A dynamic programming method for optimal scheduling in high speed downlink packet access

  • Authors:
  • Patrik Björklund;Peter Värbrand;Di Yuan

  • Affiliations:
  • Linköping Institute of Technology, Norrköping, Sweden;Linköping Institute of Technology, Norrköping, Sweden;Linköping Institute of Technology, Norrköping, Sweden

  • Venue:
  • Mobility '06 Proceedings of the 3rd international conference on Mobile technology, applications & systems
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

High Speed Downlink Packet Access (HSDPA) is a data service in Wideband Code Division Multiple Access (WCDMA) networks. Resources shared among users accessing network service through HSDPA include transmission power and channelization codes. Typically, heuristic rules are used for resource allocation in HSDPA. In this paper we study optimal HSDPA scheduling. We present a mathematical model for this problem. The model involves maximizing a utility function which can be used to achieve a balance between system throughput and user fairness. The constraints in the model account for limited amount of resource in transmission power and channelization codes. To solve the model, we propose a dynamic programming method. The method has a predictable execution time, and is guaranteed to find the global optimum provided that input data has a sufficiently good granularity. Moreover, the method allows for trading solution precision for execution time. We present a simulation study to compare optimized HSDPA scheduling to some other scheduling strategies, including the round robin scheme and two heuristics. Our simulation results show that optimized scheduling leads to a similar degree of user fairness but significantly higher data throughput in comparison to heuristic schemes.