bryan - misc comments inline... when last we saw our hero (Thursday, May 09, 2002), Bryan Halvorson was madly tapping out: > Austad, Jay wrote: > > > Well, the guy that runs ingenious-nets.com(or .net, forget which) told me > > that he has a 23 mile 802.11b link, using only a 24db directional on each > > end, no amplifiers. After 23 miles, you start running into curvature of the > > earth issues, but with an antenna that is 80 feet high, you might be able to > > squeeze some more range out of it. Plus, you can toss an amp on each end > > also. > > I have some software that can do plots of general coverage and > point-to-point links. For some examples of what it can do, look at: > > http://fox.sector14.net/~bryan/radio/2g_coverage.jpg > > Warning this is a 990k jpg! > This is a coverage map from my home in Apple Valley and assumes a 40 > foot tall antenna on the receive end with 15db of gain in the antenna > system. The different colors are signal level in dbm. A -82 dbm signal > level is the minimum required for a full bandwidth connection for the > Orinoco cards. > > http://fox.sector14.net/~bryan/radio/n0buu-ka0ztt.jpg > > This is a point-to-point plot from my place to a friend's place just off > of 35E and Grand Ave in St. Paul. The receive sensitivity is wrong in > this one but it shows that the link could possibly work but will > probably be flakey because of the closeness of the hill in the middle. these are some impressive outputs. i'm particularly impressed with the x-section that you've got displayed in the 2nd image that you've linked to. - of interest to me is the display of the fresnel zone and what i surmise is the straight line topology as calculated by USGS data? this is a really nice tool - do tell what is it? > > Say we did build a sweet wireless network... It's obviously going to have > > multiple connections to the internet. So how do we plan on doing routing? > > Do we want people to go out the nearest internet connection? Or, to get > > more complicated, out the *best* internet connection? > > This is kind of what Andy was getting at. Hot spots would probably each > have their own Internet connection while a metro wide network will > probably only have a single connection to the Internet. The design of > each type of system is very different. > > I think that it's a bit too early to try to build a metro wide network > at this time. I think we should work on Hot spots for now and we can > revisit the issue in a few months when we see how the group is growing. i would concur with this - there are a few different mechanisms for the routing of traffic to the nearest (and best) internet gw in the overlay objective. but given the complexities associated with this and the required routing infrastructure to accomplish this appropriately this is likely a good task discussion for down the line. are we building to a consensus yet? {snipped - .signatures ...} -- steve ulrich sulrich at botwerks.org PGP: 8D0B 0EE9 E700 A6CF ABA7 AE5F 4FD4 07C9 133B FAFC
