A resource allocation method and system based on imitation learning

Abstract

This application discloses a resource allocation method and system based on imitation learning. The resource allocation method specifically includes the following steps: establishing a system model; determining an objective function based on the established system model; optimizing the determined objective function; obtaining a suboptimal strategy based on the optimized objective function; optimizing the suboptimal strategy to obtain an optimal strategy; and allocating resources according to the optimal strategy. This application considers the uneven distribution of users and different channel conditions in a single cellular cell, as well as the interference between cellular users and D2D users caused by multiplexing. It performs continuous interference cancellation at the receiver, considers user fairness, and performs power allocation. Through joint optimization, the final power allocation result is faster and more accurate.

Images