Flying Near the Electronic Proving Ground: GPS and RF Interference in Southern Cochise County

Most articles about drone GPS loss are written by people who have read about it. This one is about a specific place where it is a designed feature of the environment rather than an anomaly.

The US Army Electronic Proving Ground at Fort Huachuca is one of a very small number of ranges in the country authorized to conduct open-air GPS testing. Not simulation — live sky. The site was chosen more than sixty years ago precisely because the high-desert basin, ringed by mountains, forms a natural RF barrier that keeps Tucson's electromagnetic noise out.

That barrier works in both directions, and the same bowl your property sits in is the bowl the testing happens in.

What actually goes on there

The EPG runs both indoor and outdoor GPS work. The indoor lab uses constellation simulators to model receivers anywhere on earth, but as one Army engineer put it, the results still have to be verified outside in a real operational environment using live sky.

Part of that work is testing how systems behave when GPS is degraded or denied — both natural interruption and deliberate jamming scenarios. The Army is explicit that engineers develop test techniques to minimise the area of impact so that unintentional interference is not encountered in surrounding areas, and that the EPG's spectrum manager coordinates with the FAA, the FCC, other DoD agencies and local partners specifically to protect other GPS users in the region.

The installation also hosts extensive UAS training and broader electronic-warfare and communications testing. Local reporting over the years has noted the reverse problem too — that ordinary consumer devices in a growing city crowd the same spectrum the range depends on.

What GPS degradation looks like on a small drone

It rarely looks like a clean failure. It looks like the aircraft becoming subtly untrustworthy, and the tells are worth knowing whether you fly yourself or are watching someone else fly your property.

• Satellite count dropping without weather or a change in position to explain it.
• Drifting in a hover — the aircraft slowly walking off station in still air, because it has fallen back to visual positioning or to nothing.
• A home point that moves, which is the dangerous one, because return-to-home will fly to the new one.
• Position mode dropping out to attitude mode, which usually shows on the controller and needs the pilot to notice it.
• Waypoint or mapping missions refusing to start, or a grid flight going ragged mid-mission.
• Video link degradation, which is a separate band but often correlates because the underlying cause is broad-spectrum.

The failure that hurts is not losing GPS. It is losing GPS without noticing, continuing the mission, and then triggering a return-to-home to a home point that is four hundred yards from where the pilot is standing.

How we fly differently down there

This is the operational answer, and it is unglamorous.

Closer in. We fly at distances where the aircraft stays comfortably and unambiguously in sight, well inside what the link budget would allow. If GPS goes, the recovery is manual flight home, and manual flight home only works if you can see it.

More reserve than the arithmetic requires. A degraded aircraft flown home manually against wind uses more battery than a planned return. The reserve is there for that, not for an extra pass.

Home point verified, not assumed. Checked on the ground before launch and re-checked if the aircraft has been up for a while.

Manual RTH altitude set for the terrain, not left at default. In broken foothill ground the default is sometimes below the ridge behind you.

We land on the first anomaly. Not troubleshoot at altitude, not "see if it settles." Land, work out what happened on the ground, launch again if it is explained. This costs a few minutes and has never once cost an aircraft.

Mapping missions get re-flown rather than patched. A photogrammetry block flown through a period of positional degradation produces geotags that are quietly wrong, and the resulting model looks fine and is not. If the flight log shows a satellite drop mid-grid, that block gets re-flown. This is the reason we would rather find out about a problem than not.

The other things to check in the same area

Interference is one layer. Around Fort Huachuca there are three more, and they are administered separately from one another:

The last row is the point of this article. The first three are questions you can answer at a desk before you drive. The fourth is not an approval you obtain; it is a set of habits you fly with. The airspace side is covered in more detail here, and the county-wide restriction list here.

What this means if you are hiring somebody

An operator driving down from Phoenix or in from out of state has no reason to expect any of this. Their flight planning is not wrong; it is simply built for an environment where GPS is a utility that works. Their first indication that this is a different place will be an aircraft doing something unexpected over your client's roof, your job site, or your field.

Under 14 CFR § 107.19 the remote pilot in command is responsible for the safety of the operation and has to determine that the aircraft is in a condition for safe operation. In this corner of the state, part of that determination is environmental, and it is not written on any chart.

Two questions worth asking anyone quoting you work near Sierra Vista, Huachuca City, Whetstone or Hereford:

1. "What do you do if you lose GPS mid-flight?" An immediate, specific answer means they have thought about it. A pause means they have not.
2. "How do you check whether a mapping flight was affected?" The answer should involve the flight log and ground control, not the look of the output.

We are based an hour north of there and we fly it regularly. If you have work in the south end of the county and want the environment factored into the plan rather than discovered on the day, send us the address and we will tell you what we would do differently on it.