Channel states dependent fair service: a new packet scheduling algorithm for CDMA

  • Authors:
  • Fei Long;Gang Feng;Chee Kheong Siew

  • Affiliations:
  • ICIS Research Lab, School of Electrical and Electronic Engineering, Nanyang Technological University, S2.1 B4-04, Singapore 639697, Singapore;ICIS Research Lab, School of Electrical and Electronic Engineering, Nanyang Technological University, S2.1 B4-04, Singapore 639697, Singapore;ICIS Research Lab, School of Electrical and Electronic Engineering, Nanyang Technological University, S2.1 B4-04, Singapore 639697, Singapore

  • Venue:
  • Computer Networks: The International Journal of Computer and Telecommunications Networking
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we address the problem of packet scheduling for high speed CDMA, which can provide high speed data service for mobile users. We first investigate the appropriate multiple access schemes in high speed CDMA. We find through mathematical analysis that TDM is more efficient than CDM for downlink in high speed CDMA networks in some general situations. Based on this finding, we propose and evaluate an efficient scheduling scheme, namely Channel States Dependent Fair Service for CDMA (CSDFSC), for packet scheduling in high speed CDMA networks. CSDFSC tries to maximize channel throughput under fairness and transmission power constraints. This scheme differentiates best-effort and guaranteed services by assigning different queue weights to various traffic classes. By considering the difference between the pre-defined and actually occupied bandwidth proportion in the priority computation, CSDFSC exhibits good fairness properties for best-effort flows. We compare the performance of CSDFSC with that of WISPER and WRR by using ns-2 simulations, where VBR video and TCP traffic flows are considered. The numerical results show that CSDFSC outperforms them in terms of throughput, link utilization and average end-to-end packet delay, while retaining a low implementation complexity.