CNN-aided multiple PU spectrum sensing in uncalibrated massive antennas SU system

Spectrum Sensing Algorithm by Multi-Objective Optimization Theory and Fuzzy Integral Method

“…It was compared with the five extended algorithms based on a convolutional neural network. The Monte Carlo calculator (MCS) was used to verify its performance [8]. M Karimi and others intend to study spectrum sensing and resource optimization in multiband cognitive radio (CR) systems.…”

“…In formula (8), u and v obey normal distribution. First, the reverse learning method is introduced into the initialization process of the population to improve population diversity and prevent the algorithm from falling into local extremum; Secondly, through the calculation of the Pareto ranking and crowding distance of the population, 10% of the individuals are selected from the first-order Pareto frontier solution as the global leader, and the diversity of the population is obtained through reverse learning method, and Levy mutation is introduced between the candidate population and the reverse candidate population to produce offspring.…”