Fair scheduling in wireless packet networks
IEEE/ACM Transactions on Networking (TON)
Downlink scheduling in CDMA data networks
MobiCom '00 Proceedings of the 6th annual international conference on Mobile computing and networking
IEEE/ACM Transactions on Networking (TON)
Packet service in UMTS: delay-throughput performance of the downlink shared channel
Computer Networks: The International Journal of Computer and Telecommunications Networking - 01/14/2002
Wcdma for Umts
WCFQ: an opportunistic wireless scheduler with statistical fairness bounds
IEEE Transactions on Wireless Communications
Convergence of proportional-fair sharing algorithms under general conditions
IEEE Transactions on Wireless Communications
Cross-layer scheduling strategy for UMTS downlink enhancement
IEEE Communications Magazine
Joint power control and intracell scheduling of DS-CDMA nonreal time data
IEEE Journal on Selected Areas in Communications
Supporting rate guarantee and fair access for bursty data traffic in W-CDMA
IEEE Journal on Selected Areas in Communications
| Hi-index | 0.02 |
In Universal Mobile Telecommunication Systems (UMTS), the Downlink Shared Channel (DSCH) may be used to provide streaming services. The traffic model for streaming services is different from the continuously backlogged model used in much of the literature. Each connection specifies a required service rate over an interval of time. In this paper, we are interested in determining how k DSCH frames should be allocated among a set of I connections. We need a scheduler that is channel-aware, so that channels presently enjoying low fading losses can be exploited to achieve higher aggregate throughput. On the other hand, the scheduler is also required to be fair, so that each connection obtains a throughput as close as possible to what it requires. We introduce the notion of discrepancy to capture the inherent trade-off between aggregate throughput and fairness. We show that the discrepancy criterion provides a flexible means for balancing efficiency, as measured by aggregate throughput, and fairness. Our problem, then, is to schedule mobiles so as to minimize the discrepancy over the control horizon. We provide a simple low-complexity heuristic called ITEM that is provably optimal in certain cases. In particular, we show that ITEM is optimal when applied in the UMTS context. Next, we compare the performance of ITEM with that of other algorithms, and show that it performs better in terms of both fairness and aggregate throughput. Thus, ITEM provides benefits in both dimensions--fairness and efficiency--and is therefore a promising algorithm for scheduling streaming connections.