TESTING WLAN EQUIPMENT DATA RATE AND SENSITIVITY

Many suppliers of radio equipment for WLAN, do not state output power and sensitivity for reception of data at different data rates. These parameters are needed to be able to calculate a correct linkbudget. Therefore we make tests on equipment that is new on the market, or in the cases where we think that something does not work properly. On this page we describe in short what we have found. Read more about the test method at the bottom of the page.

ORINOCO
These PCMCIA cards which mostly are used as client cards in portable computers, USB adapters, or Ethernet converters, are also used as radio cards in access points like AP-500 and AP-1000. They have been on the market for many years under a variety of names. Sensitivity: almost always better than the -82 dBm that the manufacturer states. Typically -85 dBm at 11 Mbit/s data rate. The new card with Ruby chipset however has less sensitivity, around the -82 dBm stated in the data sheet. Output power: 15 dBm

NETGEAR FM411P AP/ROUTER
Sensitivity -82 dBm at 11 Mbit/s. 2 dB less sensitivity than Orinoco at 1-2 Mbit/s data rate. Very robust physical design. Connector type RP-SMA. Only 2-3 Mbps efficient bandwidth compared to the Orinoco cards.

NETGEAR WG602AP
Sensitivity -83 dBm at 11 Mbit/s. About 15 dB less (ie -68 dBm) at highest measured bandwidth (23 Mbit/s) at IEEE 802.11g setting with the Linksys IEEE 802.11g card. Connector type RP-SMA.

LINKSYS WRT54G
The first product on the market for IEEE 802.11g. Sensitivity -82 dBm at 11 Mbit/s. Sensitivity -67 dBm at 54 Mbit/s, and another 3 dB for every step 48, 36, 24, 18 Mbit/s. We where only able to achieve 36 Mbit/s data speed through open air. To reach up to 54 Mbit/s, we had to connect the units galvanically!

Test method

We set up a link with a client card and access point (AP) which looks as follows.

The two stations are galvanically separated by placing two antennas in a shielded box. The isolation between these two antennas is 15 dB. The reason for this galvanic isolation is that some equipment uses direct current on the coaxial cable to control diversity, and we do not know how the controlling algorithms for this work for all brands.

As a reference we use two Orinoco cards where we have removed the built in antennas to avoid leakage.

The test method for bandwidth is TPTEST (see other page on this subject) and we always use the same computers and operating systems for testing. It is not always easy to use bandwidth as a means to see at what data rate the packets are transmitted. Data rate is decided for each frame. In the client program for the Orinoco cards the data rate is stated per frame, and this is what we use as reference when testing other cards.

If you have questions, please contact Tonny Vilnersson or Niklas Gunhamn.