Support #4057

verify that re-using ARFCN+BSIC pairs within a BSS works

Added by neels over 1 year ago. Updated about 1 year ago.

Target version:
Start date:
Due date:
% Done:


Spec Reference:


Any ARFCN+BSIC pair may be used any number of times in a BSS.
If a given cell reports good RXLEV for a specific ARFCN+BSIC, the eligible target cells shall be only those listed as explicit neighbors for the source cell.
However, in handover_start(), it looks like any cell with a matching ARFCN+BSIC would be picked by bts_by_neighbor_ident().

Verify in a test that explicit neighbors work as expected, even if their ARFCN+BSIC is used multiple times within the BSS,
i.e. that only the specific cell which is listed as explicit neighbor is picked as target, not another one with the same ARFCN+BSIC that is not listed as explicit neighbor for the handover source cell.

The OsmoBSC manual claims that this works, but the code to me looks like it wouldn't.


#1 Updated by laforge about 1 year ago

  • Assignee changed from neels to osmith

#2 Updated by osmith about 1 year ago

  • Status changed from New to Resolved
  • % Done changed from 0 to 100

neels replaced the mentioned function bts_by_neighbor_ident() with a new function find_handover_target_cell(), that resolves exactly this problem:

You've also added a related test already, "f_tc_ho_neighbor_config_3":

HO to bts 2 works, explicitly listed as neighbor; no ambiguity because bts 3 is not listed as neighbor

#3 Updated by neels about 1 year ago

that's accurate. Apologies, it seems I failed to update this issue back then :/

Also available in: Atom PDF

Add picture from clipboard (Maximum size: 48.8 MB)