Research Information System
Thesis Type: Postgraduate
Institution Of The Thesis: Gazi University, Fen Bilimleri Enstitüsü, Turkey
Approval Date: 2025
Thesis Language: Turkish
Student: Gökhan KAYA
Supervisor: Özgür Ertuğ
Open Archive Collection: AVESIS Open Access Collection
Abstract:
This thesis presents the development of a GPS-supported, frequency-hopping, location based communication system designed for use across Turkey. Based on Software Defined Radio (SDR) technology, the system is specifically aimed at minimizing frequency interference and ensuring uninterrupted communication. Turkey is divided into 8 and 16 geographical regions, with each region assigned different starting frequencies. While each region operates within the same 50 MHz bandwidth, the starting frequency for frequency hopping is determined uniquely for each region. The system was developed using HackRF devices and the GNU Radio platform to establish communication between two radio platforms. By utilizing GPS time data, the system dynamically adjusts the frequency hopping process based on the geographic location. Furthermore, a deep learning algorithm operates separately for each region, learning the local radio frequency usage and enabling the system to avoid heavily used or interference-prone frequencies. In particular, frequencies like 136 MHz, which are heavily used, are recognized by the deep learning model, preventing the system from hopping to these frequencies. The system offers a reliable, high-quality, and efficient communication infrastructure for both military and civilian applications. This study makes significant contributions to optimizing spectrum usage, particularly in regions with high radio traffic and heavy interference. By presenting an effective approach to spectrum management, the system aims to enhance communication quality and minimize interference.
Key Words : GPS-supported communication systems, frequency hopping, deep learning, Software Defined Radio (SDR), interference avoidance, spectrum management, HackRF