HPPLMN search - difference mfw 1.3.1 and 1.3.2

Sorry to bother again, hopefully this is the last question related to my others ones according the details of HPPLMN search:

nRF9160 SICA B1A
NB-IoT, PSM: PTAU 24h, Act 4s
SIM-Card: IMSI 90140..., HPPLMN 26201 interval: 2h

Using "%XDATAPRFL=0" for devices statically located, I observe, that the mfw 1.3.1 doesn't execute a HPPLMN search, while using the mfw 1.3.3 executes them frequently.

In both cases I used a different nRF9160 SICA B1A with the same application, but with different modem firmware.

The 1.3.2 changelog contains:

- Fixed an issue where modem’s network selection algorithm did not correctly handle global PLMN IDs,
  i.e. PLMN IDs starting with ‘9’. This could have caused delays in network selection in scenarios
  where such global networks were available.

So: does this fix affect the behavior of HPPLMN searches, if "%XDATAPRFL=0" is used? Is this change with "%XDATAPRFL=0" intended?

This change causes devices to take 10 times more energy, when that update is applied.

As I already wrote, FMPOV, it makes using a global SIM for stationary cases less attractive. I didn't found 3GPP notes about the special way of handling the 9 global SIMs, but that may also be caused of the larger amount of 3GPP specs.

Parents Reply Children
  • Hello, 

    I've asked the modem team, who provided answers as in your previous HPPLMN cases

    The mentioned fix was related to network selection algorithm in general, including HPPLMN search.

    %XDATAPRFL is not related if the device is stationary and in a stable radio conditions (no need to search for better cell).

    The fix explains the difference. MFW 1.3.1 did not recognize global PLMN IDs which is why global PLMN IDs were not considered as suitable higher priority PLMNs.

    So, if the home network (MCC starting with 9) is the only preferred network in the SIM card, it is not ever searched unless already camped on PLMN with the same MCC. As I said earlier, only networks from the current country are considered as suitable candidates in HPPLMN search. The fix in 1.3.2 was that PLMN starting with 9 is a global PLMN and can be found from anywhere and that is why it is also considered as suitable candidate in HPPLMN search. That is why the search is now initiated with MFW 1.3.2.

    Hope this provides some answer to you.


    Kind regards,
    Øyvind

  • > That is why the search is now initiated with MFW 1.3.2.

    Thanks a lot. So my observation updating to MFW 1.3.2 is true.

Related