The decision you should be able to make
Move from evidence to the smallest justified corrective action.
Core engineering
Begin with scope and timing: one direction or both, constant or intermittent, load-related, weather-related, time-of-day or change-related.
Compare current per-chain receive level, noise/interference indicators, modulation, retries, temperature, uptime and Ethernet errors with the accepted baseline.
A receive-level deficit points toward geometry, alignment, obstruction, polarisation or added loss; good receive level with poor performance points toward interference, configuration, queueing, Ethernet or processing limits.
Equipment and field checks
- Check alarms, logs, power quality, connector condition, mount movement and recent firmware/configuration changes before replacing hardware.
- Use a controlled channel or bench substitution only when it isolates a hypothesis.
- Record the before/after evidence for every change and revert changes that do not improve the measured fault.
Do not progress until
The corrective action is linked to a tested hypothesis and the link is re-validated against acceptance criteria.
Limits and cautions
- Do not raise power or disable regulatory controls as a troubleshooting shortcut.
Primary sources
Sources were checked for this technical review. Recheck regulator and equipment requirements at the point of design or installation.