Cooperative Spectrum Sensing With Ensemble Empirical Mode Decomposition and Dual-Channel Composite Neural Network for Complex Environments

Conventional spectrum sensing (SS) algorithms demonstrate subpar detection performance in complex environments. In this study, we use a strategy that integrates deep learning (DL) with signal-denoising techniques. To improve detection performance and reduce noise interference, the ensemble empirical mode decomposition (EEMD) algorithm is used to eliminate noise components from the signal. Subsequently, a Riemannian mean-based algorithm is used to fuse multiple sensing data in a multiantenna system. The fused data serve as inputs to a dual-channel composite neural network (DCCNN), leading to the development of an EEMD-DCCNN-based SS algorithm. Experimental results demonstrate that the proposed algorithm significantly outperforms traditional algorithms regarding detection performance. In particular, the proposed algorithm achieves a detection probability of 94.6% and a false alarm probability of 3.5% at SNR <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$= -18$ </tex-math></inline-formula> dB.