Deep Learning for Cooperative Spectrum Sensing
P. Shachi, K. R. Sudhindra, Marco Suma · 2020
The performance of spectrum sensing methods available have the limitations under extreme channel conditions. With the advent of deep learning in various fields, this paper is an attempt to analyze the performance of cooperative spectrum sensing (CSS) wherein deep learning method is adopted for data fusion. We study the performance of convolutional neural network (CNN) based CSS under dynamic channel condition. The dataset is synthesized for the system model considering spatial-correlation among secondary users, path loss with log-normal shadowing and different fading conditions - Rayleigh and Nakagami-m. The performance is compared with the hard fusion rules - OR, AND, Majority rules, in terms of accuracy in classification.