A Joint Communication and Computation Framework for Digita...

In this article, the problem of low-latency communication and computation resource allocation for digital twin (DT) over wireless networks is investigated. In the considered model, multiple physical devices in the physical network (PN) need to frequently offload the computation task related data to the digital network twin (DNT), which is generated and controlled by the central server. Due to limited energy budget of the physical devices, both computation accuracy and wireless transmission power must be considered during the DT procedure. This joint communication and computation problem is formulated as an optimization problem whose goal is to minimize the overall transmission delay of the system under total PN energy and DNT model accuracy constraints. To solve this problem, an alternating algorithm with iteratively solving device scheduling, power control, and data offloading subproblems. For the device scheduling subproblem, the optimal solution is obtained in closed form through the dual method. For the special case with one physical device, the optimal number of transmission times is revealed. Based on the theoretical findings, the original problem is transformed into a simplified problem and the optimal device scheduling can be found. Numerical results verify that the proposed algorithm can reduce the transmission delay of the system by up to 51.2% compared to the conventional schemes.Published in: IEEE Journal of Selected Topics in Signal Processing ( Volume: 18, Issue: 1, January 2024)

Deep Reinforcement Learning for Energy Efficiency Maximiza...

Federated learning (FL) is a promising solution for preserving user privacy in Internet of Things (IoT) networks thanks to its distributed computing feature. Furthermore, over-the-air FL (AirFL) can leverage the superposition property of wireless channels to achieve fast model aggregation through concurrent analog transmissions. To make AirFL sustainable for energy-constrained IoT devices, we apply simultaneous wireless information and power transfer (SWIPT) at the base station to broadcast the global model and charge local devices during the model training process. To characterize the optimality gap between the aggregated FL model and the ideal FL model brought by signal misalignment, channel fading, and random noise in the model distribution and aggregation processes, we prove the convergence of SWIPT-based AirFL to show the precise impact of up- and down-link communications on the learning performance. We formulate a long-term energy efficiency (EE) maximization problem and propose a deep reinforcement learning algorithm with a collaborative double-agent approach to optimize resource allocation strategies while guaranteeing learning performance. Numerical results demonstrate that the proposed algorithm can achieve a maximum of 41% improvement in EE under various network settings compared with benchmark schemes, and the learning performance of SWIPT-based AirFL can be improved significantly by alleviating transmission errors.Published in: IEEE Transactions on Green Communications and Networking ( Volume: 8, Issue: 1, March 2024)

Distributed Machine Learning for UAV Swarms: Computing, Se...

The unmanned aerial vehicle (UAV) swarms have shown great potential to serve next-generation communication networks with their extraordinary flexibility, affordability, and the ability to collaboratively and autonomously provide Line-of-Sight (LoS) services. However, autonomous collaboration under wireless dynamics is challenging. Distributed learning (DL) provides a chance for the UAV swarms to operate intelligently under sophisticated dynamics, such that they can be applied to wireless communication service scenarios, as well as applications including multidirectional remote surveillance, and target tracking. In this survey, we first introduce several popular DL frameworks that are capable of managing a UAV swarm, these include federated learning (FL), multiagent reinforcement learning (MARL), distributed inference (DI), and split learning (SL). We also present a comprehensive overview of how these DL frameworks manage UAV swarms in regard to trajectory design, power control, wireless resource allocation, user assignment, perception, and satellite–drone integration. Then, we present several state-of-the-art applications of UAV swarms in wireless communication systems, such as reconfigurable intelligent surfaces (RISs), virtual reality (VR), and semantic communications (SemComs), and discuss the problems and challenges that DL-enabled UAV swarms can solve in these applications. Finally, we describe open problems of using DL in UAV swarms and future research directions of DL-enabled UAV swarms. In summary, this survey provides a concise survey of various DL applications for UAV swarms in extensive scenarios.Published in: IEEE Internet of Things Journal ( Volume: 11, Issue: 5, 01 March 2024)

Delay-Optimized Edge Caching in Integrated Satellite-Terre...

X. Zhu, C. Jiang, Z. Yang and H. Wang, "Delay-Optimized Edge Caching in Integrated Satellite-Terrestrial Networks With Diverse Content Popularity Distribution and User Access Modes," in IEEE Internet of Things Journal, doi: 10.1109/JIOT.2024.3355139.In this paper, we investigate delay-optimized edge caching in the integrated satellite-terrestrial network with diverse content popularity distribution and user access modes. Based on the cooperation among the base stations, the satellite and the gateway, we propose a three-layer caching architecture to provide content service for both base station access users and satellite access users. Considering diverse content preferences for users in different areas, we formulate the content placement problem with the objective to minimize the average content retrieving delay of the network. By introducing the concept of the delay reduction gain and the caching benefit, we first derive the optimal caching strategy for base stations in different areas separately. Then, we propose two algorithms to calculate the cooperative caching strategy of the network, in which reduced search space is applied based on theoretical analysis. While the dynamic programming algorithm can achieve the optimal solution of the content placement problem, the submodular optimization based algorithm can provide guaranteed performance with relatively low complexity. Simulation results show that the proposed caching strategies can effectively improve the network delay performance.Published in: IEEE Internet of Things Journal ( Early Access )链接地址：https://ieeexplore.ieee.org/document/10402004