2502 West Colorado Ave., Suite 203 · Colorado Springs, CO 80904 ·
Voice 719.636.2040 · Fax 719.528.5869 ·Wireless Web http://wireless.oldcolo.com















Progress Report 5

May 2d, 1996

Progress is accelerating.

1. MONTE VISTA

The Monte Vista wireless link, which we now measure at 14.5 miles (with GPS) between the omni roof top (above 2d floor) antenna at the Alamosa Rocky Mountain Internet POP, and the omni antenna on the roof of the one-story Monte Vista Middle School, is solidly operating.

There was a problem with the Monte Vista radio going out, which seemed to be caused either by spikes or outages in the power line near the router inside the classroom, which caused the system to reset (which on a FreeWave radio puts it into the 19.2 configuration mode), or by a pair of the extended indicator lines touching each other with similar effect. After rearrangement, the radio has worked continuosly without a problem, as had the other FreeWave inside RMII's office in Alamosa all the time.

Since the Monte Vista radio is physically mounted high up, outside, on the omni mast, in order to reduce RF loss, it is important that the status indicator lights be brought right down to a panel next to the Router in the control room so that the local tech, or teacher, can determine the radio's status by looking at the lights. This was planned all along, and Bob had made the drops of wires to the control room, but the lights were not configured yet, so it was impossible, by telephone, for Lori Roberts, the quite capable tech-teacher at the Middle School to report any status. The radio simply didn't work. The only 'proof of link' is to ping from the router to the other site.

The link is strong enough that Dewayne was able, using his MacIntosh notebook computer to receive the digitized voice broadcast of National Public Radio via RealAudio Internet technology. The voices were clear, and superior in quality to my ear, to the same broadcasts coming in via a 28.8kbs phone modem link to the net. An indication of brisk link speed. Dewayne had set the radio to the 'lower' speed - to get longer range - which is not marked, while the upper speed is 115kbs. At one point he measured 68kbs on the link.

We were aware that this was in spite of the fact that Rocky Mountain Internet is still having irritating problems with their reliance on US West keeping two 56kbs data links up from Pueblo to Alamosa - which, load balanced - provides the total thruput to RMII's POP (used by dial up accounts).

The radios being used were the latest (March, 1996) upgrade of FreeWaves basic radio, rated at 115kbs, 1 watt. The most important part of the upgrade was to raise the power from .3 watt to 1 full watt, because of better reciever design. It is these 1 watt radios, with better receivers, that not only is giving us a robust link 14.5 miles, with Omni, not directional antennas, and without the originally planned 5 watt antenna 'booster' at the POP location, but also, in Dewayne's opinion, will permit a direct, not relayed, signal from the Center School location, 22 miles, to the POP in Alamosa.

The total Internet link to the 25 Windows for Workgroup 486 workstations in one classroom at the Middle School is thus complete and reliable enough that, not only can Lori Roberts, the school tech/teacher use it, but also the three math science teachers taking the course from Dr. Johnston over the net can telnet into RMII on their own accounts, use network Netscape for web access, ftp to and from the workstations.

I instructed Lori to (1) start keeping a log of their experience each session, with date/time of their session, whether the link is up or down, any problems with it, and their subjective reaction to whether it seems fast, slow, or indiffernet. (with their experience accessing the same resources by modem, they now have a basis for an 'informed' - not measured - comparison. We will do the measuring technically) (2) to have each teacher and herself Telnet into wireless.oldcolo.com, which gives them access to the new Wildcat 5 NT 'BBS' software running on the NSF Project Server. And log in. Lori herself did that May 2d, following Dr. Johnston, from Massachusetts.

We recieved the discouraging news that Lori Roberts, the tech and teacher we have patiently trained from the beginning of the project on (a) the art of lans, NTs, and workstations (b) access to the Internet, pop mail, maillisting, the multi-media add ons to her workstations, graphical file production, transfer, and downloading from MIME attachments - to support the math and science graphics necessary for the course and finally (c) the wireless link, and operation of the special router Dewayne has designed for test control - that she will not be back next year. She has taken a good 'network' postion in industry.

2. CENTER SCHOOLS

More problems. Although our project Router is fully installed and integrated into Center's extensive new lan network, with a large number of installed workstations, and several multi-media centers, the modified FreeWave radio high on the mast on the roof (and chimney) of Center School District's main building, failed somehow. It has to be brought down, which takes two men on a shakey ladder, and examined.

Bob asked for voltage readings from the leads on the dead radio, which Dewayne has procured.

The teachers math/science course is proceeding on schedule, getting three teachers well acquainted with using the Internet, which experience and knowledge will be valuable when we resume using the wireless link in the classrooms in the fall after school resumes. And Cindy, the new tech, who has had her hands full mastering the new LAN network in the school, is able to resestablish Internet TCP/IP modem links to the Router - and teacher workstations - when it goes down.

3. SAN LUIS

The Router is now installed at San Luis Centennial School, linked to the replacement Zoom Modems which are operating better on all systems, than the Supra 28.8 modems we first used (which failed to have a non-volatile ram way of saving settings - the company has removed what was once there.)

I was unable to get the LAN link from their OS2 Warp Connect working before the 2 May weekend, so the server <-> router <-> modem <-> pop link is not yet working as we had operating at Monte Vista and Center in preparation for the radio replacement of the modem link to our pop router.

It didn't help to have all our special telephone lines in Alamosa go out while we were working on them. They are back.

Also, even before we reach the wireless achievement to the Centennial School District which, in the small town (800) and small district (300) also is accompanied by a lesser concentration of technically trained people, we hear that our school tech in San Luis also may not be back next year.

4. The probable loss of two key persons at the school district level simply underscores for us - which we have communicated to the School Superintendents - that, if wireless links to the Internet are to be installed and supported, the schools are going to have, or develop techs who may also be teachers, but not continue to rely on teachers, who also may learn a little tech. It is not at all probable that a school district with many workstations, lan networks, tcpip capable servers, wireless (or wired) routers and full access to the Internet, as well as supporting local dial-up to their server (BBS or Web Server mode), simply will never get the most out of all this technology and telecommunications unless they have a very competant technicians managing all this equipment, software and networks.

5. The three Tetherless Access systems, running modified Cylink radios have been receieved and are being readied for deployment in Colorado Springs, with links from Rocky Mountain Internet's downtown, high-rise building POP, to our Old Colorado City Communciations building (three stories) premises, and from thence out to West Junior High School.

6. Solectek Company is configuring three of their Airlan 2000E radio-bridges, routers, and antennas for shipment to us for installation at the Mitchell High School site in May. These are their powerful 25mile, T-1 rated systems.

7. Dewayne Hendricks has been able to deploy, in the San Fransisco Bay area, a pair of the older FreeWave wireless modems (.3 watt, but with 1 watt antenna amplifiers, from his Internet Linux server in Fremont, 22 miles across the San Francisco Bay to link to a Macintosh Mach 10 (Unix) server in San Mateo that supports an Appletalk local area network. Using 9 element yagi directional antennas with about 10db gain. The significant finding here is that these radios are operating at this extended distance in a dense metropolitan area, in the same spectrum space (902 to 928mhz) as are the Metricom Metropolitan area network Ricochet and pole top cell radios that cover both Freemont and San Mateo. i.e. there should be severe interference here if no place else. Thruput tests of performance will be done later. The first configuration of the radios is not optimum yet.

8. I have had a pair of earlier-model Tetherless Access systems operating from my home-office to Old Colorado City's office, a quarter mile away, through two large buildings, for over a year. They have operated well, at the rated 160kbs speed, connecting to the Internet. These radios have been using the Cylink radios at .8 of a watt, using directional antennas at both ends - though they have been tested down to 1 milliwatt, and one end uses a small rubber duck omni antenna. We have now replaced the TA radios with a pair of .3 watt FreeWave radios, which are operating well over the same link, using the same antennas. The speed is noticably lower (rated at 115kbs versus the 160kbs of TAL systems), but are quite satisfactory. Measurements on these will be done also.

9. As summer approaches, we will soon learn to what extent foliage on trees on the line of sight path between the radios will affect performance.

10. The NSF Wireless Field Test for Education Web site is fully configured at .


Dave Hughes

Next