[Previous entry: "Geek Conferences in Iraq"] [Next entry: "Papers that you can read (see previous)"]
07/12/2004: "Meshblogs, part 33"
Sasha responds to my premise that we don't need lower power limits in 2.4GHz (under part 15) via his posting on Glenn's site:
As regards the 500mW limit I propose as a proactive solution in my manuscript, it is important to remember that one can today legally transmit at up to 1W; and with a exemption (which is pretty much a rubber-stamp process) and an amplifier, one can go up to 10W. The problem is not what power level one can transmit at with today’s wireless card technology;
47CFR Part 15.247 (the rules by which all WiFi and WiMax devices would need to run in 2.4GHz) contains no method by which this could be true. Nor does 47 CFR Part 15.407, which governs the U-NII band(s).
He continues
The problem is not what power level one can transmit at with today’s wireless card technology; but what will be rolled out in the future. I’m especially concerned about the WiMax technologies being proposed that allow for higher transmit powers within the same frequencies as today’s Wi-Fi systems — see this link.
with a pointer to this Which is a posting by him. Quoting again:
WiMax also "has been designed to scale from one up to 100's of users
within one RF channel". The White Paper points out that 802.16 utilizes
OFDM while 802.11 uses CDMA -- what worries me is that OFDM is built to be
usable at higher power -- thus, 802.16 could be deployed to "drown out"
802.11 transmissions. I don't know about OFDM versus CDMA, anyone want to
weigh in here?
First there are many 802.11 PHY standards. 802.11a and (most of) 802.11g do, indeed use OFDM as a modulation technique, not unlike 802.16 (WiMax). The paper Sascha points us to gets it right, but Sascha appears confused.
Second, 802.11b doesn't "use CDMA". The paper goes only so far as to call 802.11b "a basic CDMA approach", probably because 802.11b uses something like a spreading code, though to call CCK (the modulation used in 802.11b) a "spread spectrum protocol" is to do gross injustice to spread spectrum. In the 1997 standards, there were spread-specturm PHYs for both the DSSS and FHSS flavors of the now nearly-forgotten early days of 802.11, but the similarity ends here.
Third, OFDM is not "built to be useable at higher power". If anything, the design of OFDM makes building a higher-power unit more difficult, due to the Peak to Average Power Radio (PAPR) problem indemic to OFDM. This is a well-researched subject, familiar to anyone who has approached RF design with OFDM. If anything, 802.16s PHY(s) will have a worse time than will 802.11s PHYs, since they potentially use many more carriers, and the DSSS PHY (used in 802.11b) doesn't have much trouble at all. (He says, inches away from a 24-25dBm 802.11b bridge that will soon run OLSR.)
Perhaps Sasha meant that 802.11 runs CSMA (/CA)? The WiMax paper goes to great lengths to point out the superiority of the 802.16 MAC compared with 802.11's CSMA-based one. This may or may not be true, but it won't be a factor, since 802.16 is doomed, at least in unlicensed spectrum. And if Sasha did, then he seems very confused.
Sasha also states:
"A-HSLS will scale to thousands of nodes arranged in a truly non-hierarchical fashion — it’s the difference between two protocols that are most useful by major telecoms, and one that is useful for community wireless networks. A-HSLS is more useful for Community Wireless Networking purposes, while TBRPF and OLSR are more useful for major telecoms."
Yes, routing overhead is a factor in every routing protocol. But Sasha will find, with operational experience (which Chari seems to posess), that getting the data off the wireless network, and onto a wired network will ALWAYS provide the best performance.
In summary, I don't think that Sasha knows what he's talking about when it comes to RF and modulation (and hell, I barely make a dent compared to a lot of people I know). A-HSLS may be the worlds best ah-hoc routing protocol, but I doubt it. I certainly wish him well.
Me? I'll be here.
