Jeff,
I agree completely with your reasoning. As you said, 10 miles pushes the capabilities of most part-15 equipment with decent size antennas. Therefore, if there's just enough signal to deploy a LOS 10-mile link with sufficient fade margin to be reliable, then there's unlikely to be enough signal after diffraction, scattering, etc. to deploy a NLOS 10-mile link. (I have however, done 10-mile NLOS links, diffracting over ridges with foliage, with large 4-ft dishes--there's an exception to every rule!) And I further agree that OFDM is very well suited in an urban environment with strong signals but severe multipath. (Assuming that _the_particular_ OFDM system to be used is designed for the amount of multipath delay spread that one encounters in outdoor urban propagation environements (e.g. several milliseconds typically, up to 8-10 ms [ref. papers by Don Cox]) In other words, don't take an 802.11a system designed for indoor use an a maximum delay spread of 800 ns (Jim Thompson's number) and then whine because it doesn't work for an application that it was never designed for! There are, however, plenty of OFDM systems that have been designed for outdoor use (European HDTV, the Cisco Clarity gear, Alvarion, etc., etc., etc.) Btw, the Cisco Clarity gear was designed for the outdoor environment from the get-go and handled multipath delay spread up to 8 ms. Regards, Greg DesBrisay On Thu, 2003-07-31 at 12:24, Jeff King wrote: > Just so this doesn't get too far off topic what was being responding to was > someone desire to do a 10 mile NLOS path. 10miles pushes the capabilities of > almost all part 15 equipment, let alone a NLOS link. > > Some links have been posted that claim OFDM is better in a urban environment > with strong signals but severe multi-path (I think the link distances cited > where 1.5km and 3.0km respectively). I don't question this claim as I also > use OFDM for amateur radio, in which multi-path delay can approach 1ms > (Pactor 3, MT63 and Q15X25 are all amateur radio OFDM modes). > > What I am questioning is the claim that OFDM offers better long distance > penetration abilities as compared to other data methods. I've no doubt that > in a severe multipath environment, with strong signals, it can offer some > advantages, I've just seen no evidence a different data method can bend the > laws of physics, when multi-path is not a significant player. Now, if you > saying xPSK is a more sensitive modulation then X, then just say this. But I > question if OFDM, by itself, comparing apples to apples, offers any intrinsic > penetration advantages. > > This is what I am looking for this white paper to answer.... > > Then again, those that know, already know, so maybe it is time for the > moderator to give this thread a well deserved death? > > -- > Jeff King, [EMAIL PROTECTED] on 07/31/2003 > > > On Thu, 31 Jul 2003 06:38:45 -0700, Patrick Leary wrote: > >Judd, You are comparing apples to bananas to oranges. I know; I make > >them all (DSSS, FHSS, OFDM, and 900MHz). 900MHz produces better > >penetration soley as a matter of frequency, not modulation. And if > >you had a 900MHz that also did OFDM, you'd learn it would be far > >more capable than even my own 900MHz FHSS or anyone else's 900MHz > >DSSS. > > > >You are welcome to think OFDM is "useless." You should know though > >that every major WLAN company disagrees, as do the engineers in the > >IEEE. 802.11a is OFDM based. 802.11g is OFDM based. 802.16a is OFDM > >based. Within 2 years, you'll have a very difficult time finding any > >DSSS in WLAN world. The consensus is all there that OFDM represents > >the core of the next generation. Even Aperto, Proxim, Airspan, and > >others historically resistant to it, have now said OFDM will form > >their core products. > > > >As for UWB, no one with any knowledge would disagree that it is am > >extremely capable technology and can do all the best of everything. > >You neglect the reality of the regulatory environment. The FCC > >barely approved what minimal version they did. It matters not one > >wit what something can do if the feds do not permit the technology > >to be implemented to maximum effect. > > > >Lastly, you should be candid with this community and admit your > >visceral hatred of all things Alvarion. Lord knows I have the > >collection offlist threats and attacks from you over the past 2 > >years. It is also why no positive Alvarion comments or Alvarion > >people are allowed on his "uncensored" list. That is also why Judd > >will always take a position contrary to my position, regardless of > >its intellectual merit. > > > >Comments about our being proprietary are silly to the absurd. In > >almost every standard to date you'll find us as part of the core > >team creating it. In wireless broadband, there was no standard to > >implement. EVERY vendor building specific for wireless broadband had > >no choice but to implement "proprietary." There was no standard to > >implement. Now there is with 802.16a and where is Alvarion? We were > >in 802.16a in the begiunning. We are chiefly the ones responsible > >for creating harmonization between ETSI HiperMAN and the 802.16a > >(which was a very tough battle our scientists won that is key to > >industry adoption going global). Oh, and we hold the number 2 and 4 > >positions (VP and Treasurer) at WIMAX. And we announced with Intel > >just a few weeks ago that we will be the first to use their > >standards-based 802.16a silicon. > > > >Alvarion, Motorola, Proxim, Trango, Aperto, Airspan, Navin, > >Beamreach, Cambridge, Netro, Hybrid, Vyyo, P-Com, Western Multiplex, > >Malibu, Arraycom, Waverider, WiLAN, Redline, and the few others I > >missed...they all have proprietary products at the moment. Now that > >is an inconvenient fact isn't it? > > > >As for affordability? It is true, you find us too expensive. > >However, most of the market of legitimate operators disagree. Ours > >books are public. Examine them for yourself for the proof by > >comparing our results to any and all of our peers. > > > >- Patrick > > > > > >-----Original Message----- From: [EMAIL PROTECTED] > >[mailto:[EMAIL PROTECTED] Sent: Wednesday, July 30, 2003 11:08 > >PM To: '[EMAIL PROTECTED]' Subject: Re: [BAWUG] 802.11b Long > >Range non line of sight > > > > > >With DSSS and FHSS, you don't need LOS either. But there are > >extreme limits on distance and when it comes to getting links on a > >sectored or omni directional ptmp system, your going to be hit-and- > >miss. > > > >>From what I've heard, the 900Mhz stuff works well, very well, NLOS > >>like we > >see on our cell phones, where a general wall or building isn't going > >to kill your signal to an unusable amount. But at 2.4Ghz and > >5.8Ghz, even with OFDM and AP's that cost $2500+ each and $600-1000 > >CPE, your not going to only battle hit-and-miss coverage, but then > >you begin the battle price vs widespread acceptance of the > >technology. > > > >UWB might be the next big step, where, instead of a complete loss of > >a connection, you only lose part of the connection that is blocked > >and the throughput may fall, but may still be usable at 100Mbit of > >sustained throughput, even with error rates. > > > >Personally, I think that OFDM is useless unless it becomes > >affordable. Alvarion has never brought equipment down to a generally > >affordable level, in contrast to other existing equipment solutions. > >So I don't have much faith in anything that Alvarion claims, even if > >it is true and does work, cuz we don't want to go broke implementing > >proprietary solutions that give no consideration to current market > >demands, including price requirements for acceptance. > > > >By widespread acceptance, I mean that at some point, the equipment > >would become fairly "standard" for the industry. > > > >Judd > > > >Jeff King wrote: > > > >>Thanks Patrick. What I am looking for is the "white paper" that > >>will > >qualify > >>your statement: "With OFDM, you DON'T need LOS." in the context of > >>the > >title > >>of this thread (or at least the frequency domain). > >> > >>-- Jeff King, [EMAIL PROTECTED] on 07/29/2003 > >> > >>On Tue, 29 Jul 2003 17:15:04 -0700, Patrick Leary wrote: > >>>Until I can link to our paper, here are some resources to study > >>>OFDM. > >>> > >>>http://www.palowireless.com/ofdm/tutorials.asp > >> > >>-- general wireless list, a bawug thing <http://www.bawug.org/> > >>[un]subscribe: http://lists.bawug.org/mailman/listinfo/wireless > > > > > > > > > >-- general wireless list, a bawug thing <http://www.bawug.org/> > >[un]subscribe: http://lists.bawug.org/mailman/listinfo/wireless > > > > > >This mail passed through mail.alvarion.com > > > > > >**************************************************************************** > >******** This footnote confirms that this email message has been > >scanned by PineApp Mail-SeCure for the presence of malicious code, > >vandals & computer viruses. > >**************************************************************************** > >******** -- general wireless list, a bawug thing > ><http://www.bawug.org/> [un]subscribe: > >http://lists.bawug.org/mailman/listinfo/wireless > > > > > -- > general wireless list, a bawug thing <http://www.bawug.org/> > [un]subscribe: http://lists.bawug.org/mailman/listinfo/wireless -- general wireless list, a bawug thing <http://www.bawug.org/> [un]subscribe: http://lists.bawug.org/mailman/listinfo/wireless
