 2009 John Wiley & Sons, Inc.

424

Chapter 12

N

WLAN Troubleshooting

Layer 2 retransmissions As we have discussed throughout this chapter, various problems can cause frames to become corrupted. If frames are corrupted, they will need to be retransmitted and throughput will be affected.

Weather
When deploying a wireless mesh network outdoors or perhaps an outdoor bridge link, a WLAN administrator must take into account the adverse affect of weather conditions. The following weather conditions must be considered: Lightning Direct and indirect lightning strikes can damage WLAN equipment. Lightning arrestors should be used for protection against transient currents. Solutions such as lightning rods or copper/fiber transceivers may offer protection against lightning strikes. Wind Because of the long distances and narrow beamwidths, highly directional antennas are susceptible to movement or shifting caused by wind. Even slight movement of a highly directional antenna can cause the RF beam to be aimed away from the receiving antenna, interrupting the communications. In high-wind environments, a grid antenna will typically remain more stable than a parabolic dish. Other mounting options may be necessary to stabilize the antenna from movement. Water Conditions such as rain, snow, and fog present two unique challenges. First, all outdoor equipment must be protected from damage caused by exposure to water. Water damage is often a serious problem with cabling and connectors. Connectors should be protected with drip loops and coax seals to prevent water damage. Cables and connectors should be checked on a regular basis for damage. A radome (weatherproof protective cover) should be used to protect antennas from water damage. Outdoor bridges, access points, and mesh routers should be protected from the weather elements by using appropriate National Electrical Manufacturers Association (NEMA) enclosure units. Precipitation can also cause an RF signal to attenuate. A torrential downpour can attenuate a signal as much as 0.08 dB per mile (0.05 dB per kilometer) in both the 2.4 GHz and 5 GHz frequency ranges. Over long-distance bridge links, a system operating margin (SOM) of 20 dB is usually recommended to compensate for attenuation due to rain, fog, or snow. Air stratification A change in air temperature at high altitudes is known as air stratification (layering). Changes in air temperature can cause refraction. Bending of RF signals over longdistance point-to-point links can cause misalignment and performance issues. K-factor calculations may be necessary to compensate for refraction over long-distance links. UV/sun UV rays and ambient heat from rooftops can damage cables over time if proper cable types are not used.

Copyright © 2009 John Wiley & Sons, Inc.

Exam Essentials

425

Summary
In this chapter, we discussed numerous 802.11 coverage considerations. Troubleshooting for coverage, capacity, and performance problems can quite often be avoided with proper network design and comprehensive site surveys. We discussed the many causes of layer 2 retransmissions and the negative effects on the WLAN because of retries. Because wireless should always be considered an ever-changing environment, problems such as roaming, hidden nodes, and interference are bound to surface. Tools such as protocol analyzers and spectrum analyzers are invaluable when troubleshooting both layer 2 and layer 1 problems. We discussed and compared the differences between multiple channel architecture and single channel architecture. We also discussed the many performance variables that can affect both range and throughput. We discussed the challenges that are unique to both voice and data WLAN deployments. Finally, we discussed weather conditions that can impact outdoor RF communications and the steps that might be necessary for protection against Mother Nature.

Exam Essentials
Explain the causes and effect of Layer 2 retransmissions. Understand that layer 2 retransmissions can be caused by multipath, hidden nodes, mismatched power settings, RF interference, low SNR, near/far problems, and adjacent cell interference. Layer 2 retransmissions affect throughput, latency, and jitter. Define dynamic rate switching. Understand the process of stations shifting between data rates. Know that dynamic rate switching is also referred to as dynamic rate shifting, adaptive rate selection, and automatic rate selection. Explain why disabling the two lower 802.11b/g data rates is often recommended. Explain the various aspects of roaming. Understand that roaming is proprietary in nature. Know the variables that client stations may use when initiating the roaming process. Understand the importance of proper coverage cell overlap. Describe latency issues that can occur with roaming. Understand why crossing layer 3 boundaries can cause problems and what solutions might exist. Define the differences between adjacent channel interference and co-channel interference. Understand the negative effects of both adjacent cell interference and co-channel interference. Explain why channel reuse patterns minimize the problems. Know what to consider when designing channel reuse patterns at both 2.4 GHz and 5 GHz in a multiple channel architecture. Explain the differences between MCA and SCA wireless LAN design. Understand that MCA uses cell sizing to meet capacity needs, whereas SCA uses channel stacking to meet capacity needs. Explain the virtual BSSID and other aspects of an SCA design.

Copyright © 2009 John Wiley & Sons, Inc.

426

Chapter 12

N

WLAN Troubleshooting

Identify the various types of interference. Know the differences between all-band, narrowband, wideband, physical, and intersymbol interference. Understand that a spectrum analyzer is your best interference-troubleshooting tool. Explain the hidden node problem. Identify all the potential causes of the hidden node problem. Explain how to troubleshoot hidden nodes as well as how to fix the hidden node problem. Define the near/far problem. Explain what causes near/far and how the problem can be rectified. Identify performance variables. Explain all the variables that affect both the range of RF coverage and the throughput that can result within a basic service set. Understand the consequences of weather conditions. Explain the problems that might arise due to water conditions, wind, lightning, and air stratification. Explain how these problems might be solved.

Copyright © 2009 John Wiley & Sons, Inc.

Key Terms

427

Key Terms
Before you take the exam, be certain you are familiar with the following terms: adjacent cell interference airtime fairness (ATF) all-band interference antenna reciprocity Bluetooth (BT) capacity cell sizing channel blanket channel reuse channel span co-channel interference (CCI) colocation coverage cyclic redundancy check (CRC) delay spread dynamic rate switching (DRS) fast secure roaming (FSR) hidden node inter-symbol interference (ISI) layer 3 roaming Mobile IP multipath multiple channel architecture (MCA) multiple-input multiple-output (MIMO) near/far radio resource measurement (RRM) range roaming signal-to-noise ratio (SNR) single channel architecture (SCA). virtual BSSID

Copyright © 2009 John Wiley & Sons, Inc.

428

Chapter 12

N

WLAN Troubleshooting

Review Questions
1. What type of solution must be deployed to provide continuous connectivity when a client station roams across layer 3 boundaries? (Choose all that apply.) A. Nomadic roaming solution B. C. D. E. 2. Proprietary layer 3 roaming solution Seamless roaming solution Mobile IP solution Fast secure roaming solution

If the access points transmit on the same frequency channel in an MCA architecture, what type of interference is caused by overlapping coverage cells? A. Intersymbol interference B. C. D. E. Adjacent cell interference All-band interference Narrowband interference Co-channel interference

3.

What variables might affect range in an 802.11 WLAN? (Choose all that apply.) A. Transmission power B. C. D. E. CSMA/CA Encryption Antenna gain Physical environment

4.

What can be done to fix the hidden node problem? (Choose all that apply.) A. Increase the power on the access point. B. C. D. E. Move the hidden node station. Increase power on all client stations. Remove the obstacle. Decrease power on the hidden node station.

5.

Layer 2 retransmissions occur when frames become corrupted. What are some of the causes of layer 2 retries? (Choose all that apply.) A. Multipath B. C. D. E. Low SNR Co-channel interference RF interference Adjacent cell interference

Copyright © 2009 John Wiley & Sons, Inc.

Review Questions

429

6.

What scenarios might result in a hidden node problem? (Choose all that apply.) A. Distributed antenna system B. C. D. E. Too large coverage cell Too small coverage cell Physical obstruction Co-channel interference

7.

What are some of the negative effects of layer 2 retransmissions? (Choose all that apply.) A. Decreased range B. C. D. E. Excessive MAC sublayer overhead Decreased latency Increased latency Jitter

8.

Several users are complaining that their VoWiFi phones keep losing connectivity. The WLAN administrator notices that the frame transmissions of the VoWiFi phones are corrupted when listened to with a protocol analyzer near the access point, but are not corrupted when listened to with the protocol analyzer near the VoWiFi phone. What is the most likely cause of this problem? A. RF interference B. C. D. E. Multipath Hidden node Adjacent cell interference Mismatched power settings

9.

A single user is complaining that her VoWiFi phone has choppy audio. The WLAN administrator notices that the user’s MAC address has a retry rate of 25 percent when observed with a protocol analyzer. However, all the other users have a retry rate of about 5 percent when also observed with the protocol analyzer. What is the most likely cause of this problem? A. Near/far B. C. D. E. Multipath Co-channel interference Hidden node Low SNR

10. What type of interference is caused by overlapping cover cells with overlapping frequencies? A. Inter-symbol interference B. C. D. E. Adjacent cell interference All-band interference Narrowband interference Co-channel interference

Copyright © 2009 John Wiley & Sons, Inc.

430

Chapter 12

N

WLAN Troubleshooting

11. Based on RSSI metrics, concentric zones of variable data rate coverage exist around an access point due to the upshifting and downshifting of client stations between data rates. What is the correct name of this process, according to the IEEE 802.11-2007 standard? (Choose all that apply.) A. Dynamic rate shifting B. C. D. E. Dynamic rate switching Automatic rate selection Adaptive rate selection All of the above

12. Which of these weather conditions is a concern when deploying a long-distance point-topoint bridge link? (Choose all that apply.) A. Wind B. C. D. E. Rain Fog Changes in air temperature All of the above

13. What variables might affect range in an 802.11 WLAN? (Choose all that apply.) A. Wavelength B. C. D. E. Free space path loss Brick walls Trees All of the above

14. Which WLAN architecture can use the 40 MHz OFDM channel capabilities of an 802.11n access point in the 2.4 GHz ISM band? A. Multiple channel architecture B. C. D. E. Single channel architecture Distributed WLAN architecture Unified architecture None of the above

15. Which of the following can cause roaming problems? (Choose all that apply.) A. Too little cell coverage overlap B. C. D. E. Too much cell coverage overlap Free space path loss CSMA/CA Hidden node

Copyright © 2009 John Wiley & Sons, Inc.

Review Questions

431

16. What are some problems that can occur when an access point is transmitting at full power? (Choose all that apply.) A. Hidden node B. C. D. E. Adjacent cell interference Co-channel interference Mismatched power between the AP and the clients Intersymbol interference

17. Why would a WLAN network administrator consider disabling the two lowest rates on an 802.11b/g access point? (Choose all that apply.) A. Medium contention B. C. D. E. Adjacent cell interference Hidden node Co-channel interference All of the above

18. Which type of interference is caused by multipath? A. Intersymbol interference B. C. D. E. All-band interference Narrowband interference Wideband interference Physical interference

19. In a multiple channel architecture (MCA) design, what is the greatest number of nonoverlapping channels that can be deployed in the 2.4 GHz ISM band? A. 3 B. C. D. E. 12 11 14 4

20. Colocating access points in the same physical space is one method of meeting capacity needs. Colocation scales best in the enterprise using which WLAN architecture? A. Multiple channel architecture B. C. D. E. Single channel architecture Distributed WLAN architecture Unified architecture None of the above

Copyright © 2009 John Wiley & Sons, Inc.

432

Chapter 12

N

WLAN Troubleshooting

Answers to Review Questions
1. 2. B, D. The only way to maintain upper-layer communications when crossing layer 3 subnets is to provide either a Mobile IP solution or a proprietary layer 3 roaming solution. E. In an MCA architecture, if all the access points are mistakenly configured on the same channel, unnecessary medium contention overhead is the result. If an AP is transmitting, all nearby access points and clients on the same channel will defer transmissions. The result is that throughput is adversely affected: Nearby APs and clients have to wait much longer to transmit because they have to take their turn. The unnecessary medium contention overhead that occurs because all the APs are on the same channel is called co-channel interference (CCI). In reality, the 802.11 radios are operating exactly as defined by the CSMA/CA mechanisms, and this behavior should really be called co-channel cooperation. A, D, E. The original transmission amplitude will have an impact on the range of an RF cell. Antennas amplify signal strength and can increase range. Walls and other obstacles will attenuate an RF signal and affect range. CSMA/CA and encryption do not affect range but do affect throughput. B, C, D. The hidden node problem arises when client stations cannot hear the RF transmissions of another client station. Increasing the transmission power of client stations will increase the transmission range of each station, resulting in increased likelihood of all the stations hearing each other. Increasing client power is not a recommended fix because best practice dictates that client stations use the same transmit power used by all other radios in the BSS, including the AP. Moving the hidden node station within transmission range of the other stations also results in stations hearing each other. Removing an obstacle that prevents stations from hearing each other also fixes the problem. The best fix to the hidden node problem is to add another access point in the area that the hidden node resides. A, B, D, E. If any portion of a unicast frame is corrupted, the cyclic redundancy check (CRC) will fail and the receiving 802.11 radio will not return an ACK frame to the transmitting 802.11 radio. If an ACK frame is not received by the original transmitting radio, the unicast frame is not acknowledged and will have to be retransmitted. Multipath, RF interference, low SNR, hidden nodes, mismatched power settings, near/far problems, and adjacent cell interference may all cause layer 2 retransmissions. Co-channel interference does not cause retries but does add unnecessary medium contention overhead. A, B, D. The hidden node problem arises when client stations cannot hear the RF transmissions of another client station. Distributed antenna systems with multiple antenna elements are notorious for causing the hidden node problem. When coverage cells are too large as a result of the access point’s radio transmitting at too much power, client stations at opposite ends of an RF coverage cell often cannot hear each other. Obstructions such as a newly constructed wall can also result in stations not hearing each other. B, D, E. Excessive layer 2 retransmissions adversely affect the WLAN in two ways. First, layer 2 retransmissions increase MAC overhead and therefore decrease throughput. Second, if application data has to be retransmitted at layer 2, the timely delivery of application

3.

4.

5.

6.

7.

Copyright © 2009 John Wiley & Sons, Inc.

Answers to Review Questions

433

traffic becomes delayed or inconsistent. Applications such as VoIP depend on the timely and consistent delivery of the IP packet. Excessive layer 2 retransmissions usually result in increased latency and jitter problems for time-sensitive applications such as voice and video. 8. E. An often overlooked cause of layer 2 retransmissions is mismatched transmit power settings between an access point and a client radio. Communications can break down if a client station’s transmit power level is less than the transmit power level of the access point. As a client moves to the outer edges of the coverage cell, the client can “hear” the AP; however, the AP cannot “hear” the client. If the client station’s frames are corrupted near the AP but not near the client, the most likely cause is mismatched power settings. D. If an end user complains of a degradation of throughput, one possible cause is a hidden node. A protocol analyzer is a useful tool in determining hidden node issues. If the protocol analyzer indicates a higher retransmission rate for the MAC address of one station when compared to the other client stations, chances are a hidden node has been found. Some protocol analyzers even have hidden node alarms based on retransmission thresholds. Another way is to use request to send/clear to send (RTS/CTS) to diagnose the problem.

9.

10. B. Overlapping coverage cells with overlapping frequencies cause adjacent cell interference, which causes a severe degradation in latency, jitter, and throughput. If overlapping coverage cells also have frequency overlap, frames will become corrupt, retransmissions will increase, and performance will suffer significantly. 11. B. As client station radios move away from an access point, they will shift down to lower bandwidth capabilities by using a process known as dynamic rate switching (DRS). The objective of DRS is upshifting and downshifting for rate optimization and improved performance. Although dynamic rate switching is the proper name for this process, all these terms refer to the method of speed fallback that a wireless LAN client uses as distance increases from the access point. 12. E. Highly directional antennas are susceptible to what is known as antenna wind loading, which is antenna movement or shifting caused by wind. Grid antennas may be needed to alleviate the problem. Rain and fog can attenuate an RF signal; therefore, a system operating margin (also known as fade margin) of 20 dB is necessary. A change in air temperature is also known as air stratification, which causes refraction. K-factor calculations may also be necessary to compensate for refraction. 13. E. Higher-frequency signals have a smaller wavelength property and will attenuate faster than a lower-frequency signal with a larger wavelength. Higher-frequency signals therefore will have shorter range. In any RF environment, free space path loss (FSPL) attenuates the signal as a function of distance. Loss in signal strength affects range. Brick walls exist in an indoor physical environment, while trees exist in an outdoor physical environment. Both will attenuate an RF signal, thereby affecting range. 14. B. The 802.11n draft amendment defines the use of 40 MHz OFDM channels. Using 40 MHz channels at 2.4 GHz is not possible in an MCA design due to adjacent cell interference caused by the frequency overlap of the 40 MHz channels. However, a single 40 MHz channel blanket could be deployed at 2.4 GHz with a single channel architecture. 40 MHz 802.11n channels do not scale in the 2.4 GHz ISM band within an MCA design, but a single 40 MHz channel blanket can scale at 2.4 GHz with a SCA design.

Copyright © 2009 John Wiley & Sons, Inc.

434

Chapter 12

N

WLAN Troubleshooting

15. A, B. Roaming problems will occur if there is not enough overlap in cell coverage. Too little overlap will effectively create a roaming dead zone, and connectivity may even temporarily be lost. If two RF cells have too much overlap, a station may stay associated with its original AP and not connect to a second access point even though the station is directly underneath the second access point. 16. A, C, D. A mistake often made when deploying access points is to have the APs transmitting at full power. Effectively, this extends the range of the access point but causes many problems that have been discussed throughout this chapter. Oversized coverage usually will not meet your capacity needs. Oversized coverage cells can cause hidden node problems. Access points at full power may not be able to hear the transmissions of client stations with lower transmit power. Access points at full power will most likely also increase the odds of co-channel interference due to bleed-over transmissions. If the access point’s coverage and range is a concern, the best method of extending range is to increase the AP’s antenna gain instead of increasing transmit power. 17. A, C. Medium contention, also known as CSMA/CA, requires that all radios access the medium in a pseudorandom fashion. Radio cards transmitting at slower data rates will occupy the medium much longer, while faster radios have to wait. Data rates of 1 and 2 MBPS can create very large coverage cells, which may prevent a hidden node station at one edge of the cell from being heard by other client stations at the opposite side of the coverage cell. 18. A. Multipath can cause intersymbol interference (ISI), which causes data corruption. Because of the difference in time between the primary signal and the reflected signals, known as the delay spread, the receiver can have problems demodulating the RF signal’s information. The delay spread time differential results in corrupted data and therefore layer 2 retransmissions. 19. A. HR-DSSS (802.11b) and ERP (802.11g) channels require 25 MHz of separation between the center frequencies to be considered nonoverlapping. The three channels of 1, 6, and 11 meet these requirements in the United States. In other countries, 2, 7, and 12; and 3, 8, and 13; and 4, 9, and 14 would work as well. Traditionally, 1, 6, and 11 are chosen almost universally. 20. B. Cell sizing is almost always the preferable method for meeting capacity needs in an MCA environment. Colocation can provide for capacity with MCA wireless LANs in a single predefined area. However, colocation does not scale in a WLAN that uses multiple channel architecture. Colocation does, however, scale within a single channel architecture design. Colocation design in a single channel architecture is often referred to as channel stacking. Each layer of multiple APs on a single channel and using the same virtual BSSID is known as a channel blanket or channel span.

Copyright © 2009 John Wiley & Sons, Inc.