HO2: handover decision for dynamic timeslots
The recent handover_decision_2.c hasn't been duly tested in the presence of dynamic timeslots.
Particularly TCH/F_TCH/H_PDCH timeslots are potentially both TCH/F or TCH/H, which might require another spin on the handover_decision_2.c code.
Aspects to consider:It is benificial if PDCH are in consecutive timeslots.
- If dynamic timeslots are in use, it makes sense to trigger re-assignment to move them as far away from dedicated PDCH as possible.
- If a dedicated TCH timeslot becomes available, it makes sense to re-assign away from a dyn TS towards a dedicated timeslot.
An unused dynamic timeslot may count as both 1 free TCH/F and 2 free TCH/H. If one TCH/H becomes used, this implicitly also reduces the number of free TCH/F. The lchan picking algorithm must not get confused by that.
A dynamic timeslot being available as PDCH is benificial to GPRS service. So, if a neighboring cell is unloaded and has good rxlev, we may want to handover away from dynamic timeslots to make room for PDCH. rxlev vs PDCH tradeoff should be configurable.