Radio Signal Interference
Radio signal interference involves the presence of unwanted, interfering radio signals that disrupt normal WLAN operations. Because of the 802.11 Medium Access Control (MAC) protocols, an interfering radio signal of sufficient amplitude and frequency can appear as a bogus 802.11 station transmitting a packet. This causes legitimate 802.11 stations to wait for indefinite periods of time before attempting to access the medium until the interfering signal goes away.
To make matters worse, radio signal interference does not abide by the 802.11 protocols, so the interfering signal might start abruptly while a legitimate 802.11 station is in the process of transmitting a packet, as illustrated in Figure 4-6. If this occurs, the destination station will receive the packet with errors and not reply to the source station with an acknowledgement. In return, the source station will attempt to retransmit the packet, adding overhead on the network.
Radio signal interference causes wireless clients and access points to hold off transmitting, which causes delay and reduced throughput. As a result, interference lasting for longer periods of time (referred to as duty cycle) will cause more damage to the signal, and interference present for shorter periods of time may have less impact on the signal. The resulting decrease in performance caused by interference can make browsing websites and downloading files sluggish and severely limit the number of active voice users. In cases where interfering signals are strong enough, the wireless clients might not be able to access the WLAN at all for an indefinite period of time. This is rare but possible. As a result, companies need to be aware of potential sources of radio signal interference, such as microwave ovens and cordless phones, operating within the WLAN environment.
Of course this all leads to network latency and unhappy users. In some cases, 802.11 protocols will attempt to continue operation in the presence of radio signal interference by automatically switching to a lower data rate, which also slows the use of wireless applications. The worst case, which is fairly uncommon, is that the 802.11 stations will hold off until the interfering signal goes away completely, which could be minutes, hours, or days.
Figure 4-6 Radio Signal Interference Can Damage 802.11 Frames En Route Between Client Radios and Access Points
The sections that follow explore the various sources of radio signal interference.
Microwave Oven Interference
Microwave ovens emit damaging interfering signals at up to 25 feet or so from an operating oven. Some microwave ovens emit radio signals that occupy only a third of the 2.4-GHz band, whereas others occupy the entire band. This means that it might be possible to tune an access point near a microwave oven to a noninterfering channel. Keep in mind that the microwave oven’s interfering signal is present only while the oven is operating, which might not be often. For example, a microwave oven in a break room might be in use only a few times during lunch and sit idle during the rest of the day. Thus, microwave ovens present within the facility might not significantly impact the use of the WLAN.
Figure 4-7 shows the output of a Bantam spectrum analyzer depicting the presence of microwave oven interference at ranges of 10 and 25 feet. These measurements were taken while the microwave oven was cooking a bag of popcorn, which takes about 2 minutes. As you can see, the interfering signals occupy the majority of the 2.4-GHz band. What is seen in the image is actually a quasi-peak value, which is the peak signal value with a discharge that takes a few seconds. In this case, it is not possible to completely avoid the interference by changing the access point channels. Each microwave oven has a different interference signature. In fact, we have tested other microwave ovens that have had more interference in the upper third of the 2.4-GHz band, which makes it possible to avoid the interference by tuning the access points near the microwave oven to the lower RF channels.
Figure 4-7 Radio Frequency Interference Originating from a Microwave Oven
Cordless Phone Interference
Cordless phones cause interference with WLANs. Today, a person can purchase cordless phones that operate in a variety of unlicensed frequency bands: 900 MHz, 2.4 GHz, and 5 GHz. A cordless phone will cause interference only with WLANs operating in the same frequency bands as the phone. For example, a 2.4-GHz cordless phone can interfere with WLANs operating in the 2.4-GHz band but not the 5-GHz band.
Many of these phones use direct-sequence spread spectrum (DSSS) and automatically choose the least congested channel. As a result, they can tune around WLAN access points. If there are several phones actively in use at the same time, however, the phones might monopolize the entire spectrum and cause interference with the WLAN access points. Figure 4-8 shows the output of a Bantam spectrum analyzer depicting the presence of interference approximately 2 feet from a DSSS cordless phone. Some cordless phones use frequency-hopping spread spectrum (FHSS), which distributes radio signals over the entire band. As a result, they nearly always cause significant interference with WLANs operating in the same frequency band as the phone.
The use of 2.4-GHz cordless phones is very common. In this case, consider implementing 802.11ac or the 5-GHz version of 802.11n to avoid interference. As a last resort, the use of cordless phones should be prohibited in cases where it is critical to maximize the capacity of a WLAN, such as when maximizing the number of active voice users.
Figure 4-8 Radio Frequency Interference Originating from a Cordless Phone
Similar to 802.11 WLANs, Bluetooth devices operate within the 2.4-GHz band. The difference is that Bluetooth uses FHSS (at 1,600 hops per second) to hop over the entire 2.4-GHz band. 802.11, on the other hand, occupies only approximately one third of the 2.4-GHz band. As a result, Bluetooth hops all over 802.11 transmissions. Figure 4-9 shows the output of a Bantam spectrum analyzer depicting the presence of interference approximately 2 feet from a Bluetooth headset. Adaptive frequency hopping, introduced by the Bluetooth Special Interest Group (SIG) and part of most Bluetooth devices, helps reduce the impact of Bluetooth interference, but it is still wise to be aware of the impacts and implement other anti-interference methods.
Figure 4-9 Radio Frequency Interference Originating from a Bluetooth Headset
Because of the potential for collisions, 802.11 (and Bluetooth) networks can suffer lower performance. The full impact of radio signal interference depends on the use and proximity of Bluetooth devices. Interference can occur only when both Bluetooth and 802.11 devices transmit at the same time. Users might have Bluetooth devices in their PDAs or laptops, but no interference will exist if their applications are not using the Bluetooth radio to send data. Some Bluetooth applications, such as printing from a laptop or synchronizing a tablet to a desktop, use the radio for only a very short period of time. In this case, the Bluetooth devices will generally not be active long enough to noticeably degrade the performance of an 802.11 network. For example, a user might synchronize his tablet to his desktop when arriving at work in the morning. Other than that, his Bluetooth device might be inactive the rest of the day.
The biggest impacts occur when a company implements a larger-scale Bluetooth network, such as one that enables mobility for doctors and nurses using wireless devices throughout a hospital. If the Bluetooth network is widespread and under moderate to high levels of utilization, the Bluetooth system will probably offer a substantial number of collisions with an 802.11 network residing in the same area. In this case, Bluetooth and 802.11 would have difficulties coexisting, and performance would likely suffer.
In addition to use, the proximity of the Bluetooth devices to 802.11 client radios and access points has a tremendous effect on the degree of interference (see Figure 4-10). The transmit power of Bluetooth devices is generally much lower than that of 802.11 WLANs. Thus, an 802.11 station must be relatively close (within 10 feet or so) of a transmitting Bluetooth device before significant interference can occur. A typical application fitting this scenario is a laptop user having Bluetooth supporting connections to a tablet and printer and 802.11 for accessing the Internet and corporate servers. The potential for interference in this situation is enormous, especially when the user is operating within the fringe area of the 802.11 network (that is, at the range that just supports the required data rate). The low-power Bluetooth signal will likely drown out the weaker 802.11 signal because of the distance of the access point.
Figure 4-10 Bluetooth Devices Offer the Greatest Impact on Wireless LANs in Fringe Coverage Areas
Neighboring Wireless LAN Interference
With the growing proliferation of WLANs, it is likely that neighboring wireless networks will exist in the area where an organization installs the WLAN. In most cases, the neighboring networks will implement a version of 802.11 that is compatible with your network. The advantage of this is that both your network and the neighboring one coordinate access to the common medium.
Interference between different WLANs occurs when the access points within range of each other are set to the same RF channel. As a result, a radio cell from one network overlaps a radio cell of the other network. In this case, with no data traffic occurring within each of the overlapping cells, there is insignificant interference. The only interference results from the periodic transmission of 802.11 access point beacons and other management frames.
Figure 4-11 shows the output of a Bantam spectrum analyzer depicting the presence of interference approximately 50 feet from a neighboring WLAN. In this case, the neighboring WLAN access point is set to one of the lower RF channels in the 2.4-GHz band. To counteract this potential for interference with its WLAN, the organization should avoid the lower RF channels near the side of the building next to the neighboring access point. It is likely possible to use the lower RF channels farther inside the building because the additional distance and walls would attenuate the signal coming from the neighboring WLAN to levels that will not cause interference.
Figure 4-11 Radio Frequency Interference Originating from a Neighboring Wireless LAN
When client devices in both cells begin to actively send and receive data over the WLAN, the client devices will share access of the medium, based on the 802.11 MAC protocols. (See Chapter 6, “IEEE 802.11 Medium Access Control (MAC) Layer,” for details on how these protocols work.) They will share the capacity of the RF spectrum, which reduces the potential performance that the networks can deliver. Because the interactivity between the client devices does not result in a significant increase in retransmissions, similar to interference from other sources such as microwave ovens (because the client devices are sharing the medium politely), the performance degradation might not be noticeable as compared to a client device operating from one of the cells while the neighboring cell has no activity. Thus, interference from neighboring networks is a problem only as utilization increases.