Performance of Deep Learning Assisted Visible Light Communications Impaired by Blockages

Shaik P., Keçeci C., Garg K. K., BOYACI A., Ismail M., Serpedin E.

2023 IEEE Global Communications Conference, GLOBECOM 2023, Kuala-Lumpur, Malezya, 4 - 08 Aralık 2023, ss.4704-4709, (Tam Metin Bildiri) identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/globecom54140.2023.10436826
  • Basıldığı Şehir: Kuala-Lumpur
  • Basıldığı Ülke: Malezya
  • Sayfa Sayıları: ss.4704-4709
  • Anahtar Kelimeler: Blockage, BPSK, Deep Learning, QAM, VLC
  • İstanbul Ticaret Üniversitesi Adresli: Evet

Özet

This study investigates the performance of visible light communications (VLCs) in the presence of blockages. An indoor office scenario with a single VLC access point serving the user nodes in the presence of human blockages is examined. System performance is assessed through closed-form expressions for outage probability and symbol error rate for binary phase shift keying and quadrature amplitude modulation. A deep neural network for symbol detection is deployed at the receiver. Performance metrics illustrate that the blockages cause significant impact on signal detection. Computer simulations corroborate the correctness of the obtained analytical expressions.