I asked Dave Hughes about these on the WELL and boy did I get a great answer. Dave Hughes is a real pioneer, I consider him legendary in his time. And I get to talk to him frequently. What an honor. #417 of 435: Paul Terry http://www.spring.com/yapp.html (terry) Mon Jan 13 '97 (00:46) 2 lines Dave, can you discuss the details of setting up the Freewave? What is the process of setting it up? Topic 15 [wireless]: Wireless IP: New alternatives to leased lines #418 of 435: Dave Hughes (dave) Mon Jan 13 '97 (02:39) 87 lines Piece of cake. When they ship (in pairs) FreeWave has usually set them up to talk to each other (by radio number, which is printed on the bottom of the radio), at the highest speed, both radio and port (115Kbps). To set any other way, you plug into the 9 pin port of any PC with a terminal program at 19,200 baud 8N1, and hit the reset button on the front of the radio panel. A menu comes up. It shows you, under each item, what the settings are. From memory I can't remember them all, but the key ones are: Mode - you have to have one in either P to Point Master, while other one is a slave, Point to Multipoint master, and slave, or relay. Port - you set the speed of the RS232 port, from 1200 to 115,000 as attached to your radio when the radios are operating. Call Book - you list *which* radios, by number, this radio will connect with (0r in the case of multi-point, up to 9 other radios) The rest of the items are radio and radio diagnostic, which you don't need to fiddle with, unless there are real reasons to change packet l;engths etc, and you really know what you are doing. (we have had no reason to change factory set defaults in the 10-20 radios we have set up from my front room to Mongolia). Then you hit the ESC key, its written to the chip, and you (re)set the other one. When both are set right the left green link light will come on. If not, no green. And if you are out of range, no green. If green, and you are attached to a PC at each end, then, just like typing over a plain attached serial line between computers, things work. And you will see the Clear to Send and the Transmit red lights working as data passes. The sets come with a 9 pin to 9 pin serial modem cable for each radio. (you can just buy an adapter if you attach to a Mac). And the 4 inch rubber duck antenna. I also bought a 20 foot RF cable, from them (with *their* antenna connectors on one end, an SMA connector with reverse thread, which is what makes it non-standard, and a standard END connector (female) at the other end. Together with a 12 inch coil antenna, with male END connector and mounting braket. This will give you 3Db of gain. So when the little rubber duck antennas don't reach, you can affix one (or both ends) one of these, and run the antenna into a roof cavity, our out a window on the side nearest the other radio, to get a link. This is what I had to do when hooking up a pair of FreeWaves between a classroom Mac at Evans Elementary School and the Terminal Server next to the Cisco router right in the Computer Room of Adams State College, about a half mile away, but at ground level, through about 10 homes and stone wall campus buildings. I could get a green link light with just the built in antenna from several locations at the school and in the classroom, but it wasn't solid (intermittent green). So by getting one of the omni's and cable and telling Roger Quintanilla (a really neat, imaginative, modem-net-using 5th grade Hispanic teacher) to just experiment, he worked out that the best location was on the outside wall, under the eves of the roof. Even that didn't give him the highest thruput, as measured by his doing ftp transfers and measuring (while his kids recorded the data, comparing it with phone modem transfers between the same services, and plotting graphs for our NSF Tests). So I put another omni in the computer room, which helped. But hey, this is as much an *art* as a science, expecially since diagnostic equipment is either so damned expensive (we have an HP spectrum analyzer that cost $15,000) or just doesn't cut it for spread spectrum signals! Dewayne and Glenn Tenney, whom I contracted with to go to Mongolia and set up those 8 sites with Freewaves, discovered that the closest 4 sites could be reached from inside the building with the little 4 inch antennas, and only the ones out to 10km away had to have the rooftop 20-30 foot omnis. And they never could reach the US Embassy directly from the DataCom base site. So had to configure one at the Academy of Sciences location to 'relay' to the Embassy, which is in a kind of depression, or hole. All of which was a commentary on the Soviet construction techniques for the buildings, with not much metal. Hey CIA, Russian commercial buildings pass 902Mhz rather nicely! (Oh, you already knew. Ok. Just trying to be helpful!) Dave. You're more than "helpful". Excellent explanation! - ptw

Dave Hughes sums up the challenges to the amateur radio packet community: An interesting thing passed onto me by Dewayne Hendricks, TAPR Ham and my colleage on some of the NSF wireless projects, is the suprising lack of growth in the TAPR (Tuscon Amateur Packet Radio) organization, a subset of the large AARL. I understand the rough numbers are about 300K AARL (just plain amateur radio) and only 2,700 TAPR (digital radio amateurs). With the explosion of cheap microcomputers and hackers - which has, enabled the technology of non-interferring frquency hopping or direct sequence, processor controlled, radio, I would have thought that the digital-radio branch of ham radio would have grown exponentially. Especially since hams can do things with digital radio (power, range) that others cannot. And the combination of rapidly advancing processor technology - as evidenced by the ability of this design team to use many off-the-shelf components to make one of the most advanced radios in the world, yet still *really* cheap (military radios using spread spectrum techniques cost upwards of $50,000 to produce). We are indeed going through a major revolution in radio, PARTICULARLY via spread spectrum with the stunning possibility that we can have millions of secure, reliable digital radios all occupying the same space and not interferring with each other to any practicable degree. THAT is revolutionary, and undercuts the whole mountain of assumptions (that all radios interfere with each other locally) upon which our regulatory empire (FCC) is built. (The Gilder 'unlimited spectrum arguement') Well, this little $500 'radio' being built by amateurs - most of whom, of course, are nevertheless radio engineers in big companies or very advanced small ones - may be the price/performance 'breakthrough' radio that will permit Joe Sixpack and Johnny's Little ISP to team up, bypass the telcos locally (with from $125 per month for 56Kbs, to $650 per month for T-1, and upfront costs from $750 to $1,500 just in the lowest-cost big metro areas - forget the rural costs) and generate a revolution in personal Internet access. With speeds *starting* at 300kbps (versus 28.8 modems), free for the first 20 miles. 10 years ago I got a 'business band' licence so I could communicate digitally, using a commercial analog radio, and a Terminal Node Controller (TNC) developed by TAPR, giving me no more than 9,600 baud of throughput, not secure, and requiring a licence, because two such radios would interfere with each other on the same bands, unlike spread spectrum radios (when they are built right). Now we will be able to do this. Why this prospect (with over 50 companies making commercial versions right now, albeit starting at $1,000 per radio - while consumers by the millions don't flinch as spending $1,500 for their computers, and then $20 a month for s-l-o-w web access) doesn't exite people, I don't fully understand.

The success story of the year, from Dave Hughes! It works! Cripes, after ages of battling uncoperative telcos, cripplung FCC decisions, random distribution of mountain ridges, too laid-back] ISP owners, the physics of radio wave propegation, Frensel zone laws, and sheer distance, we now have a spread spectrum radio link allthe way from Alamosa, 30.7 MILES to the tiny burg of San Luis itself. And, typical of this self-imposed, very difficult task of trying to link the terminus of higher speed Internet lines in the valley, and its ISPs, both college and commercial, in Alamosa, to the tiny town and its school down in a hole-behind-the-ridge of San Luis in that remote and pretty damned unforgiving environment of 'El Valle' and 'La Sierra' every conceivable obstacle it seems, has been put in our way. By man, nature, and ? God ? (Remember, as lisa will, that radio tower above San Luis is just a few hundred feet from the magnificent Chapel of All Saints at the summit of the Stations of the Cross steep climb. Its as if the Spirits of a faith that measures devotion by the degree of penance, decreed that putting up that radio link would NOT be easy - very hard as a matter of fact. To teach us, me (since I was the one, two years ago who said - lets do the toughest link rural Colorado - with distance and mountains in our way, for if it is easy, what do we learn?) that it would be mind-bendingly difficult to provide a 'free' 30.7 mile Internet link while the siren song of a $1,700 a month 5 year contract with the Devil US West seems so easy. teach us a lesson to challenge the Gods of Telcos, Terrain, and Technology.) But if you ping or traceroute 209.94.82.50 - you will be travelling comfortably over the wired-fibered Internet to the town of Alamosa, into Amigo Net's premises, crossing over from a Livingston Router, to one of three of our No-Name Debian Linux routers, out the serial port into a Freewave Radio at 115kbps, up the RF cable to a just finished (as we watched) 60 foot tower to a little 3 foot long yagi antenna pointed *exactly* 271 degrees from North across the town and the Rio Grande river, across the sage and sand of the valley floor, between two ranges of hills and at *exactly* 30 miles, skimming a ridge I reported on above so long ago clearing it by under 10 feet, then to, *exactly* 30.2 miles to the 95 foot level of an old Sheriff's department radio tower (where nobody could find the key to their radio shack for weeks) then into the yagi pointed back toward Alamosa, (exactly two fingers left of that long white building 10 miles away - the only instructions we could give the guy charging $80 an hour to climb that 100 foot tower in the cold October valley wind) then down 160 feet with 6dB of line loss, into a hole in the shack, and into the *first* radio, which went 'green' yesterday for the first time, then into a somewhat questionable 286 MSDOS 640K 'router' running code Phil Karn wrote years ago, called NOS, via a serial port, and then out the other serial port to a *second* radio (which has been 'green' since spring) to an 8 foot omni antenna on the shack's roof, which overlooks the town of San Luis, the Vega (the only 'commons' in the country deeded down from Spanish Law) and to with the tiny communities of San Pablo, Chama, and San Francisco in line of sight view, then down over the rocky outcropping that we feared would stop the signal on the line to the school, but didn't, to the antenna at a crazy angle on the school's roof, down via too long an RF cable into the 'communication' room with fancy, costly LAN gear put their as a grant by MCI, into the *fourth* FreeWave radio, blinking green, downthe serial cable into our linux router, THEN via an ethernet conection into a patch panel into the school's NT 4.0 Router (which is the 209.94.82.50 address) which was bought by the US West fine money, and/or to any of the Pentium Gateway workstations, with DHCP address assignments, in the classrooms, for the little brown faced kiddies of 'El Valle' to use to access the net as readily as do the affluent kids of Cherry Creek High School in Denver. (and at a hell of a lot lower cost) The returned pings offer proof of the reality. Its not all done yet, tweaking of the routers, radios, and packet switching TCP has to be done before its ready for full web prime time. But the links work. And how do I know its *exactly* 30.2 miles from the Amigo Net tower in Alamosa to the 100 foot tower in the wilderness on the Stations of the Cross and Chapel ridge? Because I used my brand new Garmin GPSII Plus Global Positioning hand held gizmo, that accessed NINE military satellites to give us location to the nearast 15 meters, at the two points, then, with a GOTO button calculated and presented on its tiny screen the distance between. Whew! This has been a two year journey to get this difficult-but-important-for-proof-of-concept digital radio link working. Its such an electronic equivalent of the spiritual Penitenties of the devout Hispanics of the San Luis Parish, and their ancient rituals, I will be moved to write it up as 'The Saga of San Luis' or "How to Link a very old Community to the most Modern" one of these days. Almost with poetic justice, we ended the final tinkering after dark, in the radio shack with no lights but our flashlight, in a chill wind, up on the bleak ridge, damned near getting stuck when we high centered a rock with still 3 hours of driving to do. But at least a line of it WILL make the draft of what I was offered to do by letter from Scientific American I read when I got back - submit a 1,000 word piece for a next May Wireless issue, together with colleage Dewayne Hendricks. For which we will be paid $1 a word. Which, if this old wireless hacker gets no reward in Digital Heaven, at least I will enjoy some, at a proper rate for my scribblings, here on earth. Yep. You didn't think I was spoofing you did you? :-) (of course we have asked kids in the school to test by ping the route twice a day to detect any problems brought about by atmospheric conditions. There should not be. But hey, this is stretching the limits. Who knows what will happen next. The hop between 71.30 (our router in Alamosa) and 82.1 (our router in the radio shack on the ridge) is the 30.2 mile wireless link. Between 82.1 on the ridge, and 82.2 (our router in the school) is a 1 mile wireless link. You can only tell where the slow-down is by doing a traceroute where the milliseconds of time between hops tells you whats fast, whats slow. Ever since Ken Sweinhart of Alamosa dumped Rocky Mountain Internet - where he was operating pretty much as franchisee, and went on his own as Amigo Net, with different upstreak circuit (T-1) providers, going to/from Amigo net has been clearly much slower. Then when you add the radio links and three routers and an NT to the mess, I observed some 400ms total times yesterday. But found that over half of it was from slowdowns in the fiber portion of the total route. Which reminds me. When I asked the chief network tech for big School District 11 in Colorado Springs to comment on the relative 'reliability' of the wireless 2Mbps link we have had up for 8 months between two highs schools and the US West T-1 links he said (1) the radios operated faster than the US West T-1s (2) *AND* the radio links were at least as, sometimes MORE reliable than the telco's! That was a suprise. Radio is supposed to be unreliable, right? Hell, here is the guy responsible for the $50,000 a year T-1s between the 55 schools, and he is saying those one-time-cost radios perform out in the Colorado snow and blow (rooftop antennas with wind and weightloading on antenna towers up to 30 feet high) as well as the supposedly 'perfect' buried fiber cables. And as if to underscore the point, the little 1/20th of a watt radios we are using to link oldcolo.com to csn.net weathered the latest severe storm (where we had 3 rooftop antennas) without dropping a packet! And none of our ISP customers were inconvenienced. I think I better start a new marketting motto for the wireless companies. "Get reliable connections to the Internet - go Wireless"

Ok: !ping 209.94.82.50 PING 209.94.82.50 (209.94.82.50): 56 data bytes 64 bytes from 209.94.82.50: icmp_seq=0 ttl=106 time=647.803 ms 64 bytes from 209.94.82.50: icmp_seq=1 ttl=106 time=292.1 ms 64 bytes from 209.94.82.50: icmp_seq=2 ttl=106 time=292.988 ms 64 bytes from 209.94.82.50: icmp_seq=3 ttl=106 time=275.659 ms 64 bytes from 209.94.82.50: icmp_seq=4 ttl=106 time=299.057 ms 64 bytes from 209.94.82.50: icmp_seq=5 ttl=106 time=276.067 ms 64 bytes from 209.94.82.50: icmp_seq=7 ttl=106 time=278.422 ms 64 bytes from 209.94.82.50: icmp_seq=8 ttl=106 time=289.177 ms 64 bytes from 209.94.82.50: icmp_seq=9 ttl=106 time=269.378 ms ^Z [1]+ Suspended bbs bash$ traceroute 209.94.82.50 traceroute to 209.94.82.50 (209.94.82.50), 30 hops max, 40 byte packets 1 aus-c21.ddc.net (206.97.234.1) 5.712 ms 3.865 ms 3.786 ms 2 sat-c41-sl6-aus.ddc.net (208.128.48.5) 8.222 ms 8.41 ms 8.205 ms 3 border1-serial3-5.Houston.mci.net (204.70.36.65) 134.656 ms 90.594 ms 28.904 ms 4 core1-fddi-0.Houston.mci.net (204.70.2.97) 15.37 ms 27.364 ms 15.919 ms 5 core3-hssi-1.Memphis.mci.net (204.70.1.34) 26.067 ms 31.074 ms 26.042 ms 6 core2.Denver.mci.net (204.70.4.193) 60.408 ms 60.843 ms 69.327 ms 7 border8-fddi-0.Denver.mci.net (204.70.152.67) 65.087 ms 63.042 ms 69.13 ms 8 rocky-mount-internet.Denver.mci.net (204.70.156.6) 64.885 ms 62.499 ms 60.883 ms 9 166.93.35.1 (166.93.35.1) 63.818 ms 64.249 ms 65.868 ms 10 166.93.2.134 (166.93.2.134) 128.994 ms 128.527 ms 121.33 ms 11 * 209.94.71.30 (209.94.71.30) 149.759 ms 126.43 ms 12 209.94.82.1 (209.94.82.1) 217.643 ms 191.592 ms 201.265 ms 13 209.94.82.2 (209.94.82.2) 265.856 ms * * 14 209.94.82.50 (209.94.82.50) 235.072 ms 248.722 ms 234.67 ms bash$

That's groundbreaking news from Dave Hughes.

More from Dave: Wireless DC COnference May 4,5 Started by: Dave Hughes (dave) on Wed, Mar 11, '98 0 responses so far EMERGING WIRELESS TECHNOLOGIES CONFERENCE AND WORKSHOP May 4th and 5th, 1998 Marvin Center, George Washington University Washington, DC Sponsored by National Science Foundation, IEEE, GWU, OCCC Go to www.gwu.edu/~cms/wireless or click on the top item in the http://wireless.oldcolo.com NSF Wireless Field Tests web site - for complete details, agenda, and online registration $295 if you register before April 1. $350 thereafter Includes 2 Continental Breakfasts, 2 Luncheons w/speakers Vigerous public policy, comparative economics, 3 world as well as domestic urban and rural, and educuational applications. Demonstration and displays of no-licence (no comm cost) spread spectrum Part 15 radios, newest Microwave technologies, new forms of satellite communications, the controversial issues surrounding the whole $2.25 Billion Universal Service Fund for school library telecom subsidies. And implications for the Virtual University and other educational instituions. Speakers include David Isenberg, author of the controversial "Rise of the Stupid Networks" in which 'smart wireless' will play a big future part in direct competition with Telephone Companies. Dale Hatfield, the Chief Technolist, Plans and Policy, FCC. Panelists wide ranging from Adam Powell (Freedom Forum) and Henry Normal (VITA) on 3d world connectivity (yeah, just call your local AT&T rep...), the MIT researcher who has proved we can have 'billions' of radios, not interferring with each other in the same space and communicating hundreds of megabits of data per second, expect to have the company now demontrating in Japan OC3 (that's 155mbps folks) radios no bigger than a dictionary there, 'distance learning' pros, Motorola and other newest satellite services folks. With a demo via satellite + ground Internet to tiny town in Montana, where 7th graders are doing real science and environmental analysis by wireless from the classroom to Big Spring Creek on Brewery Flats - field science by wireless being looked at for application to research labs and universities. And no holds barred discussions with government officials from Commerce, FCC, Dept of Education. Yeah, this aint your plain 'telco competition' and 'how to used a wired web page for Good Education' conference. Its the future, here now, and why policy makers and Congressperns otta be paying a lot more attention to these technologies rather than another self-serving tv ad by Sprint, CTIA, AT&T, or MCI... $295 including two lunches with speakers? You can't get that kind of deal in the largest hotels in DC, sans speakers. Press Passes. Dave Hughes dave@oldcolo.com

Dave Hughes (dave) Sat Mar 14 '98 (07:46) 24 lines Something happened while I was mosying around Alamosa before returning to Colorado Springs, touching base with the numerous people that have pioneering wireless down there. The lightbulb turned on in my head about what 'field science by wireless' project might be interesting in the continental US. What might be *very* interesting, exiting even, as well as being useful, would be to recruit the water, wildlife, environmental departments of a whole string of universities, AND schools, all of which are the closest to the 1,400 miles of the Rio Grande River from its headwaters near Creede, Colorado down past Del Norte, Monta Vista, Alamosa, La Jara, Antonito - in Colorado, then Espanola and all the towns it flows by or through in New Mexico, and then all the towns in Texas to the Gulf of Mexico - to put the kind of wirelessly connected sensors all along that watercourse that would produce useful data for the efforts being taken to preserve that 'Heritage River.' I could get interested in a sweeping project like that. On an interesting river that flows through the most interesting and remote places in the West and South West, and 95 or more percent of which is remote from all human habitation. What would it cost? A million dollars?

This sounds a lot like our packet system used to, only at mega speed! Wish I could get folks over here in the Brazos Valley interested again in packet. I'm gonna be telling off, but I remember seeing packet address listed instead of Internet addresses... Maybe we as Hams can agin rule! 73, Pau; N5XMV

Hams rule, indeed! As long as packet stays within certain frequency bounds and all agree to it, there is no problem with it. I just hate it messing us my listening to distant frequencies when I am DXing.

Hams rule, I'll go for that (says ka6atn)! Yep, the Internet kind of came along and made packet seem like child's play, for a while, we were the forerunners of the net with our packet networks.

Ok, time to wake this topic up again. Ok RF net geeks, go to www.tapr.org and join some of the special interest groups for APRS, Spread Spectrum, DSP, Packet Networks, HF Digital, and Linux. Then we can chat on all fronts. Packet can rise again - especially if the high speed network back bone links are established. Yea, 1200 and 9600 baud are real yawners especially when trying to digi to California. The goal is to get the high speed spread spectrum links established and have a ham network that runs parallel to the internet. I know the microwaves can handle the bandwidth. The gauntlet has been thrown down - join the SIGs and make it happen! de AA9IL P.S. Holy Toledo - a posting on this page from none other than Senior George!

Its been a year since the 'wake up' post so here goes again... This time, I hope there is some followups to this thread. I have found a good url with links to many other interesting wireless net sites. There is a movement underway to bring wireless net connectivity to the grass roots level. Right now, the big problem with high speed access is still 'the last five miles' i.e. getting the high speed link from the CO (in the case of DSL) to the house. Due to my professional affiliation with the telecom industry, I can report that this bridging is progressing just a bit faster than a glacier. There are other alternatives such as cable, satellite, and wireless - its getting there but still is sloooow. In the true spirit of hacking (in the classical sense), ham radio, and the old computer revolution mindset (ala 1970's), people are setting up neighborhood high speed wireless data links that are covering that last 5 miles and then some. Check out the following: www.toaster.net/wireless/community.html These links will take you to work in progress pages of practical wireless development that is the epitome of 'free net'. 73 de AA9IL Mike radio cosmo international Hopefuly the next posting wont be a year from now!

Here's what the guy doing this in Oahu, Hawaii has going on: To tap into information at any time anywhere. I'd like to roam the island and be able to work at the beach with internet access cheaply. Please contact me if you would like to help me expand the network Maybe a quickcam hooked up to the telescope so you can see who is at the beach ;) . Intial Target is behind the red Truck- thats the beach. The land further out is flat island and I may get out there cause, no people, and thats where the surf is at baby , makes for a horid coffe break eh? :) This picture is zoomed in and actually things are abit further than they appear - will probably have a cam with me on the beach when I start working from there when development on project I'm slated for starts - So check out the webcams page for that when its ready. The home base transceiver and a close up Currently testing horizontal and vertical dish positions. Dang trees dang ugly trees in target zone... hope they mysteriously fall down in the night. Oh Ya, thats my 74 Bronco sport with no doors :0 - twas a very good year. The setup - Laptop card/dish and Homebase cardboard box PC and card - outside dish. When one doesnt have a drimmel tool or know what they even look like, just use the serrated kitchen knifeand some good ol electric tape. I wouldnt exactly consider these weather proof though :). Dont need to take off whole end just 1/4 inch hole from top where you see circle - thats what I did on the other card to the right and soldered with silver base there to J2 connector. Later on the layout will be repeaters every x amount of miles. Currently the cost is about $300(400?) - 140- towards 2 network cards - and $150 towards antenna. But the aviator 2.4ghz cards arent 10/100 there max is 2 and also their power output is only 100mw so distance needs to be improved with a amplifier (better cable would help too - a build it your self kit is shown here maybe to complex for your average joe though . So considering switching over to what seattle wireless is using - Hrmmmm - what could I use for a trans pacific line to them???. Some things I need to look at for next steps omni/directional antennas and how best I can utilize them. Current parabolic are ok but really end product should be omni directional. A cheap low power pc/repeater (ala crusoe or risc - motorola chip) with pcmcia card attached powered hopefully by solar panel- Ie would like to stick this on top of the mountain range (ie climb stairway to heaven ladder and leave something up there on the ridge :) ) a cheap pc/repeater but is a wearable basiscally same as above but for the man on the go http://www.equip2rip.com/wirelessnet.html Interesting stuff, I'd like to set one up out at the Cedar Creek ranch.

I also contacted this guy: If anyone is interested in starting a similar wireless Idea in the Austin, TX area, please email me at hem@marlabs.com, I have already registered austinwireless.net, so let's boogie! I will be able to provide a similar web hosting for free, and also help with providing a location for one transmission point. Hem Ramachandran

How far to the beach from Cedar Creek?! Surf's up!!! That guy from Oahu needs spell-check. That isn't even good pidgin, but I think it is fascinating in possibiities!

Howdy Howdy Checked out the Oahu web page - now, I like the idea of working on the beach. Screw the work, just be on the beach. Anyway, the wireless card to 2.4ghz dish was a good hack. I did some checking around on broadband access in my area. ATT, Sprint, Roadrunner, etc - no service yet despite the fact that this is one of the fastest growing areas near chicago. I guess if I wanted high speed access NOW, I could shell out the bucks for a satellite link but there is something fundamentally wrong with that. Until then, the closest high speed access is the T1's at the university. A couple of the web sites I found for free net type wireless work were in areas a couple of hundred miles away. It ends up being either wait and wait or DIY. The DIY option is more fun. 73 de AA9IL Mike r c i

The guy who's doing the Austin wireless net called me, and he's sending me his url, apparently his website is temporarily down.

From: Hem Ramachandran To: Paul Terry Walhus Subject: RE: austinwireless.net Hi, Paul, Austinwireless.net is hosted free right now. What it need is active participation of Austinites and have our goal for a well networked Austin. Let me know if you could be a part in that front. Hem

Howdy howdy I remember seeing a posting on the austin wireless network. Although I never had a chance to try them, it seems similar to the free nets and Fido net. Anyway, I hope that there is signicant progress bringing free wireless to the areas. Most of the inexpensive wireless apps are in the ISM (industrial scientific medical) bands around 900 Mhz and the shared part 15 spectrum around 2.4ghz (which is also the satellite mode S downlink frequency). Most of the real high speed commercial point to points is up in the 23 and 39ghz - look on the roof of any high rise building in chicago and you see a cluster of 39ghz dishes. Not sure if there are any part 15 allocations up there. Look forward to hearing the progress of the Austin setup! 73 de AA9IL Mike r c i

Hi, Paul & Mike, To comment on your posting, unfortunately there hasn't been much interest in the Austin Wireless front. I have registered austinwireless.net and am still waiting for a few interested souls to come by to have this started up. The idea is similar to Seattle Wireless Network. You can visit the preliminary pages at austinwireless.net. I had 4 people respond, seeing the site. A couple of them have mysteriously disappeared!(dot com crash?) Anyway, if there is good interest, we can all come together to start cracking on it! Hem Ramachandran KB5LQD.

Let's talk about getting started, what would I need to set up a basic household wireless network, since I live way out in the country and have a big spread out place. I'd like to cover a range of about 400' max., I could probably get by with about 300' really. What equipment and software would be required, Hem?

Greetings Hem and Terry Regarding local work, there are some good 2.4ghz node transceivers described on the Seattle wireless page. Believe it or not, there is even some experimentation going on with the Cybiko handhelds to convert those into simple wireless terminals for APRS work. I picked up one for $75 - although its a toy, it does something similar to what 'real' wireless data links do. As far as the Austin area, Hem, where would you get your T1 access? Would you try to get some dark fiber off of UT, St Edwards, or ACC? I figure these would be the place to look but the again, UT might not be giving away free bandwidth. Im amazed that more of the tech-geeks have not jumped on the bandwagon. Although its been ten (!) years ago, some of the folks who ran Discovery Hall and the Robot Group would seem like the group to jump start things. Also Fringeware but they vanished as well. Anyway, regarding hardware, it will primarily be Linux boxes running to data modems that follow the IEEE standard. The only minor crux is that these boxes are part 15 FCC type accepted which means low power with the intent of local area use only. In order to get to the power levels for city wide networks, you either need height and some power or lots of micro cell type nodes linked together. You can get around the power problem by building up simple repeaters. For the minimum amount of microwave hacking, then the microcells is the best but then you have the network infrastructure to worry about. The compromise would be a wider area coverage micro cell with the low power modems running to reasonably efficient omni type antennas with high gain directional antennas for the links. Anyway, food for thought. Hem, although Im not in Austin, (Austin in spirit only), I would like to keep up with the activity and throw out any microwave 2 cents if it helps. 73 de AA9IL Mike r c i

Paul, Here is some links for you. Hope this helps better than if I try on this forum. http://consume.net/press/PR001.html http://205.159.169.11/ http://www.sflan.com/ http://seattlewireless.net http://www.pbs.org/cringely/pulpit/pulpit20010628.html

Hi, Mike, The Austin network is still in the prelim stages. We don't have a T1 connection, planning on using a DSL connection. It gets upto 1.5 Mbits/sec, so that should be okay (I am real surprised that SWB doesn't throttle the speed for a 384 K connection!) as far as connectivity is concerned. The main idea is go the Orinoco Way (http://www.orinocowireless.com/products.html?section=m56&envelope=92) Regarding roof access you mentioned, when talking to Collocation Inc in downtown Austin for our company servers, they mentioned they will give free access to the roof for customers, we were thinking about moving our servers over there, but the current economic situation warrants us to be more cautious. That would have been a good starting point. But home based ones seems to be the best way to start off with, using the products mentioned above. More opinions? Hem

I got a great link from Donn Washburn about converting now obsolete Dish Network dishes. http://www.wwc.edu/~frohro/Airport/Primestar/Primestar.html "Primestar was recently purchased by Direct TV who is phasing out all the Primestar equipment. This means that the dishes are being trashed, and are available for other uses such as the one I describe here. It is easy to make a surplus Primestar dish into a highly directional antenna for the highly popular IEEE 802.11 wireless networking. The resulting antenna has about 22 db of gain, and is fed with 50 ohm coaxial cable; usually LMR400 or 9913 low loss cable is used if the source is more than a few feet from the antenna."

Wait till I tell my son about that one. I love the can...

Howdy Howdy More tin can trivia! A three pound coffee can is also a resonant feedhorn for 1.2 - 1.4ghz A tomato paste can can be used at 3.4ghz This is the cheap and easy feedhorn for the microwave crowd. For a full description on using offset feed dishes and different types of feeds, do a google search on N1GHZ - that page has a whole section to working with dishes and different types of feeds. 73 de AA9IL Mike r c i (currently engrossed in packaging a 10ghz transverter...)

http://www.pbs.org/cringely/pulpit/pulpit20010712.html > >Remarkable! > >Bootleg 802.11b > By Robert X. Cringely Two weeks ago, I wrote about my adventure building a high-speed 10-kilometer wireless Internet connection using off-the-shelf 802.11b components and a telescope. I live in a very rural part of Sonoma County in the California Wine Country, and this was the only way I could find to get my bandwidth fix without the unacceptable latency of a satellite connection. Well, that column was so well received that I got e-mail from all over the world asking for more information so people could try to do it themselves. That is the purpose of this follow-up column. Understand that I am an idiot, and so I only know what I have done and not much else. My shallowness of knowledge in this area is epic, but if I can make it work, so can you. The hardest part of the job, in fact, is knocking on doors telling people you've been looking at their house through a telescope from across the valley. For all the tens of thousands of people who read this column each week, I was amazed to learn that almost n>obody knows that the ">I> >L>ike It" button on the right side of the page leads to five Web links associated with each column. If people had known about those links two weeks ago, most of their questions would have been answered already and I'd be trying to think of something else about which to write this week.

One question that kept coming up was about the legality of such a bootleg connection. A lot of armchair lawyers with nothing better to do told me I'd soon be headed to jail. Well, this is absolutely not so. Regulations vary from country to country, of course, but in the United States, what I did is perfectly legal under the Federal Communication Commission Part 15 rules, specifically sections 5 and 23. The rules are simple. Just don't build these gizmos and offer them for sale, and don't build more than five of them that are identical. No testing or pre-qualification is required, which was probably a surprise to the lady from Metricom who lectured me ad nauseum (and inaccurately) about this only hours before her company declared bankruptcy. For homebuilt components like these, the standard is a simple one of non-interference. Whatever mischief you do should not interfere with baby monitors or garage door openers. And the very sensible way the regs chose to validate non-interference is through a lack of

complaints. So if nobody complains, you are okay. If somebody complains and you are able to satisfy them by changing frequencies or modifying your antenna installation or just buying them a beer, then you are again okay. Nobody goes to jail. Many people asked about antennas. I bought mine from a local ham radio dealer, and they are Cushcraft yagi antennas, and fairly expensive at $249 each. Were I to do this again, I probably wouldn't use the Cushcrafts. This is not because they don't work, but because there are cheaper alternatives. True, the Cushcrafts are sleek and unobtrusive with their all-weather ABS enclosures, but with homebuilt alternatives running $10 or less to build, it is hard to justify paying the big bucks. In the links associated with this column (under the "I Like It" button, remember?) are directions for making your own yagi antenna from Pringles potato chip cans and for converting old DBS satellite antennas (like DirecTV, The Dish Network, or Primestar) into high-gain 802.11b antennas. These directions require the use of one or two large juice cans and carefully instruct that you must first drink the juice before using the cans. Ten bucks gets you not only the antenna, but also the juice.

Most questions had to do with my use of Apple Airport Base Stations. This is one area where your results may vary -- not because of technology, but because of business decisions by companies who make this stuff. Apple Airports don't immediately connect to each other, for example. You can link computers with Airport cards to Base Stations or to each other, but not Base Station to Base Station. There are two ways to overcome this problem. If you are using Airport software prior to version 1.2, there is a lucent hub manager program that can reconfigure your Base Station to act as an Ethernet-to-Ethernet bridge. If your software is version 1.2 or later, then you need a $50 firmware upgrade from Karlnet.com that accomplishes the same thing. Now there is some controversy about this firmware upgrade, which seems to be only intermittently available. Some people claim it is no longer available. Some people claim it is available only to ISPs.

Hi Terry and gang As soon as I post this, Im gonna go to the link and check it out. Last week I did a bit of cold calling to various satellite TV distributors in the Chicago land area to see if I could score some free stuff - no luck. I think most of them are new item dealers and if they remove a system for a new DBS system, its up to the owner to haul the dish to the curb for the trash pickup. In light of this, the larger ham fests are the next best thing to find the dishes and components - usually at a very reasonable price since the stuff is useless for the entertainment consuming masses. Also, I think if you stay within the confines of part 15 then you are reasonably safe. I have seen some postings on the free net wireless reflector of building amps which will probably go way beyond part 15 (my 1 watt 2ghz amp would ...) I guess caution and common sense prevail. Anyway, future comments once I explore the link and do a bit of reading up... 73 de Mike r c i

Here is a message from Jess. About Wireless in Austin. I hope he doesn't mind I posting it here.You can contact him at Jess Jess wrote: Okay, now that i have gotten a response here are some of the specifics about my plans. I have a 30+ foot antenna tower at my house that i have not had a chance to put up yet. When i get it up i will be tower mounting an old laptop of mine that will be running linux and have an 802.11 card and a packet interface to either 1 or two radios. I will also have a direct high speed link using either 802.11 or older wireless cards using an old primestar dish modified for the frequency of the cards i end up using. I am not currently sure where the high speed point to point link will be going because i am not sure if any of the people i have to link up with are close enough to being line of sight . for the 802.11 I will be building an omni derctional vertical colinear antenna of some sort. For the packet interface I am thinking about using MURS radios as they will provide much longer range than the 802.11 (atleast a few miles), they will only be able to do about 9600bps most likely (well, easily and legally, there is a possibility of getting upto 19.2 but requires some modifications to the radios and things which would not be legall, i shave no problem with doing the further research into this area that is required to do it though) which is slow but is beter than no link and would still be enough to let you do things like check your email. For even longer range omni directional links we could use packet with ham radio's but this poses some problems. Just for the record i am a licensed ham radio operator. The main restrictions with using ham radios is that it is not legal to encrypt anything over the link, have any profanity or use it for business purposes, so we would have to make sure no traffic was routed over these links except that of other ham radio operators, which could be handled in multiple ways and i do plan on putting up a ham radio link. I am located Just off of south congress near oltorf so with my tower i should have a prity good view of downtown, which could provide lots of opertunities. there are numerous things i need to do before the plan for my above node can become reality, -Obtain murs radios(2 1 for node one for me to connect to it), the ones they sell at radio shack list for $150 a piece but are supposadly constantly on sale for $100 -Obtain ham radio for node, I already have one to connect to it with just need one for the node, should be able to obtain for no more than $150, i am thinking about using a 5 watt 2 meter HT(Handheld Tranceiver) of some sort because that way it will have its own battery backup and the node itself will be able to stay up for a long time(atleast 5 hours) with no external power, that way even if we loose power our network can still be up and running. -Figure out best way to tower mount laptop and radios that will keep them protected from the weather without causing them to overheat. This shouldnt be to hard, just need to get the time to experiment with some things and get down to the hardware store. -actually put up my tower, should happen before to long -get the software all setup. The software will be slackware linux with wireless tools, ham radio tools, some custom scripts and possibally some other things (such as mobileIP) -find somewhere to link to, i do not currently have a permanent internet connection so i need to find someone to link to wirelessly. this is just the plan for this one node, i have plans and locations for other, just need to obtain the equipment, if you know anyone with old laptops they could donate, any ones that are semi broken like have a broken screen or keyboard will work just fine and are perfect since they are no longer good as laptops, or any types of radios or wireless cards,doesnt mater if they are 802.11 or not, start collectiong them, they will be very useful. One other issue that we need to discuss is routing, Ideally all nodes will be able to route even if they are not permenant ones or are mobile, i think i have figured out how to deal with this with just software that is available, well ive gota go so i will continue this email later. tell me what you think and feel free to forward this to anyone that would be interested. Jess

Intel will ship new wireless home-computing products Wednesday that support a wireless technology the giant chipmaker once competed against. Intel early last year became one of the first companies to sell technology that allows consumers to wirelessly connect their home computers and share the same Net connection. The company supported a wireless standard called HomeRF that was backed by wireless technology provider Proxim as well as giants Siemens, Motorola and Compaq Computer. But with HomeRF support sagging, Intel five months ago announced it would switch its support to 802.11b, or Wi-Fi, a competing wireless standard backed by Apple Computer, Dell Computer, Cisco Systems, Lucent Technologies spinoff Agere Systems and many others. Intel on Wednesday will announce new AnyPoint Wireless II Network products based on 802.11b that let people wirelessly link their PCs and laptops so they can share a Net connection, files and computer peripherals such as printers. The products will allow laptop users to roam around a house and surf the Web. Intel previously touted HomeRF as cheaper than 802.11b, but prices of 802.11b products have dropped considerably in the past year. Intel previously sold 802.11b products aimed at businesses and HomeRF products aimed at consumers. But Intel executives say supporting one standard will allow workers to go home and have their work laptops easily connect to a wireless home network. Analysts said the rift between companies in the home-networking market has caused a standards war similar to the VCR technology battle that pitted VHS against Betamax in the early days of videotape machines. But analysts say Intel's support for 802.11b gives the technology the edge to win out as the standard in the home. "Intel's name carries a lot of weight for consumers and that's significant," said Parks Associates analyst Kurt Scherf. "Industry support for 802.11b has been tremendous." During the past few months, HomeRF has needed to show new support from companies beyond its three main backers--Motorola, Proxim and Siemens--and it hasn't happened yet, Scherf said. "They need to prove that it isn't obsolete." Intel's AnyPoint wireless products come in two forms: a tiny wireless device that can be plugged into a desktop computer, and, for laptops, a wireless PC card. Both, available immediately, have radio transmitters and receivers built in. Intel will ship in the coming weeks a third device, called a "gateway," which connects the wireless technology to a regular Internet connection. Intel spokesman Tom Potts said the company's new wireless technology features software that makes it easy to install the home network. The product also features built-in security. "We're trying to mask all the network complexity behind our software, so in most situations, the person installing it only has to answer three questions," Potts said. Intel ranks fifth in the market for wireless networking products with 5 percent of the market, according to a survey by market analyst firm NPD Intelect. In the first five months of this year, Linksys captured 28.3 percent of the $65 million spent in wireless networking, followed by Agere, U.S. Robotics, SMC and D-Link.

Microsoft will hit Palm with a double whammy next week when it announces two new versions of its operating system for handheld computers, sources say. The software giant will announce a low-end and high-end version of its upcoming Pocket PC 2002 OS on Sept. 6 at the Demomobile conference in La Jolla, Calif., sources familiar with company's plans say. The new versions of the OS, code-named Merlin, will resemble Microsoft's upcoming Windows XP desktop OS and add 802.11b wireless networking capabilities and security. Microsoft representatives declined to comment. Although one source said the new versions of the OS are "incremental" upgrades to the current Pocket PC OS, another said they are exactly what Microsoft needs to stave off Palm from increasing its presence in the emerging corporate market for handhelds. Palm's OS shows up in handhelds from Palm, Handspring and Sony. Microsoft's current Pocket PC runs devices from Compaq Computer, Hewlett-Packard and Casio. The enterprise market is the brass ring that all companies in the handheld industry are reaching for. According to research data, Microsoft has the lead in that market. Palm maintains its lead in the overall handheld market, which is still largely made up of consumer purchases. But lately Pocket PC has been coming on strong in the small but growing corporate market because of prior relationships with business customers, according to analysts. The corporate market for handhelds has the potential for massive growth because businesses tend to buy devices in large volumes, compared with the single units that consumers purchase. "Microsoft has an advantage over its competitors given that they and their partners, such as Compaq and Hewlett-Packard, have relationships built up from previous business," ARS analyst Matt Sargent said. "They are a natural fit to sell into that market." Sargent added that corporations are more intrigued by Pocket PC-based devices because of their ability to expand beyond basic handheld functions, such as storing appointments and addresses. Santa Clara, Calif.-based Palm has been trying to build a viable strategy to attract corporate customers. In late June, Palm announced several partnerships, the biggest being with accounting giant PricewaterhouseCoopers. And earlier this month, Palm announced it will acquire the intellectual property of Be. That acquisition is expected to help beef up Palm's OS with the multimedia and communications capabilities of the Be OS. Palm was also set to acquire a mobile data management company called Extended Systems. But the deal dried up in mid-May. Microsoft may beat Palm to the corporate punch, however with the new versions of its OS. The Redmond, Wash.-based company is making an aggressive push for Fortune 500 companies, according to sources, and the upcoming OS will have more of an emphasis on business applications than the current version of Pocket PC. Microsoft apparently hasn't overhauled, but rather added specific features that corporations have been asking for, such as wireless networking capabilities and security. Enterprise customers, and specifically IT managers, tend to want gradual upgrades to avoid the hassle of making major upgrades to their networks and PCs that they support. However, Pocket PC will likely go through a major revamp if it adopts the next version of the Windows CE OS code-named Talisker. The current Pocket PC is based on Windows CE, but Microsoft representatives have declined to comment on whether Pocket PC eventually will use Talisker, which is due for release later this year. Although details regarding the differences between the two versions of Pocket PC 2002 are not clear, sources say, the two are expected to vary considerably when it comes to storage capacity. The low-end version is expected to target handhelds with 16MB of memory, while the high-end version is expected to work with devices with at least 32MB of memory. The new versions of the OS will also have software drivers to support the addition of 802.11b wireless networking cards and will include virtual private network software to ease the security concerns of IT managers, sources say. The two versions will also have a similar look and feel to Windows XP. Pocket PC 2002 will also allow Outlook users to store their e-mail on their handhelds, so they can view messages even when they aren't connected to the network. Sources say HP and Compaq will soon announce new devices that run on Pocket PC 2002