Input
Processing Status
Chains
Stations
| Role | Class | Station | GRI offset | Lock | LDC Eurofix | LDC 9th pulse |
|---|
Click a station to select it: the highlighted row drives the pulse plots and the decoded messages below.
Data messages
Times are when each message was sent, worked out from the receiver's clock (good to a second or so on a live receiver). A message sent again with the same content is listed again. Messages that did not decode are missing, so the gaps between them can be longer than the schedule.
Pulses
The blue trace is the detector's group correlation with each station's comb sidelobes hidden. Correlating against an 8-pulse comb also responds at every 1 ms offset where some pulses line up (15 lines for one group), so the raw trace looks like a train of pulses; only the peak, the group start, is meaningful, and the orange trace (and the zoom below) show the real pulses.
Every point is the median, across the newest GRI cycles (the "Median of" setting), of the signal at that phase within the cycle: a pulse that recurs every cycle survives, while an atmospheric crash that hits one cycle does not. The plots follow the signal as it arrives; choose more cycles for a steadier, less noisy picture, or "all buffered" for the smoothest one (that one is refreshed only with each analysis pass, which takes longer). The waterfall below updates continuously.
Solid ticks (M, S1, S2, ...) mark stations actually detected here. Dotted "?" ticks, when a master is locked and the chain matches a known one (data/stations.conf), mark chain members the database lists but nothing here has confidently identified yet, at the master's own group start plus that station's published emission delay -- the chain's nominal position for it, what a receiver equidistant from both stations would see. A real receiver's own propagation-delay difference (stations.h) can shift the true pulses either side of that tick, by up to several thousand µs for real chain baselines, so treat it as a reference point to search near, not an exact position.
Waterfall
GRI offsets are measured in the server's stream frame (its own sample counter, mod the GRI), so their zero point is arbitrary but stays put from update to update; a steady station drifts only if the receiver's clock and the GRI genuinely disagree. Verdicts are the server's heuristics, not proof — see the project README for what each one does and doesn't establish.