Multiagent (not deep) reinforcement learning? Modeling the problem

Question

I have N number of agents/users accessing a single wireless channel and at each time, only one agent can access the channel and receive a reward.

Each user h