monad03_walk-abba-48-night_20260903-114339

In plain words A library keeps three lists: what was delivered, what was catalogued, and who checked it out. A measurement corpus needs the same three, and a gap between them is a fact about the pipeline, not about the room. This is the ledger, per day and receiver, with every session as a row you can open.

The rate ladder

Counts and yields

Records	14,199,623
Frames the radio saw	14,307,298
Empty records	322,659
useful_yield	0.9699
capture_yield	0.9925
Empty share	0.0227
Mean rate	665.6 Hz
Bytes	9,939,737,956
Band / channel / width	5 GHz / 48 / HT20

Quote useful_yield: (records − empty records) ÷ frames the radio saw. capture_yield counts an all-zero record as delivered. A low capture_yield on 2.4 GHz is DSSS/CCK traffic that yields no channel estimate, not a quiet channel. Both are derived ratios from two independent counters and exceed 1.0 on some sessions; the counts are exact, the ratios are not.

State

Sealed	no
Decoded	decoded (12 of 12 segments)
Projected note	yes
Integrity flags	`unsealed`
Run id	`lib-drift-day-2026-09-03-05`
csid	0.2.0

Record classes

What the channel carried, by tone count and modulation. A bandwidth is printed only when every frame of the class named the same one.
Class	Records	Share	MHz
52t-legacy_ofdm	14,199,565	100.0 %	20
56t-ht	58	0.0 %	20