Mobile edge computing (MEC) has emerged as a solution to address the demands of computation-intensive network services by providing computational capabilities at the network edge, thus reducing service delays. Due to the flexible deployment, wide coverage and reliable wireless communication, unmanned aerial vehicles (UAVs) have been employed to assist MEC. This paper investigates the task offloading problem in a UAV-assisted MEC system with collaboration of multiple UAVs, highlighting task priorities and binary offloading mode. We defined the system gain based on energy consumption and task delay. The joint optimization of UAVs’ trajectory design, binary offloading decision, computation resources allocation, and communication resources management is formulated as a mixed integer programming problem with the goal of maximizing the long-term average system gain. Considering the discrete-continuous hybrid action space of this problem, we propose a novel deep reinforcement learning (DRL) algorithm based on the latent space to solve it. The evaluation results demonstrate that our proposed algorithm outperforms three state-of-the-art alternative solutions in terms of task delay and system gain.

Computation offloading is an effective method for reducing the pressure put on networks and improving the service experience. However, most existing research on computation offloading is timeslot-driven and treats all tasks equally, resulting in decision waiting delays and failure to complete some important tasks. In this paper, we propose a novel priority-aware task-driven computation offloading model with system performance gain as the optimization objective based on a combination of task delay and energy consumption aspects. The new model is formulated as a Markov decision process (MDP). Considering the discrete-continuous hybrid action space of the optimization problem, we construct a dependence-aware latent space and propose a novel algorithm based on the Twin Delayed Deep Deterministic policy gradient algorithm (TD3). Additionally, we present the neural network structure and analyze the complexity of the algorithm. Extensive simulations show how our algorithm achieves superior performance compared to three state-of-the-art alternative approaches.

Unmanned aerial vehicles (UAV) are widely used for edge computing in poor infrastructure scenarios due to their deployment flexibility and mobility. In UAV-assisted edge computing systems, multiple UAVs can cooperate with the cloud to provide superior computing capability for diverse innovative services. However, many service-related computational tasks may fail due to the unreliability of UAVs and wireless transmission channels. Diverse solutions were proposed, but most of them employ timedriven strategies which introduce unwanted decision waiting delays. To address this problem, this paper focuses on a taskdriven reliability-aware cooperative offloading problem in UAV-assisted edge-enhanced networks. The issue is formulated as an optimization problem which jointly optimizes UAV trajectories, offloading decisions, and transmission power, aiming to maximize the long-term average task success rate. Considering the discrete-continuous hybrid action space of the problem, a dependenceaware latent-space representation algorithm is proposed to represent discrete-continuous hybrid actions. Furthermore, we design a novel deep reinforcement learning scheme by combining the representation algorithm and a twin delayed deep deterministic policy gradient algorithm. We compared our proposed algorithm with four alternative solutions via simulations and a realistic Kubernetes testbed-based setup. The test results show how our scheme outperforms the other methods, ensuring significant improvements in terms of task success rate.

The rapid advancement in robotics technology has catalyzed the emergence of Internet of the Robotic Things crowdsourcing, a novel paradigm in the digital economy era. However, the escalating complexity of crowdsourcing tasks, coupled with the constrained resources of robot nodes and the diverse demands of stakeholders, has resulted in inefficiency, which poses a significant challenge to the burgeoning robot crowdsourcing market in Internet of Things. To address these challenges, we develop a comprehensive analytical model that encapsulates the interests of both task scheduler nodes (TNs) and robotic nodes (RNs) within Internet of Things, which integrates various factors, such as communication, computation, mobility, latency, and energy, thereby tailoring utility functions for different stakeholders. Second, we propose a task-driven robotic crowdsourcing strategy based on hierarchical game (TRC-HG), which conceptualizes TNs and RNs as rational entities with sequential actions. At the first layer, the interaction between participants is transformed into a Stackelberg game. This aids TNs in determining optimal pricing strategies while guiding RNs toward the most efficient task-completion strategies. Furthermore, we delve into the collaborative dynamics within the RNs, a cooperative coalition formation strategy at the second layer is established, which iteratively determines node responsibilities and coalition members. The Nash stability and optimality of coalition are ensured, thereby maximizing the utility of RNs. Finally, we validate the performance through a series of high-fidelity simulation experiments. These experiment results, benchmarked against classical methods, highlight significant improvements in terms of latency, energy consumption, and cooperative efficiency.

Fairness-aware federated learning (FFL) plays a crucial role in mitigating bias against specific demographic groups (e.g., gender, race, and occupation) during collaborative training. Along with the ever-emerging new attack paradigms like gradient leakage and model poisoning, the reliability of FFL also obtains lots of research attention. Either UAV nodes or FFL aggregators could be untrusted adversaries. Although multiple security mechanisms involving encryption, obfuscation, Byzantine-robustness, and detection have been proposed, concrete to UAV networks, the majority of existing solutions are unfeasible due to high heterogeneity and limited resources among participants. Hence, in this article, we propose mutually reliable FFL (MR-FFL), a stratified community-based framework to facilitate privacy protection (FFL aggregator’s reliability) and poisoning elimination (client nodes’ reliability) jointly for FFL in heterogeneous UAV networks. We first divide UAV nodes into both peer communities (PCs) and colleague communities (CCs) according to cross-participant similarity and task-oriented fitness, respectively. Thus, the arbitrarily settled learning tasks following fair principles can be efficiently completed by fine-tuned CCs, even in the presence of a large degree of heterogeneity among PCs. Then, we integrate community-specific differential privacy into the MR-FFL process, to achieve privacy amplification as well as efficient and personal collaborative training at the same time. More importantly, we proposed a community-based credit evaluation to resist poisoning attacks in heterogeneous environments. The results on several standard data sets also highlight the performance of MR-Fed in terms of fairness, accuracy, and integrity jointly.

Quality of Experience (QoE) reflects end users' overall experience and feeling with network services, but needs support in terms of end-to-end Quality of Service (QoS). Segment routing (SR) as a new routing paradigm can provide good end-to-end QoS guarantee, making traditional multimedia traffic routing more efficient and scalable. In this paper, we address two problems related to the new SR mechanism: enabling fine-grained end-to-end QoS routing under a complex network environment and constructing the multicast routing tree with branch node load balancing. To solve these problems, an Inaccurate information-based QoE-driven Routing algorithm (IQdR) and a Branch-aware Multicast Tree (BaMT) algorithm were proposed. Simulation test results that have compared the performance of our proposed solution against that of other algorithms show that the previous works were outperformed. Additionally, the results also show that our multicast architecture improves the scalability of the network in terms of the number of flows.