Classification of Data Corruption in Microcontroller-Based Serial-Optical Communication with Tensorflow-Lite Mikrodenetleyici Tabanlı Seri-Optik Haberleşmede Veri Bozulmasının Tensorflow-Lite ile Sınıflandırılması

Küçük Y., ÇEVİKALP H.

32nd IEEE Conference on Signal Processing and Communications Applications, SIU 2024, Mersin, Turkey, 15 - 18 May 2024, (Full Text) identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/siu61531.2024.10600911
  • City: Mersin
  • Country: Turkey
  • Keywords: artificial intelligence, embedded systems, IEC-62056, machine learning, serial-optical communication, TensorFlow-Lite
  • Eskisehir Osmangazi University Affiliated: Yes

Abstract

In electricity meters located outdoors, sunlight causes disruptions in serial optical communication and communication with the meter cannot be achieved. Dynamically adjusting the radiation intensity provides an alternative solution to this problem. A model was developed to detect the radiation intensity appropriate to the environmental conditions using Tensorflow-Lite. By operating this model in a microcontroller, the radiation intensity required for communication can be adjusted and situations in which communication is not possible can be detected in advance. As a result of this study, it was seen that the TensorflowLite model in the microcontroller enables optical communication by adjusting the radiation intensity according to variable external environmental conditions.