Backward handover can be described as network-controlled/UE-assisted mobility. Handover related information is exchanged between the UE and the source eNB via the old radio path (thus, the usage of the term‘backward’).
Specifically, the radio conditions need to be good enough for the source eNB to be able to decode the Measurement Report from the UE and subsequently prepare the target cell for handover. The radio conditions also need to be good enough for the UE to be able to decodethe Handover Command from the source eNB.
There is a short interruption in service between the time that the UE decodes the Handover Command from the source eNB and the time that the target eNB decodes the Handover Confirm from the UE. However, data forwarding and in-order delivery ensures that none of the data buffered in the source eNB is lost.
Forward handover can be described as UE-based mobility. Handover related information is exchanged between the UE and target eNB via the new radio path after the UE context is fetched by the target eNB from the source eNB (thus, the usage of the term ‘forward’).
Forward handover is successful even if the radio conditions are not good enough for the source eNB to be able to decode the Measurement Report from the UE and prepare the target cell. The success of the handover procedure even with complete failure of signaling with the source eNB makes forward handover robust to rapidly changing signal strength conditions.
This will still incur an additional delay versus the backward handover procedure and,
consequently, a longer interruption in service.
