On Jun 26, 2012, at 4:54 PM, David Burgess <apt....@gmail.com> wrote:
> That said, it's good practice to keep the beam as narrow as is practical and > reduce transmit power accordingly. This reduces the > amount of noise you are spreading to the neighbours as well as the > probability of others eavesdropping. This is one of those "RF engineering" things that doesn't belong on the pfSense list. That said, in order to make 802.11's CSMA 'work', the side-lobe strength can't be too low. There are a plethora of issues when working with the 802.11 MAC, which strongly assumes omni antennae and relatively high signal strengths. I just can't allow your statement about keeping beamwidth narrow and tx power (more EIRP) low to stand. With 802.11 devices, if you don't set "carrier detect" on the other radio's in your field of view, as the population of same increases, it is increasingly likely that one of them will decide to transmit while you're attempting to *receive* a packet, ruining your chances of successful reception, due to the changes in the channel vector. Do this often enough, and the whole thing flatlines. I'm not even going to respond to your "probability of others eavesdropping" assertion on-list. I've been round and round on these issues (and others (*)) on this list (and others), and it's almost never a 'win'. Ten years ago, I thought I understood radio, and especially 802.11. Then I embarked on a journey staffed with real experts who have forgotten more than I'll ever know on these subjects. I'm not even going to lecture on the difference between "noise" and "interference". This is the pfSense list, and 'Paul' wanted a solution for his *satellite* modems that have an Ethernet hand-off. I don't believe pfSense will do this without running a routed network, which should be fairly straight-forward to do. (Just pretend that the two satellite modems are connected to a LAN. A LAN that happens to have a lot of propagation delay, and not quite as much bandwidth as you would assume, but... a LAN none-the-less.) Running a VPN over that is straight-forward. Jim (*) Often it's the impossibility of using all three 'non-overlapping' channels in 2.4GHz Sure, the transmit masks don't overlap, but you have to look at the in-channel power from adjacent channel (or even alternate channel) operation. Hint: "adjacent channel rejection"
_______________________________________________ List mailing list List@lists.pfsense.org http://lists.pfsense.org/mailman/listinfo/list
