Optimal Wireless Information and Power Transfer Using Deep Q-Network

Table 1

DQN simulation parameters.

Dueling Deep Q-network	Value
Number of hidden layers ()	4
Number of nodes of each hidden layer	100
Learning rate ()
Mini-batch size	10
Learning frequency	5
Training starting step	200
Experience pool
Initial exploration rate ()	1
Final exploration rate ()	0.1
Exploration interval	0.001
weight replacement interval	100
Discount factor	0.9
Training episodes	40000