Energy & Power · Study deck

Low-Power Design: Firmware and Wake Policies

A sleeping node can still draw power through a pull-up, a resistor that holds an input high, or a sensor rail, the line that powers a sensor.

Battery Bruno is your guide for this deck.

awarestrategies
Battery Bruno, the module guide, in a scene from this chapter.
iotclass.org

After studying this chapter

Learning objectives

You will be able to:

  • Select a sleep mode from wake latency, state retention, and wake-source requirements.
  • Design time-based, event-driven, and adaptive duty-cycle policies.
  • Identify hidden loads from GPIOs, pull-ups, regulators, sensors, LEDs, and radios.
  • Use power gating and state reinitialization without corrupting measurements or firmware behavior.
iotclass.org

Major section

Start With the Sleeping Device

Saving power is not simply making everything sleep; the device must still notice the event and preserve its meaning.

  • The first decision is which service promise controls each wake-up.
  • An application programming interface is a defined way for software parts to request work or data; API is its short name.

Key terms

Latency
Latency means the time from an input event to the result that matters.
iotclass.org

Major section

Start With the Sleeping Device (continued)

Latency means the time from an input event to the result that matters.

  • This runway does not prove that the deepest sleep state is always best.
  • The cheapest energy is the energy a device never draws.
  • The simple route is service policy first, wake policy second, power states third, and measurement last.
A transistor transfer curve on a log scale comparing a low-threshold-voltage device and a high-threshold-voltage device, showing that neither reaches zero current below threshold.
A transistor transfer curve on a log scale comparing a low-threshold-voltage device and a high-threshold-voltage device, showing that neither reaches zero current below threshold.
iotclass.org

Major section

Minimum Viable Understanding

The deepest sleep mode is not always best; wake latency, retained state, and wake-source support matter.

  • Peripherals can dominate sleep current unless they are disabled, held in a known state, or power-gated.
  • Radio energy includes startup, association, receive windows, retries, and acknowledgments, not only transmit current.
  • Every optimization claim should be checked against a before/after current trace.

Why it matters

Duty cycling reduces average current only if active work is short and sleep current is actually low.

iotclass.org

Major section

Strategy Stack

Low-power design works best when decisions are layered from service policy down to board leakage.

  • The separate: Wake label exposes the energy consequence, preventing the first two stages from being accepted on their own.
  • The complete reading is that no-panel low-power strategy stack from service policy through wake policy, power states, load gating, and measurement evidence.
Low-power strategy stack showing service policy, wake policy, power states, load gating, and measurement evidence.
Low-power strategy stack showing service policy, wake policy, power states, load gating, and measurement evidence.
iotclass.org

Major section

Sleep-Mode Selection

Sleep-mode selection is a requirements decision.

  • A mode is viable only if it preserves the state you need, supports the wake source you need, and wakes quickly enough for the service requirement.
  • The simplified worked example below shows why sleep percentage alone is not enough: every state contributes current multiplied by time.
True-scale charge ledger for one active second and 99 sleeping seconds.
True-scale charge ledger for one active second and 99 sleeping seconds.
iotclass.org

Major section

Checkpoint: Sleep Policy Fit

The comparison makes the causal boundary explicit: oscillator and restore energy are paid once per wake, while sleep leakage grows with the interval.

  • Idle or wait-for-interrupt.
  • Wake must be nearly immediate and peripherals stay active.
  • Milliamps of standby current can destroy long-life budgets.
  • Frequent wake events need quick resume and retained RAM.
iotclass.org

Major section

Checkpoint: Sleep Policy Fit (continued)

Retained clocks and peripherals may keep current high.

  • Wake intervals are seconds to hours and limited state retention is acceptable.
  • Boot and reinitialization cost may dominate frequent wakes.
  • Full-cycle trace, wake cause, initialization time.
  • Events are rare and only a small wake circuit must remain alive.
iotclass.org

Deck summary

Key takeaways

Saving power is not simply making everything sleep; the device must still notice the event and preserve its meaning.

  • Latency means the time from an input event to the result that matters.
  • The deepest sleep mode is not always best; wake latency, retained state, and wake-source support matter.
  • Low-power design works best when decisions are layered from service policy down to board leakage.
  • Sleep-mode selection is a requirements decision.
iotclass.org

Retrieval practice

Recall check 1 of 6

Battery Bruno says: answer from memory, then check your reasoning.

Q1What evidence should approve a new low-power wake policy?

AA processor datasheet sleep-current value without board testing.
BA shorter firmware function without energy measurements.
CA modeled battery life with no measured current trace.
DA measured whole-device current trace with service checks.
Show answer

Answer: D The trace tests energy; service checks catch missed work.

iotclass.org

Retrieval practice

Recall check 2 of 6

Battery Bruno says: answer from memory, then check your reasoning.

Q2The processor sleeps, but whole-board current stays high. Where should the team look next?

AInspect powered sensors, pull-ups, regulators, LEDs and radio states around the CPU.
BInspect the CPU sleep-mode setting and confirm firmware enters that state, then accept the measured whole-board draw as the controller's expected sleep current.
CUse battery capacity rating to infer board sleep behavior.
DChange report interval to reduce current while measured board sleep remains high.
Show answer

Answer: A Loads around the processor can dominate sleep current.

iotclass.org

Retrieval practice

Recall check 3 of 6

Battery Bruno says: answer from memory, then check your reasoning.

Q3A battery node has a low sleep current but wakes too often for its service. Which design level should the team revisit?

AUse board leakage alone because wake frequency and active radio duration contribute negligible energy.
BThe battery label alone as proof that the repeated wake schedule is viable
CThe reporting latency target alone without its radio and energy cost
DThe service-level wake and reporting policy above the circuit power states
Show answer

Answer: D The strategy stack starts with what the service needs and when the device wakes.

iotclass.org

Retrieval practice

Recall check 4 of 6

Battery Bruno says: answer from memory, then check your reasoning.

Q4A deep sleep mode loses state required for the next action. How should the team judge that mode?

AAccept it from the sleep-current number alone
BCompare only the percentage of time asleep
CRequire the needed state and wake behavior
DUse the teaching example as the MCU specification
Show answer

Answer: C Viability depends on retained state, supported wake source, and response time.

iotclass.org

Retrieval practice

Recall check 5 of 6

Battery Bruno says: answer from memory, then check your reasoning.

Q5A door sensor must report a change at once but send a routine health report later. Which wake policy fits?

AEvent wake for door changes and scheduled wake for routine health reports
BLong fixed-timer wake for both door alerts and routine health reports
CContinuously powered radio for immediate door alerts and routine health reports
DBattery-full wake alone instead of door event and health scheduling
Show answer

Answer: A This hybrid policy meets both service needs.

iotclass.org

Retrieval practice

Recall check 6 of 6

Battery Bruno says: answer from memory, then check your reasoning.

Q6When does batching routine samples save radio energy?

AWhen every sample is delayed, including urgent alerts that miss deadlines
BWhen fewer radio sessions still meet the required alert timing
CChoose the larger payload because setup overhead stays dominant at that size.
DChoose longer local code execution because radio-on time remains lower than processing energy.
Show answer

Answer: B The saved sessions count only if the service promise holds.

iotclass.org

Print reference

Answers

Answer key.

  1. D · The trace tests energy; service checks catch missed work.
  2. A · Loads around the processor can dominate sleep current.
  3. D · The strategy stack starts with what the service needs and when the device wakes.
  4. C · Viability depends on retained state, supported wake source, and response time.
  5. A · This hybrid policy meets both service needs.
  6. B · The saved sessions count only if the service promise holds.
iotclass.org