NB-IoT wireless resource allocation method based on NOMA and multi-agent reinforcement learning

The invention discloses an NB-IoT (Narrow Band Internet of Things) wireless resource allocation method based on NOMA (Non-Orthogonal Multiple Access) and multi-agent reinforcement learning, and aims to solve the problem of connection density maximization in a multi-user NB-IoT scene based on an NOMA technology. Considering a more practical scene, different users have different QoS requirements and use different tone types. Different from a traditional heuristic algorithm, the method firstly models a joint optimization problem of power, resource block allocation and NOMA user pairing into a Markov decision process, and adopts an MAPPO, which is the most advanced multi-agent reinforcement learning algorithm, to solve the joint optimization problem. In consideration of the existing invalid actions, the invalid actions are shielded in the process of calculating the probability distribution of each action by the neural network, so that convergence is accelerated. Python3.8.15 is used on a VScode platform for syste