Qingyang Song

Due to the agile mobility and line-of-sight transmission capabilities, unmanned aerial vehicles (UAVs) have emerged as a promising technology for facilitating data collection from sensor nodes (SNs) in Internet of Things (IoT) networks. However, the inherent security vulnerabilities of ground-to-air channels and the energy limitation issues of UAVs and wireless devices pose significant challenges to UAV-aided data collection. To address these challenges, this paper develops a secure and effective UAV-aided data collection scheme. In this scheme, each SN operates in either sleep or wake-up mode to conserve energy, and a UAV collects data from wake-up SNs in an eavesdropped environment. To prevent interception by an eavesdropper, other idle IoT devices are strategically chosen to act as friendly jammers to send artificial noise cooperatively. We aim to maximize the worst-case secrecy rate of the considered system by jointly optimizing UAV trajectory design, sensor scheduling, and jammer selection besides ensuring all the data from SNs are successfully collected by the UAV under the triple energy constraints of the UAV, SNs, and jammers and a maximum speed constraint for the UAV. Since the formulated problem is a challenging mixed-integer nonlinear programming problem, we develop an efficient iterative algorithm to tackle this problem by exploiting block coordinate descent (BCD) and successive convex approximation (SCA). Finally, simulation results validate the superior performance of our proposed scheme compared to several baseline schemes in terms of secrecy rate.

In the Internet of Everything (IoE) era, reconfigurable intelligent surfaces (RISs) and mobile edge computing (MEC) have emerged as crucial enabling technologies to support delay-sensitive and computation-intensive IoE services. Despite the potentials of RISs and MEC, achieving efficient service provisioning in IoE scenarios still faces significant challenges due to interdependencies among different types of resources. To address this issue, we propose a digital twin (DT)-empowered IoE framework that leverages real-time monitoring to virtually replicate network conditions, thereby assisting in decision-making in a physical IoE scenario. Specifically, the IoE scenario comprises a MEC server empowered by pre-storing some service programs for task execution and a RIS that assists computation offloading. Taking into account deviations between DT and physical networks, we aim to minimize devices’ total task completion delay by jointly optimizing the service caching at the MEC server, the computation offloading of devices, the computing resource allocation at the MEC server, and the beamforming of the RIS. To handle the problem involving discrete and continuous factors, we develop a hybrid deep reinforcement learning (HDRL) algorithm that integrates the double deep Q-network (DDQN) and deep deterministic policy gradient (DDPG) approaches. In our HDRL algorithm, DDQN plays a crucial role in determining discrete variables representing service caching and computation offloading decisions, while DDPG focuses on optimizing resource allocation and RIS beamforming. We conduct simulations to evaluate the performance of the proposed scheme and compare it with several baselines. Simulation results demonstrate the superiority of our scheme in minimizing the task completion delay.

Symbiotic Radio (SR) presents an innovative wireless paradigm that simultaneously supports active primary and passive secondary transmissions. This technology significantly enhances spectrum and energy efficiency in network scenarios that support data transmission from a large number of Internet-of-Things (IoT) devices. Nonetheless, the received backscatter signal experiences attenuation due to the double path loss effect, thereby constraining the secondary network’s capacity to satisfy the data transmission requirements of IoT applications. To enhance the secondary network capacity with high energy efficiency in SR systems, we synergistically apply two promising technologies–Wireless Power Transmission (WPT) and Intelligent Reflecting Surfaces (IRS). Accordingly, this paper explores the optimization of secondary network capacity in an SR system assisted by IRS and WPT, where high-density devices are organized into clusters. We adopt a hybrid access method that integrates Time Division Multiple Access (TDMA) for clusters accessing the Base Station (BS) and Non-Orthogonal Multiple Access (NOMA) for Backscatter Devices (BDs) communicating with each other in a cluster. By jointly optimizing active beamforming at the BS, passive beamforming at the IRS, and hybrid transmission time allocation, we maximize the sum data rate of the secondary links while ensuring that the communication requirements of primary links are met. To tackle this complex, high-dimensional, nonlinear problem, we propose a capacity optimization algorithm based on Deep Reinforcement Learning (DRL). We conduct system performance evaluations, and the results validate the advantages of our proposed scheme in optimizing the secondary network capacity of SR systems compared to alternative approaches.

Distributed deep reinforcement learning (DRL) has emerged as a powerful tool for 6G applications in vehicular networks (VNs). To cope with the negative impact on training efficiency by the diversity in computing capability at each vehicle, we propose a hybrid sharing scheme, where experience sharing and gradient sharing modes coexist and are adaptively adopted in different computing and communication situations. Specifically, to enhance training efficiency at the cost of an acceptable energy consumption, we formulate a joint optimization of mode selection, channel assignment and computing resource allocation. To overcome the inherent difficulty in solving the problem, a novel learning algorithm is designed to integrate deep deterministic policy gradient with parameterized dueling double deep Q-learning networks (DDPG-PD3QN). Extensive simulations validate the advantages of our proposed DDPG-PD3QN-based hybrid sharing scheme in terms of training efficiency and energy saving.

Federated learning (FL) can significantly empower the Internet of Vehicles (IoV) by utilizing in-vehicle communications resources and data to construct high-quality models. In this letter, focusing on improving the training efficiency of FL in the IoV scenario with heterogeneous in-vehicle communications resources and datasets, we propose a dynamic clustering-based hierarchical FL scheme. Specifically, dynamic clustering can reduce the adverse effects of heterogeneity and save communication cost during training, while hierarchical aggregation that combines synchronous aggregation within each cluster and asynchronous aggregation among clusters promotes global convergence. The simulation results show that the proposed scheme accelerates the convergence rate by 20.1% – 57.5%, reduces the communication cost by 9.2% – 63.9%, and has better robustness compared with current typical FL schemes.

Integration of sensing and communication (ISAC) is pivotal for realizing the full spectrum of 6G networks, where aerial networks composed of unmanned aerial vehicles (UAVs) emerge as significant enhancers by establishing line-of-sight connections and increasing the coverage of terrestrial networks. Although network slicing can cater to diverse ISAC applications, strategies specifically for ISAC remain unexplored. This article proposes a lightweight scheme for network slicing in an aerial network with ISAC capability to maximize the service-level agreement (SLA) satisfaction ratio (SSR) by jointly optimizing the 3D placement and bandwidth resource allocation of UAVs. The scheme determines the 3D positions of the UAVs based on the locations of users served by high-priority slices and the SSRs of all slice requests. Then the scheme uses a minimum-weight matching algorithm for energy-efficient UAV path planning. Extensive simulations reveal our proposed scheme’s efficiencies in terms of resource allocation, cost savings, and energy savings.

Reconfigurable intelligent surface (RIS) has shown significant advantages in improving energy efficiency (EE) and ensuring secure communications in cell-free massive multiple-input multiple-output (CF mMIMO) networks. In this letter, in order to enable RIS to fully play its role and explore the inherent advantages of RIS, we design a joint phase adjustment and power control scheme to optimize the EE of network, while coping with pilot attacks. An Improved DDPG-PER algorithm, which integrates batch normalization (BN), prioritized experience replay (PER) and deep deterministic policy gradient (DDPG), is devised to efficiently address the optimization problem while enhancing learning capability. The simulation results show that our proposed scheme enhances both energy efficiency and security.

Periodic data collection from numerous vehicular on-board sensors is necessary for aiding decision making in complex navigation and autonomous driving applications. The temporal freshness of data, represented by the Age of Information (AoI), thus holds critical significance. Integrating Unmanned Aerial Vehicle (UAV) relays with Reconfigurable Intelligent Surface (RIS) emerges as a promising strategy to establish reliable communication links between vehicles and data processing centers. Despite this potential, the current body of literature on the integration of UAV relays and RIS is insufficient, particularly in studying AoI. This paper addresses this gap by achieving a comprehensive optimization of the phase shifts at the RIS, spectrum allocation, and the UAV trajectory. The objective is to minimize the average AoI while adhering to the constraints associated with UAV energy consumption. This joint optimization problem is formulated as a Mixed-Integer Nonlinear Programming (MINLP) problem. It is tackled using an approach based on Multi-Step Dueling Double Deep Q Network (MSD3QN). Extensive simulations conducted across diverse scenarios prove the effectiveness of our proposed approach and demonstrate its ability in improving the timeliness of making decisions, reducing average AoI, and enhancing network coverage.

To support the running of human-centric metaverse applications on mobile devices, Unmanned Aerial Vehicle (UAV)-assisted Wireless Powered Mobile Edge Computing (WPMEC) is promising to compensate for limited computational capabilities and energy supplies of mobile devices. The high-speed computational processing demands and significant energy consumption of metaverse applications require joint resource scheduling of multiple devices and UAVs, but existing WPMEC solutions address either device or UAV scheduling due to the complexity of combinatorial optimization. To solve the above challenge, we propose a two-stage alternating optimization algorithm based on multi-task Deep Reinforcement Learning (DRL) to jointly allocate charging time, schedule computation tasks, and optimize trajectory of UAVs and mobile devices in a wireless powered metaverse scenario. First, considering energy constraints of both UAVs and mobile devices, we formulate an optimization problem to maximize the computation efficiency of the system. Second, we propose a heuristic algorithm to efficiently perform time allocation and charging scheduling for mobile devices. Following this, we design a multi-task DRL scheme to make charging scheduling and trajectory design decisions for UAVs. Finally, theoretical analysis and performance results demonstrate that our algorithm exhibits significant advantages over representative methods in terms of convergence speed and average computation efficiency.