Remote ID is a beacon, not a sensor

The popular description of Remote ID is a digital license plate. That analogy is doing more damage than it looks, because a license plate implies a world where an officer can see the car, walk up to it, and read the plate. Remote ID does not work that way. It is closer to the car volunteering its plate number over the radio, on the honor system, and only while it is close enough to hear.

I spent my service as a Counter-IED and Electronic Warfare Officer, and this is one of the oldest distinctions in the trade. Identification and detection are separate problems solved by separate systems. Military aircraft carry IFF, Identification Friend or Foe, and IFF has never been how anyone finds an enemy aircraft. It answers a question you can only ask about something you already found, and it only answers when the other side wants it to. Remote ID is the civil-airspace version of that, and reading the regulation with that framing makes the whole thing snap into focus.

So here is the rule, 14 CFR Part 89, read closely, along with what each requirement does and does not give you.

What the rule actually demands

The operating requirement is short. Under § 89.110, "From takeoff to shutdown, the standard remote identification unmanned aircraft must broadcast the message elements of § 89.305." If that equipment stops working mid-flight, the person at the controls "must land the unmanned aircraft as soon as practicable."

Section 89.305 lists what gets broadcast. A standard Remote ID aircraft "must be capable of broadcasting" its identity, which is either "a serial number assigned to the unmanned aircraft by the person responsible for the production of the standard remote identification unmanned aircraft" or "a session ID"; the latitude and longitude of the control station; the geometric altitude of the control station; the latitude and longitude of the aircraft; the geometric altitude of the aircraft; the velocity; a UTC time mark; and an indication of emergency status.

That control station element is the part worth pausing on. For a standard Remote ID aircraft, the broadcast includes where the operator is standing. Most people assume Remote ID tells you about the drone. What makes it genuinely useful to a responding officer is that it tells you about the person.

The performance floor is in § 89.310, and the numbers are modest on purpose. Reported position "must be accurate to within 100 feet of the true position, with 95 percent probability." Control station altitude gets 15 feet at the same confidence. Aircraft altitude gets 150 feet. Position must be broadcast "no later than 1.0 seconds from the time of measurement to the time of broadcast," at "a rate of at least 1 message per second." And the transmission must use "a non-proprietary broadcast specification and using radio frequency spectrum compatible with personal wireless devices in accordance with 47 CFR part 15, where operations may occur without an FCC individual license."

Read that last one again, because it is the design decision that defines everything else. Part 15 is unlicensed spectrum, the same neighborhood as Wi-Fi and Bluetooth. The rule requires a short-range local broadcast that any phone can receive. It does not require the flight to be reported to the FAA, to a database, or to anyone at all. There is no central feed. There is a radio signal, and you have to be near enough to hear it.

The module that tells you less

Not every drone has Remote ID built in. The rule provides a retrofit path, and the retrofit broadcasts a different and thinner set of facts.

Under § 89.315, a remote identification broadcast module transmits the serial number of the module, the aircraft's latitude and longitude, its geometric altitude, its velocity, a UTC time mark, and then this instead of the control station: "an indication of the latitude and longitude of the take-off location of the unmanned aircraft," plus the geometric altitude of that take-off location.

Take-off location is not operator location. It is where the aircraft left the ground, which is where the operator was at that moment and not necessarily where they are now. There is also no emergency-status element. The retrofit path is a real compliance path, and it hands a receiving officer a materially weaker picture than the built-in version.

The rule compensates with a different kind of constraint. Section 89.115 attaches a condition to module use: "The person manipulating the flight controls of the unmanned aircraft system must be able to see the unmanned aircraft at all times throughout the operation." A module flight is a line-of-sight flight. The regulation is trading electronic accountability for the human kind.

The gaps that are features

Three separate provisions describe legal flight with no Remote ID broadcast at all. None of them is an oversight.

First, § 89.115 also allows operation where "the unmanned aircraft and the person manipulating the flight controls of the unmanned aircraft system remain within the boundaries of an FAA-recognized identification area throughout the operation," again with the aircraft visible to the operator at all times. FAA-recognized identification areas, generally called FRIAs, are defined sites where aircraft without Remote ID equipment can fly. They exist so that model-aircraft flying fields and the legacy fleet did not become instantly unlawful.

Second, the operating requirement has a boundary. Section 89.101 applies the rule to "persons operating unmanned aircraft registered or required to be registered under part 47 or 48" and to foreign civil unmanned aircraft in the United States. Registration is therefore the trigger, and 14 CFR 48.15 exempts an aircraft that "is operated exclusively in compliance with 49 U.S.C. 44809 and weighs 0.55 pounds or less on takeoff, including everything that is on board or otherwise attached to the aircraft." A sub-250-gram machine flown recreationally is outside the registration requirement, and therefore outside the broadcast requirement.

Third, § 89.101 excepts operations under part 91 that are "transmitting ADS-B Out pursuant to § 91.225." That one is sensible: an aircraft already broadcasting on the aviation surveillance system does not need to broadcast a second time on Wi-Fi.

Add them up and you get an honest statement of coverage. Remote ID covers registered civil aircraft flown by people who follow the rule. That is a large and useful population. It is not the population that anyone deploying a counter-drone system is worried about.

Why this is an identification system, not a detection system

Here is the operator's version of the same point, and it is the reason this distinction matters beyond regulatory trivia.

A sensor works on emissions the target cannot choose to withhold. Radar illuminates the target and reads what comes back. An acoustic array listens to the noise the propellers cannot help making. A radio-frequency direction finder listens to the control link the pilot needs in order to fly. None of those require the target's consent, which is exactly why they are sensors.

Remote ID is the opposite construction. It is a declaration the aircraft makes about itself, in a format and on a schedule the rule specifies, and every element of it can be absent for a reason that is perfectly legal, or absent for a reason that is not. A receiver sees the same thing in both cases: nothing. A drone flying with no Remote ID broadcast might be under 250 grams, or inside a FRIA, or a home-built machine, or a compliant aircraft whose module failed, or someone who simply turned it off. The silence itself carries almost no information.

Even the presence of a broadcast is a weaker signal than it looks. The identity element can be "a session ID" rather than a serial number, so a member of the public with a receiver may be looking at an identifier that means nothing to them. The position is specified to 100 feet at 95 percent probability, and the altitude to 150 feet, which is airspace-awareness quality rather than intercept quality. And the whole thing arrives over unlicensed spectrum, which is to say a channel with no authentication requirement in the operating rules I have quoted here. Any system that trusts an unauthenticated broadcast is trusting the transmitter.

That is not a criticism of the FAA. The rule was written to solve accountability, and for accountability it is a good instrument. If a drone is over a stadium and a responder can pull up its serial number and the operator's position from a phone, the situation resolves in minutes instead of hours. The failure is entirely in how the rule gets described afterward, when a compliance framework is discussed as though it were a surveillance layer.

What I could not confirm

Everything above is drawn from the text of the regulations themselves, read section by section: 14 CFR 89.101, 89.110, 89.115, 89.305, 89.310, 89.315, and 14 CFR 48.15. I did not read Part 89 in its entirety. Subpart E on means of compliance and Subpart F on design and production set requirements on manufacturers that I have not examined here, and my statement that nothing requires a network report is a statement about the operating and equipment sections I quoted, not a claim about every provision in the part.

I also did not open the rulemaking preamble, so I am making no claims about the FAA's stated intent, about what was proposed and later changed, or about the reasoning behind any specific number. Where I have said what a provision does, that is the text. Where I have said why it matters, that is my reading as an electronic warfare officer, and it is mine rather than the agency's.

I have not assessed compliance rates, enforcement practice, receiver app quality, or how well any of this works in the field. Those are empirical questions and this report is a reading of a rule.

The signal

Three things worth carrying out of Part 89.

First, Remote ID answers "who is that, and where is the pilot" for aircraft that participate. It does not answer "is something up there." Those are different questions, and only the second one is solved by sensors. If a briefing, a vendor deck, or a city council presentation treats Remote ID as drone detection, the rule contradicts it in a paragraph.

Second, the coverage boundaries are written into the regulation and are easy to check: FRIAs, the 0.55-pound registration exemption, the ADS-B exception, and the retrofit module that reports take-off location instead of where the operator is standing. A responsible plan states which of those it is exposed to instead of assuming full coverage.

Third, and most general, absence of a cooperative signal is not evidence of absence. The one lesson that transfers everywhere from electronic warfare into everyday security engineering is that a system built on voluntary declaration measures compliance, never threat. Anything relying on the target to identify itself is telling you about the honest population, and the honest population was never the problem.