İHA Operasyonlarında Uçta Hesaplama Destekli Anlamsal Veri İletimi

Göker E., Kahveci S.

URSI-TÜRKİYE 2026 XIII. Bilimsel Kongresi, Kastamonu Üniversitesi, Kastamonu, Türkiye, 7 - 09 Eylül 2026, ss.15-18, (Tam Metin Bildiri)

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Kastamonu
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.15-18
  • Karadeniz Teknik Üniversitesi Adresli: Evet

Özet

This study investigates task information transmission from UAV camera images processed with

YOLOv5 on NVIDIA Jetson Xavier NX over an 868 MHz LoRa P2P link. The detector was trained on a custom,

manually labelled dataset. LoRa used SF7, 125 kHz bandwidth and 14 dBm power; a 5 GHz Loco5AC video link

served as an independent reference under the same field conditions. In line-of-sight flight near 50 m altitude,

about -108 dBm RSSI was recorded at 5 km. For the eight-byte application payload, the Semtech time-on-air

model gives approximately 36.10 ms theoretical airtime. Findings characterize only RSSI and observed videolink

behavior.