Implications of power control in wireless networks: a quantitative study

  • Authors:

  • Ioannis Broustis;Jakob Eriksson;Srikanth V. Krishnamurthy;Michalis Faloutsos

  • Affiliations:

  • Department of Computer Science and Engineering, University of California, Riverside;Department of Computer Science and Engineering, University of California, Riverside;Department of Computer Science and Engineering, University of California, Riverside;Department of Computer Science and Engineering, University of California, Riverside

  • Venue:
  • PAM'07 Proceedings of the 8th international conference on Passive and active network measurement
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

The use of power control in wireless networks can lead to two conflicting effects. An increase in the transmission power on a link may (i) improve the quality and thus the throughput on that link but, (ii) increase the levels of interference on other links. A decrease in the transmission power can have the opposite effects. Our primary goal in this work is to understand the implications of power control on interference and contention. We conduct experiments on an indoor mesh network. Based on analysis of our experimental data, we identify three interference scenarios: a) the overlapping case, where the aggregate throughput achievable with two overlapping links cannot be improved via power control; b) the hidden terminal case, where proper power control can primarily improve fairness and, c) the potentially disjoint case, where proper power control can enable simultaneous transmissions and thus improve throughput dramatically. We find that power control can significantly improve overall throughput as well as fairness. However, to our surprise, we note that using virtual carrier sensing in conjunction with power control generally degrades performance, often to a large degree.