Near-Optimal Selection of GPR Observations for Linear Inversion (Not published)

Wang, J. (Speaker), Yarovyi, O. (Speaker)

Activity: Talk or presentationTalk or presentation at a workshop, seminar, course or other meeting

Description

Dense spatial measurement of Ground Penetrating Radar (GPR) guarantees the sampling criterion for the linear inversion. On the other hand, the oversampled data significantly increases the computational load in the inversion operation, or it requires more complex GPR systems. In this paper, the “frame potential” based approach was suggested to near-optimally select GPR measurements for linear inversion. As it is mainly based on the observation matrix, the suggested approach can be used to select the most informative observations before linear inversion or optimize the GPR spatial sampling criterion. As an example, an antenna array topology optimization problem is presented to
show the effectiveness of this approach.
Period28 Jun 2017
Event title9th International Workshop on Advanced Ground Penetrating Radar
Event typeConference
LocationEdinburgh, United Kingdom

Keywords

  • Ground penetrating radar (GPR)
  • Sensor selec-tion
  • sampling design
  • linear inversion