v5.3 of ESP-IDF reports notably lower RSSI than v5.1 (IDFGH-13966)

[mbratch]

Answers checklist.

• I have read the documentation ESP-IDF Programming Guide and the issue is not addressed there.
• I have updated my IDF branch (master or release) to the latest version and checked that the issue is present there.
• I have searched the issue tracker for a similar issue and not found a similar issue.

IDF version.

v5.3.1

Espressif SoC revision.

Chip is ESP32-PICO-V3 (revision v3.0)

Operating System used.

Windows

How did you build your project?

Command line with Make

If you are using Windows, please specify command line type.

PowerShell

Development Kit.

ESP32 Dev Kit 1

Power Supply used.

USB

What is the expected behavior?

After upgrading my ESP-IDF to v5.3.1 (from v5.1), I would expect to see no significant change in RSSI readings if the wifi environment hasn't changed.

What is the actual behavior?

After upgrading my ESP-IDF to v5.3.1 (from v5.1), what I see is that the RSSI reported by ESP-IDF using the esp_wifi_sta_get_ap_info function is significantly lower after the upgrade. Specifically, the RSSI with v5.1 was between -70 and -60. After the upgrade to v5.3.1 it dropped to between -127 and -80.

Steps to reproduce.

1. Using v5.1 ESP-IDF operate software and report RSSI
2. Observe RSSI between -70 and -60.
3. Upgrade to v5.3.1 ESP-IDF and operate software
4. Observe RSSI is now between -127 and -80.

Debug Logs.

No response

More Information.

No response

[zhangyanjiaoesp]

@mbratch
What if you put the phy lib (idf/components/esp_phy/lib/esp32)under v5.1.4 on v5.3? Will the RSSI change to be consistent with 5.1.4?

[AxelLin]

@mbratch What if you put the phy lib (idf/components/esp_phy/lib/esp32)under v5.1.4 on v5.3? Will the RSSI change to be consistent with 5.1.4?

@zhangyanjiaoesp Did you try it? This seems something you can easily verify.

[zhangyanjiaoesp]

@mbratch
We have already conducted internal testing(using the STA example and softAP example ) and did not observe such RSSI differences between the two versions you mentioned. And our colleagues mentioned that there haven't been any changes to the PHY library regarding RSSI in these two versions. Did you use deep sleep? Could you share how you performed your testing?

[mbratch]

@zhangyanjiaoesp thank you very much for the prompt reply and the information regarding internal testing.

I did not perform a deep sleep, and my test process was not very formal but just observational looking at an RSSI indicator that I defined and my units under test (I have four of them) are stationary in a test setup. Of course, with wifi, there are natural fluctuations, and I tried to take that into account. But based upon your comments, I suspect perhaps I didn't capture enough data over a period of time to firmly determine what is going on.

I will run more tests when I have the opportunity.