A Neuro-Fuzzy edge based spectrum sensing processor for cognitive radios

Spectrum sensing is a mechanism in cognitive radio to improve spectral efficiency with the aid of allowing unlicensed users to access the idle portions of spectrum opportunistically. In this paper, a Neuro-Fuzzy (NF) edge based spectrum sensing is proposed. To our best knowledge, it is the first time that a spectrum sensing method based on NF is introduced. The presented method used the built-in thresholding capability of neuro-fuzzy network. The simulations indicate that it can operate perfectly at Signal-to-Noise Ratio (SNR) equals -10 dB while maintaining the probabilities of false-alarm and detection within acceptable bounds. This work is realized in FPGA platform and demonstrates operational performance of 72.67 μs in sensing time interval at clock frequency of 30 MHz.