eduzhai > Applied Sciences > Engineering >

Semi-Supervised Learning with GANs for Device-Free Fingerprinting Indoor Localization

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 6 pages

Abstract: Device-free wireless indoor localization is a key enabling technology for theInternet of Things (IoT). Fingerprint-based indoor localization techniques area commonly used solution. This paper proposes a semi-supervised, generativeadversarial network (GAN)-based device-free fingerprinting indoor localizationsystem. The proposed system uses a small amount of labeled data and a largeamount of unlabeled data (i.e., semi-supervised), thus considerably reducingthe expensive data labeling effort. Experimental results show that, as comparedto the state-of-the-art supervised scheme, the proposed semi-supervised systemachieves comparable performance with equal, sufficient amount of labeled data,and significantly superior performance with equal, highly limited amount oflabeled data. Besides, the proposed semi-supervised system retains itsperformance over a broad range of the amount of labeled data. The interactionsbetween the generator, discriminator, and classifier models of the proposedGAN-based system are visually examined and discussed. A mathematicaldescription of the proposed system is also presented.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×