eduzhai > Applied Sciences > Engineering >

Configuration Learning in Underwater Optical Links

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 25 pages

Abstract: A new research problem named configuration learning is described in thiswork. A novel algorithm is proposed to address the configuration learningproblem. The configuration learning problem is defined to be the optimizationof the Machine Learning (ML) classifier to maximize the ML performance metricoptimizing the transmitter configuration in the signal processing communicationsystems. Specifically, this configuration learning problem is investigated inan underwater optical communication system with signal processing performancemetric of the physical-layer communication throughput. A novel algorithm isproposed to perform the configuration learning by alternating optimization ofkey design parameters and switching between several Recurrent Neural Network(RNN) classifiers dependant on the learning objective. The proposed MLalgorithm is validated with the datasets of an underwater optical communicationsystem and is compared with competing ML algorithms. Performance resultsindicate that the proposal outperforms the competing algorithms for binary andmulti-class configuration learning in underwater optical communicationdatasets. The proposed configuration learning framework can be furtherinvestigated and applied to a broad range of topics in signal processing andcommunications.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×