ChinaCom2009-Wireless Communications and Networking Symposium

Research Article

Frame Synchronization and Buffer Overflow Control Schemes for Handover in HSDPA Networks

  • @INPROCEEDINGS{10.1109/CHINACOM.2009.5339717,
        author={Yeong-Sheng Chen and Mu-Sheng Liao},
        title={Frame Synchronization and Buffer Overflow Control Schemes for Handover in HSDPA Networks},
        proceedings={ChinaCom2009-Wireless Communications and Networking Symposium},
        keywords={e Synchronization; Buffer Overflow Control; Handover; High-Speed Downlink Packet Access (HSDPA)},
  • Yeong-Sheng Chen
    Mu-Sheng Liao
    Year: 2009
    Frame Synchronization and Buffer Overflow Control Schemes for Handover in HSDPA Networks
    DOI: 10.1109/CHINACOM.2009.5339717
Yeong-Sheng Chen1,*, Mu-Sheng Liao1,*
  • 1: Department of Computer Science National Taipei University of Education Taipei 106, TAIWAN
*Contact email:,


In the high-speed downlink packet access (HSDPA) mechanism, the radio network controller (RNC) sends frames to all cells in the active set. A user equipment (UE) communicates with one cell, which is called serving cell, in the active set of cells. When the UE performs handover (switches from the old serving cell to the new serving cell), the new serving cell must be informed the status of the buffer in the old serving cell so as to transmit the next correct frames to the UE. This is referred as frame synchronization. Also, it should be assured that when the buffer of a nonserving cell is full, the previously received frames in the buffer can be safely dropped without causing any lost of frames during frame synchronization. This is referred as buffer overflow control. This paper investigates the frame synchronization and buffer overflow control mechanisms and proposes more efficient and complete schemes than the previous work. For the buffer overflow control scheme, we derive a better lower bound on the buffer size of the nonserving cell. Therefore, the system resource can be better utilized. For the frame synchronization scheme, we use the circular queue to implement the buffer of the old serving cell. With the proposed scheme, the frame synchronization information can be represented as a pointer of the circular queue and a status variable. Thus, the size of this information is very small and can be efficiently transmitted to the new serving cell through one uplink channel, which is of limited bandwidth. This scheme can reduce the time for frame synchronization processes and therefore will enhance the system performance. Analytical analysis and illustrative examples show that the proposed scheme is more efficient and complete than the previous work.