In-Network Fusion for High Interference Signal Detection within CBRS Band

In this paper, we present an advanced signal detection framework for shared spectrum. The primary goal is to detect of radar pulses and 5G signals, ensuring seamless coexistence in shared spectrum environments, particularly within the Citizens Broadband Radio Service (CBRS) band. We propose a novel in-network fusion-based solution which relaxes the Federal Communications Commission (FCC) requirement of achieving a radar detection accuracy of at least 99% for a signal-to-interference-noise ratio (SINR) ≥ 20 dB to SINR ≥ 0 dB. This adjustment enables a robust detection of radar pulses even in the presence of interfering 5G signals which can introduce up to 20 dB more interference. By utilizing emulated digitized radar signals for spectrogram analysis and adopting our in-network fusion-based approach, we demonstrate 99% signal detection for SINR ≥ 0 dB.