AP and client power mismatch

Why it happens

Every Wi-Fi exchange runs both ways. The access point sends, and the client answers with its own data and acknowledgments. Path loss is the same in both directions. Transmit power is not.

One vendor’s configuration guide calls this an asymmetric power issue. An access point can transmit far, while a battery-powered client replies with less power than the access point needs to hear it. The guide puts the least powerful clients at around 10 dBm for iPhones and 14 dBm for most laptops¹ and advises matching access point power to them. An industrial design guide for EtherNet/IP gives the same rule: access point power typically matches the client adapters and bridges in the cell.

The client cannot see the problem, because it judges its link by what it receives. Apple documents that an iPhone holds its access point until the signal it hears passes -70 dBm². Intel’s roaming setting moves the signal threshold at which its adapter starts to scan. Run the access point 10 dB hotter than the client and, with equal antenna gains, the access point hears that client at about -80 dBm while the client still reads -70 dBm. The client stays while its uplink weakens. Its frames start to fail, it retries and steps down its rate, and the link can break with a good signal on the client’s screen.

6 GHz writes a gap into the rules. A low-power indoor access point may radiate 5 dBm/MHz³ and its clients -1 dBm/MHz. A client of a standard power access point must stay at least 6 dB below it. The FCC reasoned that clients need less power because traffic is asymmetrical. Control traffic is not. An EtherNet/IP adapter on a vehicle produces its I/O data at the requested rate, whatever the controller sends it.

Transmit power control does not close the gap. Under 802.11h, a client sets its local maximum to the regulatory maximum minus the access point’s power constraint. That can turn clients down. It cannot turn them up. NIST warns that more transmit power is not necessarily an improvement and suggests a link budget for each device. A louder access point does look better on a survey, because a survey that listens to beacons measures only the downlink. The control link fails on the way back.

How to confirm it

1. At the spot where the client fails, read the client's signal as the access point hears it, from the controller or access point, and the access point's signal as the client hears it, from the client. Compare the two.
2. Capture over the air near that spot as the measurement guide describes, and count retries and data rates separately for frames to the client and frames from it.
3. Compare the access point radio's configured transmit power with the client's transmit power from its data sheet or driver settings.
4. Check whether automatic power control has raised the access point's power since the cell was designed.
5. In 6 GHz, check the access point's power class. A client of a low-power indoor access point runs 6 dB below it by rule.
6. Find where along the route the client's received signal crosses its roam trigger, and note what the access point hears from the client at that point.

The measurement guide shows how to probe, capture, and time a drop step by step.

Ben Rutter. "AP and client power mismatch." OT Wireless, published October 8, 2026. https://otwireless.com/causes/ap-and-client-power-mismatch/

Rutter, B. (2026, October 8). AP and client power mismatch. OT Wireless. https://otwireless.com/causes/ap-and-client-power-mismatch/

@misc{rutter2026apandclientpowermismatch,
  author = {Rutter, Ben},
  title = {{AP and client power mismatch}},
  year = {2026},
  howpublished = {\url{https://otwireless.com/causes/ap-and-client-power-mismatch/}},
  organization = {OT Wireless},
}