Internet Radios Tests at Oyster

On November 19, 2001, PI Dave Hughes and I met with the staff of the VCR/LTER for some assessment, strategy, and planning. Present were Dr. John Porter, Co-PI for the site; researchers Randy Carlson and Phil Smith; and Dr. Raymond Dueser of Utah State University.

Since the 'Oyster' wireless project was a totally new project - extension to a 4th site after Alaska, Wisconsin, and Puerto Rico, as PI with a fixed budget, Dave Hughes wanted to be sure just how far our project could purchase radios and associated equipment, install them, and how far John Porter's LTER project could support it.

And since Dave Hughes planned to let Tom Williams do as much of the work on the Oyster site as possible, he wanted to see just what new ground would be broken.

The first new matter was a substantial 'cross water' link. The second was the effort to make that link support video on the island. And the third was, in contrast to the other projects, much more emphasis on biology and wildlife data gathering than on environmental factors.

While it was not clear from the meeting what we could do, Dr. Dueser's work with Raccoons on the island - the desire to track them continuously - was a future challenge.

By the end of the dinner meeting it was clear that the Oyster project could expand in many directions, which could take many a year. We would have to set priorities. But it confirmed the need for a high speed - 1mbps at the very least, for the backbone link to Hog Island. From which all else could flow.

On November 20 we tested two IP-ready radios -- the NovaRoam 900 and the WiLan Hopper Plus model 22-09.

NovaRoam and Hopper radios

Comparison of Features

The NovaRoam 900 is a true Internet router, able to isolate one LAN broadcast group from another. The frequency hopping radio, with a power output of 800 mW, can move data at speeds ranging from 20,000 to 1,000,000 bits per second, over distances of tens of miles at the slowest speed to its 1mbps speed at a quarter mile or so. As data rate increases, range decreases.

The potential value of these radios is in the true IP routing, in a higher power, lower frequency (obstacle penetration) radio.

The WiLan Hopper-Plus is branded as an Ethernet Bridge. Its 500 mW direct-sequence radio sends data at a fixed rate of 2,000,000 bits per second (2 Mbps). There are myriad options for testing and measuring link quality. It has lower output power than either the Freewave (1 watt) or the NovaRoam (800mW) at its 500mW, but is also in the 900mhz frequency range; thus it should handle vegetation better than any 2.4 or 5.8ghz radios in the field.

One team went out to Hog Island. The boat trip takes 45 minutes to an hour. The team took a 12v motorcycle battery for the test, rather than a heavy full sized vehicle battery (to haul up the tower by rope), 6 foot LMR400 cable, connectors, antennas, radios, all in our waterproof Pelican Case. We wanted to be very sure no moisture got in the connectors we would be connecting and reconnecting between radios and antennas. That alone could invalidate the tests. Fortunately both the NovaRoam and the Wi-Lans can use the same Reverse TNC adapters on the cables with standard N Connectors on the other end.

It remains a big pain to deal with the FCC requirement that the connectors on Part 15 Radios be 'proprietary' instead of standard. Cables take more work than all the radios and antennas put together.

LTER Project Boat

Range Test

We tested both radios over the 14 mile distance between the VCR/LTER headquarters in the 'Farmhouse', across Hog Island Bay to the Broadwater Tower, using the same antennas as we had previously used to test FreeWave radios (see Diary 38): 9 dBi Yagi antennas at both sites, plus a 6dBi omnidirectional antenna on Broadwater Tower. The weather was cool, windy, and wet, especially up on the tower, where Phil Smith worked with two radios. PI Hughes worked with the other two radios on the second floor of the farmhouse in Oyster using his laptop inside, connected by Ethernet, to produce pings, or read packet quality in the radios.

The Farmhouse that is the LTER Headquarters

The Wi-Lan radios failed to link up, even Yagi-to-Yagi. The only encouraging indication was with the unit on Broadwater Tower, on which a lone orange light showed that the tower radio was receiving data from its partner. No such result at the LTER headquarters, however. The 500 mW radios could not both span the distance.

The NovaRoam radios did link up, when set for lower speed of 159kbps, the recommended setting for over 10 miles. Although the Yagi-to-omni link would not work, Yagi-to-Yagi did -- particularly when the antennas were horizontally polarized. (Polarization, for a Yagi antenna, refers to the orientation of the tines -- up-and-down for vertical polarization, or left-to-right for horizontal. Omnidirectional antennas, for the purposes of this writeup, are always vertically polarized.)

Dave Hughes PI linking NovaRoam from 2d Floor to Hog Island.

Attempts to link from ground level on Hog Island back to the Farmhouse, using a 6db antenna also failed. We wanted to rule out any effect the all metal tower might have on the signal.

John Porter, LTER PI holding antenna on Hog Island

Conclusions

In some ways we confirmed the obvious:

(1) 800 milliwatts gets you greater range than 500 milliwatts; and

(2) 130,000 bits per second gets you greater range than 2,000,000 bits per second.

Even though the NovaRoam radios passed the test that the WiLan units failed, we decided to give the WiLan radios a second chance due to the desire higher bandwidth that would be needed if we were to run any live video applications. This decision was reinforced by telephone consultation with John Kinghorn of Wi-Lan's Technical Support department, who informed us that (1) the orange light at Broadwater Tower indicated that a link had been half-established, and (2) a higher-gain Yagi antenna would likely solve the problem. We were, in short, very close to a robust link.

Mr. Kinghorn also confirmed another characteristic of this WiLan radio: even when there is no data being moved, these radios are always transmitting, thus tending to saturate the 902-928 MHz frequency band. Since we intend to operate them between a rural farmhouse and an uninhabited island, we believe this behavior will not cause anyone any inconvenience.

After all reassembled at the Farmhouse, a second planning session was held, in which PI Hughes and PI Porter came to the agreement that the LTER would provide all the infrastructure equipment - batteries, recharging devices, whether solar or wind power, masts, and waterproof cases - while our project would provide the Radios, antenna cables/connectors, and expertise getting them installed.

The Farmhouse Planning Session

Wi-Lan Revisited

On December 7, Tom Williams went back to Oyster, and everyone performed another test of the WiLan radios, this time using 13 dBi Yagi antennas made by Cushcraft. We also refined our testing techniques a bit.

A 13 dBi Yagi is more than twice the length of a 9 dBi Yagi. No real surprise there -- remember that each 3 dB represents a factor of about 2-to-1.

Randy Carlson with a 13dBi and 9dBi Yagis

As we had hoped, the extra 4 dB gain at each end paid off nicely with a solid link between the farmhouse (headquarters) and Broadwater Tower. We knew this because, having established a connection, we were able to get some very useful metrics out of the WiLan radios link statistics. This is one of that brand's major strengths -- a fairly detailed performance measurement available through the serial port. (For oscilloscope users, there is even more detailed information available through the "Link Test" connection on the back.) It displays information on data reliability, as well as reception strength for both base and remote radios.

Hopper-Plus Loopback Display

Hopper lights indicating moving data and link

Due to the greater detail of the WiLan loopback display, we were able to confirm -- with numbers -- what had been apparent to observation in the previous test: at this distance, at 900 MHz frequency (and perhaps also relating to the amount of water between points), horizontal polarity was more effective than vertical polarity. We lost a few bits (and had lower signal strength) with vertical polarity, and could not really establish a link at all when we used the 6 dBi omni at Broadwater Tower, whereas reception was flawless with two horizontally polarized Yagis.

Not Quite Out of the Woods Yet

Although this test moves us closer to the dream of a relatively high bandwidth wireless Internet cloud on Hog Island, we still have a few challenges ahead.

One challenge is to extend the link from Broadwater Tower at the south end of the island, to Machipongo Station at the north end. In order to use Broadwater Tower as a relay point, it may be necessary either to use larger antennas or to add another radio. Hopefully the former, since the latter solution would require even more power (solar array, batteries, voltage regulation, inverter, and the like) at the site.

Another challenge is FCC approval of the link. A 13 dBi (or greater) antenna on a 500 milliwatt radio is over the legal limit per the Part 15 rules, so once we determine what antennas will be needed for this link we will need to obtain an FCC waiver in order to actually deploy them.