haibomei / uav-ris-drl Goto Github PK

In the file of environment modeling, two boolean values RIS and P_shift are generated in the link rate function. I want to ask the meaning of these two boolean values and the meaning of the formulas in the three if branches generated by these two boolean values, because I haven't seen them in the paper.

The formula for the optimal phase shift is not reflected, M=100
On the contrary, M, which does not have an optimal phase shift, is calculated using a formula

代码中finish的结束条件为什么设置成累积吞吐量是否小于uk,是则不结束，那么一旦有一个节点满足了，不就会结束整个任务，但是为什么最终代码的平均吞吐量相差这么多

Hi,

I have read the original paper, which is a interesting work! However, I think there are some issues, which make me feel confused. Thus, I wish to discuss with you about the details. Potential issues are as follows:

1. For DDPG scheme, I don't find how to predict the user selection by it, which is a {0,1} discrete problem.
2. How can both DRL scheme meet all constraints listed in formula 8?
3. The number of time slot is 600. For DRL scheme, finishing the whole communication task is a episode. However, the algorithm you describe is not like the rule of DRL.