| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
Dear Vladimir,
I wrote a short answers for your reply comments to the comment #418.
Please see the colored text below.
Best Regards,
Seung Joo
1. It is assumed [though not said explicitly] that collaborating BSs have to transmit identical PHY bursts [not "packets"]. So all MAC PDUs in the bursts must be the same. To accomplish that, MAC instants at the BSs probably need operate with same SDUs and perform same operations [fragmentation, packing, adding sub-headers, PHS, ARQ]. For example, how two BSs may decide that two SDUs at hand are two copies of the same SDU? SDUs are not supposed to carry any ID. It is not explained where such synchronization comes from. It is hard to imagine practical implementation of transceivers exchanging [probably very fast] signals to reach complete synchronization of MAC state machines.
=> Yes, you are right. To support SHO, a kind of central controller may be required, but this issue should be discussed in NET-MAN adhoc. The explanation of the detailed synchronization operation is beyond the scope of this contribution. In this contribution, we simply consider BSs which can support S-HO and try to provide possible ways to do so.
2. Another problem is possible CID collisions that may require reassignment of CIDs to service flows, even for those not related to the SS entering soft combining. An air interface mechanism is needed to resolve such collisions.
=> We added additional MAP-IEs(e.g., HO_Anchor_Active_DL/UL_MAP_IE,
HO_Active_Anchor_DL/UL_MAP_IE) to resolve the CID collision issue in the revised
contribution.