Performance Analysis and Optimization of CRNs Based on Fixed Feedback Probability Mechanism with Two Classes of Secondary Users
Table 1
PSO algorithm to find the optimal value of feedback probability.
Algorithm: find the optimal value of feedback probability
Step 1
Initialize the number of particles t, acceleration factors and , maximum value of the speed, maximum value of the inertia weight and the minimum value , maximum iteration number , upper limit and the lower limit of the feedback probability
Step 2
Set a random initial feedback probability and initial velocity for each particle, and make the value satisfy ,; set the initial iteration number to 0
Step 3
Formula (25) is used to calculate the cost function value corresponding to the feedback probability of each particle
Step 4
Search for the current individual optimal value (the minimum value of the cost function) of the particle and the corresponding individual optimal probability
Step 5
Find the minimum value of the cost function in the whole particle swarm, that is, the global optimal value, and assign the feedback probability corresponding to the global optimal value to , i.e.,
Step 6
Calculate the weight according to the formula
Step 7
Update the speed and feedback probability of each particle. Take the particle as example: ,, where and are an random variables uniformly distributed from 0 to 1, and is kept in the range
Step 8
The number of iterations is increased by one, i.e., ; if , return to the third step; otherwise, output the optimal feedback probability