Read-Zone Evidence: Dwell Time, Inventory Rounds, and Filtering
Ada re-derives the chapter’s own dwell time, inventory-round opportunities, and read-zone filtering rates
ADA · CALCULATION AUDIT
Read-Zone Evidence: Dwell Time, Inventory Rounds, and Filtering
RFID evidence is physics plus bookkeeping. The radio field gives the tag a short read window; the ledger proves which raw observations became accepted business events.
A tagged tool cabinet should identify its 18 tools while ignoring 3 nearby bench tools, yet its raw trace shows 63 observations for those 18 real items. A moving portal has a different denominator: 32 tagged cartons pass through a 2.4 m read zone at 1.6 m/s, giving about 1.5 s of dwell time and, at a 200 ms inventory round, roughly 7.5 rounds before the cartons leave the zone — and the pilot actually reads 29 of the 32. This audit asks the question those two examples invite: what evidence turns 63 raw observations and a 29-of-32 pilot into a trustworthy read-zone claim, rather than just an RF success story?
Companion to the chapter How RFID Identifies Objects — every number here comes from that chapter.
Convert motion into dwell time
1. Convert motion into dwell time. The portal example gives a 2.4 m read zone and a carton speed of 1.6 m/s.
That is the physics constraint: a carton is only inside the useful RF evidence boundary for about 1.5 s.
Convert dwell time into inventory opportunities
2. Convert dwell time into inventory opportunities. The chapter states an observed inventory round time of 200 ms, which is 0.2 s.
The result is not a guaranteed read count. It says the system has about seven and a half inventory opportunities before the cartons leave the zone, so misses still need a cause: orientation, material, anti-collision timing, antenna aim, or middleware timeout.
Audit the read-zone ledger
| Evidence question | Chapter values | Audit arithmetic | Checked result |
|---|---|---|---|
| Did the cabinet accept the in-scope tools? | 18 expected tools, 18 accepted |
18 / 18 = 1.00 |
100% accepted verified |
| Did the cabinet reject nearby bench tools? | 0 accepted from 3 nearby tools |
0 / 3 = 0 |
0% stray acceptance verified |
| How much raw observation noise was filtered? | 63 raw observations, 18 accepted events |
63 - 18 = 45; 45 / 63 = 0.714 |
45 observations filtered, about 71.4% of the raw trace |
| How strong was the portal pilot? | 29 reads from 32 cartons |
29 / 32 = 0.90625; 3 / 32 = 0.09375 |
90.6% read rate, 9.4% miss rate |
Review rule: accept the RFID claim only when the read-zone length, object speed, round timing, accepted-event denominator, rejected-nearby denominator, and duplicate filter are all recorded with the pilot trace.
Every number above is taken from the chapter’s own portal and cabinet examples and re-derived step by step.