4th International ICST Conference on Wireless Internet

Research Article

Transmission Power Selection for Ad hoc Networks

Download553 downloads
  • @INPROCEEDINGS{10.4108/ICST.WICON2008.4853,
        author={Jianping He and Jiahai Yang and Changqing An and Xuenong Li},
        title={Transmission Power Selection for Ad hoc Networks},
        proceedings={4th International ICST Conference on Wireless Internet},
        keywords={Ad hoc networks BER SINR Power selection},
  • Jianping He
    Jiahai Yang
    Changqing An
    Xuenong Li
    Year: 2010
    Transmission Power Selection for Ad hoc Networks
    DOI: 10.4108/ICST.WICON2008.4853
Jianping He1,*, Jiahai Yang1,*, Changqing An1,*, Xuenong Li1,*
  • 1: Network Research Center, Tsinghua University Tsinghua National Laboratory for Information Science and Technology, Beijing, P.R. China, 100084
*Contact email: hejp06@mails.tsinghua.edu.cn, yang@cernet.edu.cn, acq@tsinghua.edu.cn, lxn@tsinghua.edu.cn


In wireless ad hoc networks, as an important technology, power control is widely used to increase energy efficiency and improve network throughput. Power control is composed of two components: one component handles the problems caused by asymmetric channel; the other component is responsible for selecting transmission power. In this paper, we focus on transmission power selection for ad hoc networks using 802.11 DCF MAC protocol. In this paper, we model the network with power control in joint with PHY layer and MAC layer. Then we prove that in considering the noise, interference from other nodes and packet length, it is NP-hard to obtain the optimal transmission power for every node even if it knows the complete knowledge of the network. After that, we propose two heuristic transmission power selection algorithms, one for energy-oriented power control and the other for throughputoriented power control, which are time efficiency and able to improve the network performance largely. Finally, we evaluate our algorithms under different topologies and in different packet size. The results show that our algorithms can improve the network performance largely when compared with 802.11 and traditional power control algorithms.