Jim, You're absolutely correct.
Most technologies have trade-offs. OFDM deals well with delay spread up to the limit of the cyclic prefix and then it fails due to ISI. Outdoor propagation envirnments encounter delay spreads of several milliseconds, up to a realistic maximum of 10 ms [ref. papers by Cox]. Indoor delay spreads are a few hundreds of nanoseconds max. [ref. papers by Devarsivatham, Rappaport, etc.]. According to one of your earlier postings, the delay-spread limit for 802.11a is 800 ns. The delay-spread limit for the defunct Cisco Clarity OFDM gear was 8 ms. Many folks with practical measurement experience will argue that a limit of 5 ms is fine for systems that use directional antennas. (Patrick, what is the delay-spread limit for Alvarion OFDM gear?) What keeps coming through in your postings is that you've tried 802.11a outdoors and it failed miserably. Duh. That only means that 802.11a doesn't work outdoors, it does not mean that another implentation of OFDM doesn't work outdoors. On the contrary, there are numerous examples of OFDM systems that work great outdoors (most notably, the European HDTV standard). These systems don't require magic to work, they work because they were carefully engineered to do so. So do us all a service by being careful with your terminology. If you don't like 802.11a, then say you don't like 802.11a, not "OFDM". Regards, (Really!) Greg On Wed, 2003-07-30 at 04:13, Jim Thompson wrote: > > You're making my point, but may not understand 'why'. Some of the deployment > scenerios you describe are extremely difficult, and changing > modulation may not be enough (heavy understatement). > > I only became sarcastic when you started asserting some 'magic' to > OFDM over single-carrier (or spreading) systems. Most technologies > have trade-offs. OFDM deals well with delay spread up to the limit of > the GI, and then it fails due to massive, overwhelming ISI. > > RF is physics. Physics is mostly math. Modulation is math. You > offered emperical data as 'proof', but proofs require mathematics. > > Being completely open, I'm sure your product works in an interesting > number of scenerios. Really. I hope you sell a ton. > > But in the end propagation is physics, and "you canno violate the laws > of physics, captian." > > The emperical data you've offered are short textual descriptions of > deployments. I'm interested either in the modeling (or measurements) > that show your product works, or in the actual data from the > deployments you describe. > > OK? > > Jim > > Patrick Leary writes: > > All things "fail" when used beyond their capabilities and/or intent, so I > > don't regard such things as failures of the technology so much as failure of > > the human implementing it. For example, would you be the type to drive a car > > into a lake, then call the car's inability to reach the other side a failure > > of the product? > > > > Jim, the larger point is that your sarcasm is unwanted and unwarranted. I > > see now how you respond once your intellectual arguments themselves fail. > > You switch to personal insult. Accordingly, my dialogue with you has ended. > > Enjoy your "last word" - no doubt you are the kind who must have it. > > > > - Patrick Leary > > > > -----Original Message----- > > From: Jim Thompson [mailto:[EMAIL PROTECTED] > > Sent: Monday, July 28, 2003 3:56 AM > > To: Patrick Leary > > Cc: [EMAIL PROTECTED] > > Subject: RE: [BAWUG] 802.11b Long Range non line of sight > > > > > > > > I'm sure you have many successful deployments. I'm sure your products > > are all winners, and your shareholders think you walk the earth in > > peace. > > > > I'm interested in where things fail(ed), and 'why'. > > > > Jim > > > > Patrick Leary writes: > > > Fair enough, so here is a list of a few of the areas where we tested > > and/or > > > have implemented OFDM in scaled deployments so far: > > > - extremely urban-type low building density of a major city in South > > America > > > - heavily foliated coastal hills, across bays, and urban settings in New > > > Zealand > > > - low mountains with heavy deciduous tree coverage in rural western > > Maryland > > > - coastal Southern California with little vegetation > > > - flat arid, with marginal density in the Texas Panhandle > > > - heavily foliated and urban landscapes in Hungary > > > - deep in the very heavy coniferous high mountains in rural British > > Columbia > > > in Canada > > > - an urban landscape in the Russian steps > > > > > > I'd say this is a fairly strong cross section of environments. Have any > > more > > > negatives you might like to guess about on things you have never seen? > > > > > > Patrick Leary > > > Alvarion > > > > > > -----Original Message----- > > > From: Jim Thompson [mailto:[EMAIL PROTECTED] > > > Sent: Friday, July 25, 2003 3:41 AM > > > To: Patrick Leary > > > Cc: 'Tim Pozar'; Ladjicke Diouf; [EMAIL PROTECTED]; > > > [EMAIL PROTECTED] > > > Subject: RE: [BAWUG] 802.11b Long Range non line of sight > > > > > > > > > > > > Empirical data can only demonstrate results in the environment you > > > studied. > > > > > > Proofs require mathematics. > > > > > > Patrick Leary writes: > > > > Tim, I am not sure if you are talking about OFDM or DSSS. With OFDM, you > > > > DON'T need LOS. Of course its not going to connect forever with NLOS, > > but > > > > for a few miles, it is a no brainer. We have ample empirical data that > > > > proves it. > > > > > > > > Patrick > > > > > > > > -----Original Message----- > > > > From: Tim Pozar [mailto:[EMAIL PROTECTED] > > > > Sent: Thursday, July 24, 2003 8:27 PM > > > > To: Ladjicke Diouf > > > > Cc: Patrick Leary; [EMAIL PROTECTED]; [EMAIL PROTECTED] > > > > Subject: Re: [BAWUG] 802.11b Long Range non line of sight > > > > > > > > > > > > On Thu, Jul 24, 2003 at 02:26:26PM -0700, Ladjicke Diouf wrote: > > > > > Can somebody shed some light on how OFDM helps NLOS, I thought it was > > > just > > > > > > > > > a coding scheme like DSSS for 802.11b > > > > > > > > You still need LOS. OFDM of DSSS will handle interference (ie. > > > > "smearing" ) of the signal better with the lower symbol rate. Things > > > > like multipath will be less of an issue. Still an issue, but less > > > > pronounced. > > > > > > > > Tim > > > > -- > > > > Snail: Tim Pozar / LNS / 1978 45th Ave / San Francisco CA 94116 / USA > > > > POTS: +1 415 665 3790 Radio: KC6GNJ / KAE6247 > > > > "Be who you are and say what you feel because the people who mind > > > > don't matter and the people who matter don't mind." - Dr. Seuss > > > > > > > > > > > > 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 > > > > > > > > > > -- > > > "Speed, it seems to me, provides the one genuinely modern pleasure." > > > -- Aldous Huxley (1894 - 1963) > > > > > > > > > > > > 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. > > > > > **************************************************************************** > > > ******** > > > > > > > -- > > "Speed, it seems to me, provides the one genuinely modern pleasure." > > -- Aldous Huxley (1894 - 1963) > > > > > > > > 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 > > > > -- > "Speed, it seems to me, provides the one genuinely modern pleasure." > -- Aldous Huxley (1894 - 1963) > > -- > 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
