Dynamic Spectrum Access Based on SDR Frequency Offset for Drone Communication Under Public Protection and Disaster Relief

The contradiction between spectrum shortage and spectrum underutilization has inspired active research on Dy-namic Spectrum Access (DSA). DSA uses unused white space spectrum to achieve greater efficiency on a limited spectrum. DSA requires the use of software-defined radio (SDR) to change the frequency band to white space. However, SDR requires tuning time to change the frequency band, and there is no time synchronization between SDR devices. In this article, we use SDR to change different frequency bands, use SDR tuning time to prevent SDR from improperly competing for spectrum with other SDRs, and reduce interference caused by multiple SDRs using the same frequency band at the same time. The SDR used by DSA in the process of changing frequency bands will generate tuning time, but we noticed that additional tuning time was observed when the SDR switched between the PPDR frequency band (694–894 MHz) and the telecom carrier band (1–6 GHz). Correctly designing the tuning time between distributed SDR devices is critical to DSA. We change different frequency bands according to each SDR, record the tuning time and additional tuning time of each SDR, and use reinforcement learning to control the allocation of blank spectrum to coordinate the use of idle spectrum.