Peripheral Latency Governs Both Power and Timeout

Peripheral Latency Governs Both Power and Timeout

Ada re-derives this chapter’s own numbers step by step, at full precision

ADA · CALCULATION AUDIT

Peripheral Latency Governs Both Power and Timeout

A BLE peripheral on a 30 ms connection interval with peripheral latency 4 has a supervision-timeout floor of (1 + 4) × 30 × 2 = 300 ms, so a 2 s timeout is legal but a 200 ms one is not, while the same latency stretches its idle listening gap to 250 ms. Push latency to 9 on a 1 s interval and the gap and the timeout floor jump to 10 s and 20 s. This audit re-derives each and asks whether the shared (1 + latency) factor means a power tweak that lets the radio sleep longer silently widens how long a dead link goes undetected.

Companion to the chapter BLE Field Debugging and Internals — every number here comes from that chapter.

Try

A BLE peripheral on a 30 ms connection interval with peripheral latency 4 has a supervision-timeout floor of (1 + 4) × 30 × 2 = 300 ms , so a 2 s timeout is legal but a 200 ms one is not, while the same latency stretches its idle listening gap to 250 ms . Calculate this case.

Observe

This audit re-derives each and asks whether the shared (1 + latency) factor means a power tweak that lets the radio sleep longer silently widens how long a dead link goes undetected. Check shows this.

Explain

The design meaning is the coupling: raising latency to let the radio sleep longer multiplies the idle gap and the minimum legal supervision timeout by the identical (1 + latency) factor, so a power tweak that widens the sleep window silently widens how long a dead link goes undetected -- which is exactly why a high-latency sensor is forced into a long supervision timeout. Check confirms it.

See the relationship before changing it

The figure reads from left to right. The blue input is peripheral latency. The middle card names the page’s rule. The green output is idle gap. The arrow matters: change the input, apply the rule once, then read the result with its unit.

Walk the arrow. Latency lets the peripheral skip events, so the idle gap grows.

Derive the baseline in four named moves

1. 1

   Name the input. The chapter baseline is 4 events.

2. 2

   Name the relationship. gap = (1 + latency) x 50 ms

3. 3

   Substitute with units. (1 + 4) x 50 ms = 250 ms

4. 4

   Read the result. Keep the unit beside the value, then use the result only inside the technical boundary below.

Predict, then change peripheral latency

Try Predict how idle gap responds when peripheral latency moves. Calculate peripheral latency; compare idle gap with that prediction.

Peripheral latency

4 events

Chapter baseline

Idle gap

Observe Return to 4 events. Recheck idle gap with peripheral latency at its chapter value.

Explain Latency lets the peripheral skip events, so the idle gap grows.

Check yourself

What should you do before trusting a moved-slider result?

Answer: Predict its direction, apply the displayed relationship, keep the units, and compare the reset value with the chapter’s worked baseline.

What does this small model leave out?

Answer: Only peripheral latency moves here. The idle gap calculation excludes field effects listed below.

Honesty note. The idle gap result holds peripheral latency as the only changing input. It is not a field prediction.

Technical boundaries

Beyond the scope of “Peripheral Latency Governs Both Power and Timeout” sit RF collisions, connection-event retries, clock drift, OS scheduling, peripheral firmware stalls, or battery voltage sag; “Peripheral Latency Governs Both Power and Timeout” therefore reports only its named fixtures.

Ada: The Under-the-Hood section uses the same (1 + peripheral_latency) multiplier twice – once for the idle listening gap and once for the supervision-timeout floor. Let me confirm both, because that shared factor is the reason you cannot tune one without moving the other.

Idle listening gap, (1 + latency) x interval:

• 50 ms interval, latency 4: (1 + 4) x 50 = 250 ms.
• 1 s interval, latency 9: (1 + 9) x 1,000 = 10,000 ms = 10 s.

Supervision-timeout floor, (1 + latency) x interval x 2:

• 30 ms interval, latency 4: (1 + 4) x 30 x 2 = 300 ms, so a 2 s timeout is valid while a 200 ms timeout is below the floor and illegal.
• 1 s interval, latency 9: (1 + 9) x 1,000 x 2 = 20,000 ms = 20 s, so a 10 s timeout – half the floor – is invalid for that latency.

The design meaning is the coupling: raising latency to let the radio sleep longer multiplies the idle gap and the minimum legal supervision timeout by the identical (1 + latency) factor, so a power tweak that widens the sleep window silently widens how long a dead link goes undetected – which is exactly why a high-latency sensor is forced into a long supervision timeout.

Every number above is taken from the chapter’s own material and re-derived step by step.