Optimization of LoRa SF Allocation Based on Deep Reinforcement Learning

Figure 7

Different learning rate of training DQN.