For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Performance of a Burst Switching Scheme for CDMA-Based Wireless Packet Data Systems
Sung Kyung KIM Meejoung KIM Chung Gu KANG
IEICE TRANSACTIONS on Communications
Publication Date: 2003/03/01
Print ISSN: 0916-8516
Type of Manuscript: PAPER
Category: Wireless Communication Switching
radio resource management, burst switching, QBD analysis, packet scheduling, code division multiple access,
Full Text: PDF(764.7KB)>>
Emerging requirements for higher rate data services and better spectrum efficiency are the main issues of third-generation mobile radio systems. In particular, a new concept of burst switching has been introduced for supporting the packet data services in the CDMA-based wireless system. In the burst switching system, radio resources are allocated to users for the duration of data bursts, which is a series of packets, as opposed to the conventional packet switching scheme. To implement the burst switching scheme, three different states (active, control hold, dormant states) are defined and two transition timers are employed to release the fundamental and supplemental code channels, respectively, at certain instances. Furthermore, the system is subject to burst admission control policy, with which a burst is admitted only when the number of currently available channels is greater than the admission threshold. Since there exists a trade-off between the additional packet access delay during a burst and resource utilization depending on the time-out value of the transition timer and burst admission threshold, it is critical to understand the performance characteristics in terms of the underlying design parameters. In this paper, we develop an analytic model and present a Quasi-Birth-Death (QBD) queueing analysis for evaluating the performance of burst switching schemes. This work focuses on the trade-off studies for optimizing the time-out value of the transition timer so as to minimize the average delay performance. Theoretical performance measures are derived by means of the matrix geometric method and furthermore, some simulation results are presented to validate the proposed analytical approach.