Joint Optimization of Jamming Link and Power Control in Communication Countermeasures: A Multiagent Deep Reinforcement Learning Approach

Table 4

The comparison of the maximum success jamming rate among different approaches of each link.

Type	Method
	MASAC	MADDPG	CA-DDPG
Link 1	0.8616	0.8692	0.9102
Link 2	0.9448	0.9632	0.9308
Link 3	0.9616	0.9866	0.9652
Link 4	0.8988	0.9106	0.9276
Link 5	0.9676	0.9574	0.9468
Simultaneously jamming all links	0.8694	0.7824	0.7760