Alignment tone disappeared long after 256QAM was added. I thought it was more of an oops than a temporary but intentional removal.
If it was a case of having to throw a feature overboard to make room and meet a deadline, it would be the new ETSI specs not 256QAM. Which would still support your theory they are at the limits of what the hardware is capable of. From: Colin Stanners Sent: Tuesday, March 17, 2015 9:27 PM To: [email protected] Subject: Re: [AFMUG] ePMP 2 GHz deployments >From my understanding: The 450 platform uses the standard "decode 2 signals separately" approach. Linear V vs H, or slant +45 vs -45 has ~30db isolation depending on the antennas so there's enough margin between the polarizations for the signals to be independant. But going from any linear to any slant polarization and vice-versa has only ~3db isolation so having 2 radio paths between different antenna types means the signals interfere heavily with each other. The radio would fall down to MIMO A (and in the days before 450 did MIMO A, it tried to use MIMO B with painful results). In comparison, one interesting thing about .N wifi chipsets including the one in the ePMP is that they have an approach where they use both antennas to receive two slightly unsynced samples of both signals and they decode the original signals based on differences in the sums that were received. That's why consumer routers can be MIMO using multiple antennas of the same polarization located right next to each other, in houses with many reflections. In this case, loss of most isolation between signals is not a big problem as long as there is still a time or phase-based difference. But I'm thinking that that approach may not work well over longer outdoor links? I assume adding the above capabilties would be too much for the Canopy FPGAs, which (based on the temporary removal of the alignment tone in the SMs) may be getting maxed out with the 256QAM feature that was added. On Tue, Mar 17, 2015 at 8:24 PM, Kurt Fankhauser <[email protected]> wrote: That just boggles my mind. Why doesn't the 450SM work with both antenna types then? Why would a cheap $100 SM radio have a more sophisticated antenna design than a $250 FPGA radio? I can personally attest that you CAN NOT hook up a 2.4ghz 450 slant SM to an AP that's running a V/H antenna. Tried that with horrible results.... Kurt Fankhauser Wavelinc Communications P.O. Box 126 Bucyrus, OH 44820 http://www.wavelinc.com tel. 419-562-6405 fax. 419-617-0110 On Tue, Mar 17, 2015 at 9:20 PM, Mathew Howard <[email protected]> wrote: It works with any SM, there's nothing special about the antennas. On Tue, Mar 17, 2015 at 7:54 PM, Josh Luthman <[email protected]> wrote: Any SM or just the ones that were made to support slant? Josh Luthman Office: 937-552-2340 Direct: 937-552-2343 1100 Wayne St Suite 1337 Troy, OH 45373 On Mar 17, 2015 8:51 PM, "TJ Trout" <[email protected]> wrote: There was an explanation made that was over my head even with 20 years rf experience but basically the sm doesn't care sees them both the same strength and no compromise On Mar 17, 2015 4:35 PM, "Josh Luthman" <[email protected]> wrote: I'm curious to know what everyone is using for 2 GHz sectors. Primarily linear vs slant but also OEM vs aftermarket sectors. PS. I heard through the grapevines that the (integrated) CPEs can do slant and linear. Is this true? Is there no loss in signal regardless of your AP/Sector being 90*/45*? If the patch antenna can do both, doesn't that mean it isn't focusing on one or the other and had to compromise? Josh Luthman Office: 937-552-2340 Direct: 937-552-2343 1100 Wayne St Suite 1337 Troy, OH 45373
