Internet of Things (IoT) devices can apply mobile edge computing (MEC) and energy harvesting (EH) to provide high-level experiences for computational intensive applications and concurrently to prolong the lifetime of the battery. In this paper, we propose a reinforcement learning (RL) based offloading scheme for an IoT device with EH to select the edge device and the offloading rate according to the current battery level, the previous radio transmission rate to each edge device, and the predicted amount of the harvested energy. This scheme enables the IoT device to optimize the offloading policy without knowledge of the MEC model, the energy consumption model, and the computation latency model. Further, we present a deep RL-based offloading scheme to further accelerate the learning speed. Their performance bounds in terms of the energy consumption, computation latency, and utility are provided for three typical offloading scenarios and verified via simulations for an IoT device that uses wireless power transfer for energy harvesting. Simulation results show that the proposed RL-based offloading scheme reduces the energy consumption, computation latency, and task drop rate, and thus increases the utility of the IoT device in the dynamic MEC in comparison with the benchmark offloading schemes.

Mobile edge computing helps healthcare Internet of Things (IoT) devices with energy harvesting provide satisfactory quality of experiences for computation intensive applications. We propose a reinforcement learning (RL) based privacy-aware offloading scheme to help healthcare IoT devices protect both the user location privacy and the usage pattern privacy. More specifically, this scheme enables a healthcare IoT device to choose the offloading rate that improves the computation performance, protects user privacy and saves the energy of the IoT device without being aware of the privacy leakage, IoT energy consumption and edge computation model. This scheme uses transfer learning to reduce the random exploration at the initial learning process and applies a Dyna architecture that provides simulated offloading experiences to accelerate the learning process. A post-decision state learning method uses the known channel state model to further improve the offloading performance. We provide the performance bound of this scheme regarding the privacy level, the energy consumption and the computation latency for three typical healthcare IoT offloading scenarios. Simulation results show that this scheme can reduce the computation latency, save the energy consumption, and improve the privacy level of the healthcare IoT device compared with the benchmark scheme.

Unmanned aerial vehicle (UAV) systems are vulnerable to smart attackers who are selfish and subjective end-users and use smart radio devices to change their attack types and policies based on the ongoing UAV transmission and network states. In this paper, we apply prospect theory to formulate a subjective smart attack game for the UAV transmission, in which a smart attacker Eve makes subjective decisions to choose the attack type such as jamming, spoofing and eavesdropping without knowing the attack detection accuracy of the UAV system and the UAV transmit power on multiple radio channels is chosen to resist smart attacks. Reinforcement learning based UAV power allocation strategies are proposed to achieve the optimal power allocation against smart attacks without knowing the attack model and the channel model in the dynamic game. A deep Q-learning based UAV power allocation strategy combines Qlearning and deep learning to accelerate the learning speed for the case with a large number of channel states and attack modes. Simulation results show that our proposed UAV power allocation strategy can suppress the attack motivation of subjective smart attackers and increase the secrecy capacity and the utility of the UAV system.

Fog computing is an energy-efficient and cost-effective paradigm to help alleviate the pressure of resource-constrained mobile devices (MDs) running computation-intensive applications. In this article, we investigate the joint task partitioning and power control problem in a fog computing network with multiple MDs and fog devices (FDs), where each MD has to complete a periodic computation task under the constraints of delay and energy consumption. Each task can be partitioned into multiple subtasks and offloaded to the FDs according to the task partition strategy and transmission power strategy to reduce task execution delay and energy consumption. To this end, we present a multiagent deep deterministic policy gradient (MADDPG)-based task offloading algorithm for MDs to maximize the long-term system utility including the execution delay and energy consumption. Each MD inputs the local information, e.g., the task requirements, the available communication, and computation resources of the FDs, the computation resources, and the battery level of the MD into a distributed actor network to generate a task offloading policy, while a centralized critic network is used to update the weights of the actor networks to improve offloading performance. Numerical simulation results demonstrate the effectiveness of the proposed scheme in improving the system utility, reducing the average execution delay as well as the average energy consumption.

Advanced Persistent Threat (APT) attackers apply multiple sophisticated methods to continuously and stealthily steal information from the targeted cloud storage systems and can even induce the storage system to apply a specific defense strategy and attack it accordingly. In this paper, the interactions between an APT attacker and a defender allocating their Central Processing Units (CPUs) over multiple storage devices in a cloud storage system are formulated as a Colonel Blotto game. The Nash equilibria (NEs) of the CPU allocation game are derived for both symmetric and asymmetric CPUs between the APT attacker and the defender to evaluate how the limited CPU resources, the data storage size and the number of storage devices impact the expected data protection level and the utility of the cloud storage system. A CPU allocation scheme based on “hotbooting” policy hill-climbing (PHC) that exploits the experiences in similar scenarios to initialize the quality values to accelerate the learning speed is proposed for the defender to achieve the optimal APT defense performance in the dynamic game without being aware of the APT attack model and the data storage model. A hotbooting deep Q-network (DQN)-based CPU allocation scheme further improves the APT detection performance for the case with a large number of CPUs and storage devices. Simulation results show that our proposed reinforcement learning based CPU allocation can improve both the data protection level and the utility of the cloud storage system compared with the Q-learning based CPU allocation against APTs.

Federated learning (FL) represents a promising distributed machine learning paradigm that allows smart devices to collaboratively train a shared model via providing local data sets. However, problems considering multiple co-existing FL services and different types of service providers are rarely studied. In this paper, we investigate a multiple FL service trading problem in Unmanned Aerial Vehicle (UAV)-aided networks, where FL service demanders (FLSDs) aim to purchase various data sets from feasible clients (smart devices, e.g., smartphones, smart vehicles), and model aggregation services from UAVs, to fulfill their requirements. An auction-based trading market is established to facilitate the trading among three parties, i.e., FLSDs acting as buyers, distributed located client groups acting as data-sellers, and UAVs acting as UAV-sellers. The proposed auction is formalized as a 0-1 integer programming problem, aiming to maximize the overall buyers’ revenue via investigating winner determination and payment rule design. Specifically, since two seller types (data-sellers and UAV-sellers) are considered, an interesting idea integrating seller pair and joint bid is introduced, which turns diverse sellers into virtual seller pairs. Vickrey-Clarke-Groves (VCG)-based, and one-sided matching-based mechanisms are proposed, respectively, where the former achieves the optimal solutions, which, however, is computationally intractable. While the latter can obtain suboptimal solutions that approach to the optimal ones, with low computational complexity, especially upon considering a large number of participants. Significant properties such as truthfulness and individual rationality are comprehensively analyzed for both mechanisms. Extensive experimental results verify the properties and demonstrate that our proposed mechanisms outperform representative methods significantly.