Search

In this paper, we study how to optimize the federated edge learning (FEEL) in UAV-enabled Internet of things (IoT) for B5G/6G networks, from a deep reinforcement learning (DRL) approach. The federated learning is an effective framework to train a shared model between decentralized edge devices or servers without exchanging raw data, which can help protect data privacy. In UAV-enabled IoT networks, latency and energy consumption are two important metrics limiting the performance of FEEL. Although most of existing works have studied how to reduce the latency and improve the energy efficiency, few works have investigated the impact of limited batteries at the devices on the FEEL. Motivated by this, we study the battery-constrained FEEL, where the UAVs can adjust their operating CPU-frequency to prolong the battery life and avoid withdrawing from federated learning training untimely. We optimize the system by jointly allocating the computational resource and wireless bandwidth in time-varying environments. To solve this optimization problem, we employ a deep deterministic policy gradient (DDPG) based strategy, where a linear combination of latency and energy consumption is used to evaluate the system cost. Simulation results are finally demonstrated to show that the proposed strategy outperforms the conventional ones. In particular, it enables all the devices to complete all rounds of FEEL with limited batteries and meanwhile reduce the system cost effectively.

In this paper, we investigate mobile edge computing (MEC) networks for intelligent internet of things (IoT), where multiple users have some computational tasks assisted by multiple computational access points (CAPs). By offloading some tasks to the CAPs, the system performance can be improved through reducing the latency and energy consumption, which are the two important metrics of interest in the MEC networks. We devise the system by proposing the offloading strategy intelligently through the deep reinforcement learning algorithm. In this algorithm, Deep Q-Network is used to automatically learn the offloading decision in order to optimize the system performance, and a neural network (NN) is trained to predict the offloading action, where the training data is generated from the environmental system. Moreover, we employ the bandwidth allocation in order to optimize the wireless spectrum for the links between the users and CAPs, where several bandwidth allocation schemes are proposed. In further, we use the CAP selection in order to choose one best CAP to assist the computational tasks from the users. Simulation results are finally presented to show the effectiveness of the proposed reinforcement learning offloading strategy. In particular, the system cost of latency and energy consumption can be reduced significantly by the proposed deep reinforcement learning based algorithm.