Does the ePMP support Matrix A MIMO? I ask this because I've seen contradictory answers form users in the forum, but no official word.
One of my clients has some existing PMP 320 WiMAX gear on 3.65 GHz, and put an ePMP sector on the same tower. They put up an ePMP Force 200 on 5 GHz at a user site and per my calculations, the signal should have been pretty comparable. It's in a wooded area, from a mountainside AP, and 5 GHz has slightly more foliage attenuation than 3.65 GHz. But the 5 GHz barely connected while the 320 did connect quite well. One possible explanation is that the 320 went into Matrix A mode while the ePMP insisted on Matrix B.
I believe the Qualcomm chips are capable of Matrix A operation, in which the two chains transmit the same thing and the receiver picks the better signal in real time. It is extremely useful in the woods. Is this in the ePMP and if not, is it a possible firmware enhancement? Thanks.
From Sri:
"MIMO-A which is Matrix A is a little bit more. It uses a technique called STBC (Space Time Block Code) where we use space time coding to receive the same data on two polarities and put together one good copy out of it. This is not done on ePMP and is not native to 802.11 devices. Matrix A helps with nLOS, NLOS or interference situations. Specifically with interference, the advantage you get is that if one polarization is interfered with, you still can get a good copy from the other polarization. In nLOS and NLOS situations where you are, say shooting through trees, you can have both polarizations being messed with but at the receiver it has a better chance of putting together a good copy using two corrupted pieces of data.
PMP 450/450i does true MIMO-A. This gives 450 an advantage over ePMP for tough shots"
I wonder if the 802.11-based chips are up to doing Matrix A with driver help. It would be nice. The 450 family is, well, rather pricey, and I'm looking at sectors with relatively few potential users, due to the terrain. I raised the question of 450 vs. ePMP on 2.4 GHz some time ago, I think on the 450 forum, and there was no consensus that the 450 would do better in the woods. The big Force 200 dishes may help make up some of the differnce. I wonder how they compare in that regard to other vendors "competitive" solutions, like the un-elevated Ubiquitis.
We did just put some 450i's in on 900 MHz, replacing Canopy 100s, in a wooded area, and the performance is amazing. But the lower frequency does most of the NLOS part of the work.