Wireless LAN Banwidth can be impacted by outside interferences, such as Bluetooth.

The widespread adoption of IEEE 802.11b/g wireless LAN (WLAN) technology within the medical community has been a key factor in the integration of patient care and information technology. Touted as an inexpensive alternative to conventional hard-wired networking, it relies on utilizing the unlicensed 2.4-GHz industrial, scientific and medicine (ISM) band in combination with direct-sequence spread spectrum (DSSS) technology to free the user from cumbersome cables while increasing mobility.

A comparative newcomer to the WLAN environment, IEEE 802.15–more commonly known as Bluetooth (BT) technology—also shares the 2.4-GHz ISM band. While traditionally associated with extremely short-range or personal-area networks (PANs) and using the frequency-hopping spread spectrum (FHSS) technology, the new generation of higher-powered BT-enabled equipment is capable of extended range communications, with much higher power levels.

The randomness of problem events, however, make traditional short-term troubleshooting captures using a standard WLAN network analyzer challenging. In one example, a company made multiple, long-term packet captures over the course of several days from several points within its WLAN. The network engineer deployed a distributed analyzer with long-term capture and storage capabilities and strategically placed 802.11 capture probes in multiple locations.

The captured data revealed that while normal WLAN data speed was 11 Mbps immediately preceding the random outages, user-perceived network slowdowns, combined with AP/client rate-shifting from 11 Mbps to 5.5 Mbps to 2 Mbps, were observed. Network errors and WLAN cyclic redundancy check errors were observed to significantly increase from a baseline of less than 5% to well above 60% immediately prior to the workstations losing connectivity to the network.

In addition, an evaluation of the analyzer’s corresponding expert system events revealed numerous instances of WLAN interference, wireless physical errors, excessive client retries and numerous wireless data rate changes. The network engineer observed that the users complaining of WLAN disconnects were also using the new Bluetooth headsets, and that disconnects seemed to occur when the user was more than 10 feet away from the workstation.

A previously undiscovered interference source was suspected but where the interference was coming from was not clear until the Bluetooth headsets were evaluated. While the two WLAN technologies utilize completely differing technologies (DSSS vs. FHSS), they both occupy the same 2.4 GHz ISM band.

Under the new IEEE 802.15 rules, Bluetooth has been modified to allow for three separate power levels. This capability is employed in many of the new generation Bluetooth-enabled office devices, including the users’ headsets in question.

Additional captures showed that as a user was actively using the headset and moved more than 10 feet away from the base station, the power level changed from 1 milliwatt to more than 2.5 milliwatt. Although a low power level will not affect 802.11b performance, when the headset power shifts into high, it can and does disrupt 802.11b signals. This near tripling of the radiated FHSS power level, combined with the location of the base-station almost directly adjacent to the DSSS workstation NIC card, plus the overlap in the RF spectrum, manifested itself in seemingly random device or network failures.

Both short- and long-term solutions were available. The network engineer moved the 802.15 Bluetooth headset base stations as far away from the 802.11b NIC cards as possible. A long-term solution was to deploy non-overlapping IEEE 802.11a equipment wherever the Bluetooth devices were located.

.: Back to Menu :.