As stated here:
http://documentation.netgear.com/wag102/enu/202-10120-01/v1/WAG102-4-10.html
and in many other sources, WEP key can be 64bit, 128bit (factors of 2) and 152bit length.
Factor of 2 length I do understand. It is convenient to use in low-level processing, where you have to use boolean logic and save everything in bits.
Any idea why the 152bit key was selected?
Best Answer
I will preface this by saying this is a terrible answer, first because I can't find any of the references and am going off of memory. Second because I haven't put much thought into WEP for so long that my memory fails me.
As to why it was 152-bit, IIRC, this was a 128-bit standard WEP key with an additional 24-bit extension. The 24-bit extension was one of several proprietary extensions in an attempt to make WEP more secure (or at least appear more secure). Cisco's MIC and TKIP won out as the most widely recognized extensions to make WEP more secure (more commonly known as WPA).
152-bit WEP ultimately faded away as many proprietary extensions do when they don't get enough industry support and/or are over shadowed by better extensions.
Ultimately, it really doesn't matter at this point though as WEP is a fundamentally broken technology and WEP, WPA/TKIP and WPA2/TKIP are all defunct technologies. By this I mean that the 802.11n amendment says an AP has to disable the high throughput data rates and function as an 802.11a/g device if any of the above are enabled.