基于深度强化学习的多基站计算卸载策略

Multi base station computing offload strategy based on deep reinforcement learning

针对多个基站覆盖下的移动用户如何选择最佳的卸载策略问题,提出一种与匹配博弈论结合的双延迟深度确定性策略梯度(GT-TD3)算法来实现用户计算任务的卸载。以最小化时延及能耗为目标提出优化问题,使用多基站博弈算法求解用户与其对应卸载基站的最佳匹配方案,使用深度强化学习算法求解系统开销最小化的优化问题。仿真结果表明,所提算法相较4种对比算法可以有效减少MEC系统总花费,为移动用户提供更高效的服务。

Aiming at the problem that how to choose the best offloading strategy for mobile users under the coverage of multiple base stations,a dual-delay deep deterministic policy gradient(GT-TD3)algorithm combined with matching game theory was proposed to realize offloading of user computing tasks.The optimization problem was proposed with the goal of minimizing the delay and energy consumption.The multi-base station game algorithm was used to solve the optimal matching scheme between users and their corresponding unloaded base stations.The deep reinforcement learning algorithm was used to solve the optimization problem of minimizing system overhead.Simulation results show that,compared with the three baseline algorithms,the proposed algorithm can effectively reduce the delay and energy consumption of the MEC system,and provide more efficient services for mobile users.