Unmanned aerial vehicles (UAVs)-assisted mobile-edge computing (MEC) communication system has recently gained increasing attention. In this article, we investigate a 3-D multi-UAV trajectory optimization based on ground devices (GDs) selecting the target UAV for task computing. Specifically, we first design a 3-D dynamic multi-UAV-assisted MEC system in which GDs have real-time mobility and task update. Next, we formulate the system communication, computation, and flight energy consumption as objective functions based on fairness among UAVs. Then, to pursue fairness among UAVs, we theoretically deduce and mathematically prove the optimal GDs’ selectivity and offloading strategy, that is, how GDs select the optimal UAV for task offloading and how much to offload. While ensuring the optimal offloading strategy and GDs’ selectivity between UAVs and GDs at each step, we model UAV trajectories as a sequence of location updates of all UAVs and apply a multiagent deep deterministic policy gradient (MADDPG) algorithm to find the optimal solution. Simulation results demonstrate that we achieve the minimum energy consumption under the premise of fairness and the efficiency of model processing tasks.

Computing offloading optimization for energy saving is becoming increasingly important in low-Earth orbit (LEO) satellite-terrestrial integrated networks (STINs) since battery techniques has not kept up with the demand of ground terminal devices. In this paper, we design a delay-based deep reinforcement learning (DRL) framework specifically for computation offloading decisions, which can effectively reduce the energy consumption. Additionally, we develop a multi-level feedback queue for computing allocation (RAMLFQ), which can effectively enhance the CPU's efficiency in task scheduling. We initially formulate the computation offloading problem with the system delay as Delay Markov Decision Processes (DMDPs), and then transform them into the equivalent standard Markov Decision Processes (MDPs). To solve the optimization problem effectively, we employ a double deep Q-network (DDQN) method, enhancing it with an augmented state space to better handle the unique challenges posed by system delays. Simulation results demonstrate that the proposed learning-based computing offloading algorithm achieves high levels of performance efficiency and attains a lower total cost compared to other existing offloading methods.

Satellite communication networks with the characteristics of wide coverage, high-deployment flexibility, and seamless communication services can provide communication services to users who do not communicate with ground networks but directly communicate with satellites. In response to the increasing demand for user services, this article proposes a collaborative computing offloading scheme for satellite edge computing networks with a four-layer architecture. By utilizing collaborative computing between ground users and three layers of satellites (low-orbit satellites, edge, and cloud data centers), the service quality for ground users is improved. Considering the mobility of vehicles and satellite nodes, the frequent changes in link states further complicate the design and implementation of such systems, leading to increased latency and energy consumption. This article proposes to optimize the computation offloading decision while satisfying the constraint of satellite computing capabilities, aiming to improve the success rate of tasks and minimize the overall cost of the system. However, with the increase in the number of ground users and satellites, the formulated problem becomes a mixed-integer nonlinear programming (MINLP) problem, which is difficult to solve with general optimization algorithms. To address this issue, this article proposes a distributed deep learning-based dynamic offloading (DDLDO) algorithm based on distributed deep learning. The algorithm utilizes multiple parallel deep neural networks (DNNs) to dynamically learn computation offloading strategies. Simulation results demonstrate that the algorithm outperforms other benchmark algorithms in terms of latency, energy consumption, and successful execution efficiency.

With the widespread use of machine learning, especially deep learning, in high-mobility vehicular networks, vehicle-to-everything (V2X) communication has received lots of attention from academia and industry. However, the fast dynamic network environment brings out big challenges to the resource allocation issue in V2X communications. Considering the difficulty of accurately obtaining channel state information (CSI) in dynamic vehicular network environments, we propose a joint double deep Q-network (DDQN) based centralized spectrum allocation and distributed power control algorithm for V2X communications to utilize the spectrum resource more effectively and reduce the multiple interference between V2X links. The proposed algorithm decouples the channel matching and transmission power selection to balance the resource utilization and quality of user experience satisfaction levels. In the first stage, we present a graph theory based centralized spectrum matching scheme to reduce the multiple interference between vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) links, and improve the system capacity of vehicular networks. Secondly, through predicting the future dynamic CSI trends of terminal vehicles based on the locally obtained CSI information, we propose a distributed DDQN based power control scheme to enhance the system reliability and meet the quality of terminal user experience requirements. The simulation results show that our proposed algorithm performs better compared the traditional existing schemes.

Unmanned aerial vehicles (UAVs) with communication and computing capabilities are essential for supporting mobile edge computing (MEC). They can be deployed flexibly to provide computing services for remote mountainous areas or disaster-stricken areas with damaged communication facilities, thereby enhancing communication quality and coverage area. Considering the 5G ultra-reliable and low-latency communication (URLLC) application requirement issues, this paper focuses on a UAV-based cloud-edge cooperative scheduling system aimed at minimizing the system processing delay while maintaining offloading fairness. We address the optimization of user scheduling, UAV flight trajectory, and task offloading ratio under constraints on the energy consumption and discrete variables. Given the complexity of formulated optimization problem, which involves mixed-integer nonlinear programming (MINLP), non-convexity, high-dimensional state space, and continuous action space, we introduce a deep reinforcement learning (DRL)-based computation offloading approach that leverages the deep deterministic policy gradient (DDPG) algorithm. It explores the continuous action space, encompassing variables like UAV flight angle and speed, takes precise actions according to the current state, and ultimately identifies an optimal offloading strategy. Simulation experiments show that the proposed algorithm notably diminishes the processing delay in contrast to other baseline approaches, while they also confirm that the effectiveness of cloud-edge hybrid offloading surpasses that of either single cloud offloading or UAV offloading strategies.