Service-Oriented Power Allocation for High-Speed Railway Wireless Communications

Tao Li, Ke Xiong, Pingyi Fan, Khaled Letaief

Research output: Contribution to journalArticle

15 Citations (Scopus)


The rapid development of high-speed railways (HSRs) all over the world is drawing much more attention on high-mobility wireless communication. For the wireless links that connect the passengers on the train to the cellular network, it is very essential to employ an appropriate power allocation strategy to guarantee the reliability and efficiency of information transmission. Therefore, this paper concentrates on evaluating the transmission performance of wireless links in the HSRs and attempting to derive an optimal power allocation strategy of this scenario. Considering the fact that the information transmitted between the train and the base station usually has diverse quality-of-service (QoS) requirements for various services, a QoS-based achievable rate region is utilized to characterize the transmission performance in this paper to instead of traditional single throughput. Based on it, a QoS-distinguished power allocation algorithm is derived to achieve the largest achievable rate region. It is proved that the traditional water-filling algorithm and the channel inverse algorithm can be regarded as two specific cases of this new algorithm. Besides, the robust performance of the proposed strategy is also discussed in detail under a non-uniform motion scenario, and its relative performance loss is evaluated in terms of energy consumption. Finally, we present a typical implementation example in a Rice fading environment when the data rate requirements are prior known.

Original languageEnglish
Article number7924308
Pages (from-to)8343-8356
Number of pages14
JournalIEEE Access
Publication statusPublished - 1 Jan 2017
Externally publishedYes



  • Diverse quality-of-service (QoS) requirements
  • hybrid information
  • optimal power allocation algorithm
  • QoS-based achievable rate region

ASJC Scopus subject areas

  • Computer Science(all)
  • Materials Science(all)
  • Engineering(all)

Cite this