Separate function from transport
ICAO defines the C2 link as the data link between a remotely piloted aircraft
and its remote pilot station for managing the flight. The
ICAO terminology page
is deliberately functional: it does not say the link must use one band, modem,
network, or antenna.
The same separation appears in
ITU-R Report M.2171.
Its control-link description includes telecommands sent toward the aircraft and
non-payload telemetry returned to the control station. The report treats payload
as a separate subsystem, while also recognizing other non-payload flows such as
navigation information, surveillance data, and air-traffic-services relay.
Scroll horizontally to compare all columns.
The rows can share a bearer, but their consequences differ. A dropped image
packet may reduce a deliverable. A stale aircraft-state message may undermine
supervision. A delayed command may affect the ability to direct the aircraft.
Treating all three as “the video link” prevents meaningful requirements.
C2 is more than stick position
In an automated mission, the uplink may carry mode changes, route revisions,
approvals, or abort commands instead of continuous control-surface inputs. The
downlink may carry the state needed to supervise those commands. Low average
data rate does not mean low criticality.
This distinction connects directly to the
authority test for automation and autonomy.
If a human retains authority to redirect or terminate a mission, the
architecture must identify the information and command paths that make that
authority usable. It must also define what happens when either path is delayed,
corrupted, or absent.
Telemetry deserves equal care. Position alone is not enough to explain aircraft
state. A useful set may include navigation validity, flight mode, command
acknowledgement, link quality, energy estimate, contingency state, and payload
status. Which items are necessary depends on the operation and safety argument.
Payload data can have different failure tolerances
Mission data often tolerates buffering, retransmission, adaptive compression, or
delayed delivery. C2 may not. Combining them on one network can be efficient,
but only if resource allocation prevents payload traffic from starving a more
critical flow.
Video makes the trade especially visible. Higher resolution, frame rate, bit
depth, or lower compression increases demand. If video is only a framing aid,
the system might degrade its quality while preserving command and state traffic.
If the operation depends on video for a safety function, that assumption changes
the performance requirement and must be justified explicitly.
Onboard recording is not automatically a backup for every mission. It protects
data only if the aircraft returns and the storage remains intact. A low-rate
preview plus onboard full-resolution capture may be a sound mission-data design,
but it does not replace the C2 path.
Build requirements from consequences
For each information flow, document at least:
- source and destination;
- message content and units;
- expected and worst-case rate;
- maximum useful age or latency;
- integrity and authentication needs;
- availability target and coverage assumptions;
- priority when capacity is constrained;
- behavior after delay, corruption, duplication, or loss;
- crew indication and recovery method;
- recorded evidence for post-flight review.
Then map those flows to radios, networks, antennas, frequencies, service
providers, and onboard buses. This exposes common-mode failures. Two screen
icons do not provide redundancy if both services depend on the same modem,
antenna, power rail, tower backhaul, or software process.
Preflight checks should test the architecture
14 CFR 107.49
requires a Part 107 remote pilot in command to ensure that control links between
the ground control station and aircraft are working properly before flight. A
green connection indicator is a starting point, not a complete architecture
check.
A mission-specific check can confirm:
- commands reach the intended vehicle and produce acknowledgements;
- required state fields are current and plausible;
- payload traffic cannot mask C2 degradation;
- antennas and equipment match the planned geometry and configuration;
- loss thresholds and contingency actions are configured as intended;
- the crew can recognize which path has failed;
- recordings preserve the evidence needed to diagnose a problem.
Link performance is conditional. Terrain, antenna orientation, installation
loss, interference, network loading, weather at some frequencies, and aircraft
attitude can change it. A coverage claim without configuration, environment,
traffic assumptions, and a failure response is not yet an operational
requirement.
Define the information flows first, assign their consequences, and only then
select or assess the transports. This keeps a fast payload stream from being
confused with a safe control architecture and makes every link claim testable.
