Deep Reinforcement Learning-Based Joint Optimization for Transmitted Power of D2D User and Cellular User
Targeting at the interference between D2D user and cellular user in D2D communication underlay cellular network system,deep enhancement learning-based transmission power optimization(DTPO)algorithm is proposed.The interference is mitigated by opti-mizing the transmit power of the devices.The power allocation problem is generally modeled as a NP-hard combinatorial optimization problem with linear constraint.Then deep reinforcement learning(DRL)algorithm is used to optimize the transmit power for both D2D users and cellular users,and the sum-rate is maximized.Simulation results show that DTPO algorithm affords similar performance with exhaustive search algorithm.
D2D communication underlay cellular network systeminterferencetransmitted powerdeep reinforcement learningsum-rate
国家科技期刊平台
NSTL
万方数据
