1. The signal that had to find the best path

Carl Sagan’s Contact returns as the recalled work because the message from Vega reaches Earth through whatever path the universe provides. A LEO desktop has more choices. It can transmit directly to a ground station, or it can route through a relay constellation. The link rate can be high when the path is good and low when it is not. The wondering is how the platform makes these choices without a human adjusting the dials.

This entry wonders about adaptive link rates and relay routing.

The link budget changes during a pass. At low elevation the signal is weak; at high elevation it is strong. If the platform uses a fixed high data rate, it may lose the link at low elevation. If it uses a fixed low rate, it wastes capacity at high elevation. Adaptive link rate solves this by matching the modulation and coding to the measured signal quality.

The platform can monitor received signal strength, bit error rate, or frame error rate. When the margin is high, it switches to a faster mode. When the margin drops, it falls back to a more robust mode. The decision needs hysteresis: switching modes too often wastes time and risks dropouts.

Adaptive rate also interacts with power and thermal budgets. A high-rate mode may require more transmit power and generate more heat. The autonomy must verify that the power and thermal managers can support the chosen mode before using it.

Direct downlink is simple and cheap when ground station coverage is adequate. It gives the operator full control of the radio and the data path. But a LEO orbit may spend most of its time out of view of any single station.

Relay constellations provide near-continuous coverage but introduce dependency on a third party. The platform must use the relay’s protocols, frequencies, and scheduling. The relay may also charge for data volume or connection time.

A hybrid architecture is attractive: direct downlink for bulk data when a station is available, and relay for command, telemetry, and urgent data between passes. The autonomy must choose the path for each data item based on availability, cost, latency, and data rate.

4. Routing as an optimization problem

At any moment the platform may have several possible routes:

  • direct to ground station A with a good link;
  • direct to ground station B with a marginal link;
  • relay constellation X with continuous but expensive connectivity;
  • relay constellation Y with intermittent coverage.

The autonomy can score each route by latency, cost, data rate, and reliability. The choice may change during a pass as geometry changes. A data item with a tight deadline might take the first available route, even if it is more expensive. A large file might wait for the cheapest high-bandwidth route.

5. What this changes

  • Adaptive link rate maximizes throughput across a pass but requires coordination with power and thermal budgets.
  • Relay routing adds cost and dependency but can dramatically increase connectivity.
  • The autonomy should choose routes based on availability, cost, latency, and data rate.
  • The next entry will wonder about missed pass recovery and fault tolerance.