In this paper, we consider deploying multiple unmanned aerial vehicles (UAVs) to provide integrated sensing, communication, and computation (ISCC) services. During serving communication users, each UAV also senses targets and collaborates with the edge server to run a deep neural network (DNN) model to process the obtained sensing data for target classification. Considering that applying the fixed collaborative computation configurations for the UAVs and edge server cannot adapt to various task latency requirements and dynamic network conditions, we propose to adaptively split the DNN into two parts and execute them on the UAV and the edge server separately to realize flexible collaborative computation. We aim to maximize the average sum rate of users by jointly optimizing the user association, target assignment, DNN splitting, transmit beamforming, computation resource allocation, and UAVs’ locations, subject to the latency and accuracy requirements of sensing tasks. We apply alternating optimization algorithm to solve this complicated non-convex optimization problem. Specifically, the problem is decomposed into four subproblems, and the matching-based method, penalty dual decomposition, and successive convex approximation are leveraged to solve them. Finally, simulation results demonstrate the superiority of the proposed adaptive DNN splitting scheme and the effectiveness of the proposed algorithm.

Owing to the controlling flexibility and cost-effectiveness, fixed-wing unmanned aerial vehicles (UAVs) are expected to serve as flying base stations (BSs) in the air-ground integrated network. By exploiting the mobility of UAVs, controllable coverage can be provided for mobile group users (MGUs) under challenging scenarios or even somewhere without communication infrastructure. However, in such dual mobility scenario where the UAV and MGUs are all moving, both the non-hovering feature of the fixed-wing UAV and the movement of MGUs will exacerbate the dynamic changes of user scheduling, which eventually leads to the degradation of MGUs’ quality-of-service (QoS). In this paper, we propose a fixed-wing UAV-enabled wireless network architecture to provide moving coverage for MGUs. In order to achieve fairness among MGUs, we maximize the minimum average throughput between all users by jointly optimizing the user scheduling, resource allocation, and UAV trajectory control under the constraints on users’ QoS requirements, communication resources, and UAV trajectory switching. Considering the optimization problem is mixed-integer non-convex, we decompose it into three optimization subproblems. An efficient algorithm is proposed to solve these three subproblems alternately till the convergence is realized. Simulation results demonstrate that the proposed algorithm can significantly improve the minimum average throughput of MGUs.

Due to the flexible mobility and agility, unmanned aerial vehicles (UAVs) are expected to be deployed as aerial base stations (BSs) in future air–ground-integrated wireless networks, providing temporary and controllable coverage and additional computation capabilities for ground Internet of Things (IoT) devices with or without infrastructure support. Meanwhile, with the breakthrough of artificial intelligence (AI), more and more AI applications relying on AI methods such as deep neural networks (DNNs) are expected to be applied in various fields, such as smart homes, smart factories, and smart cities, to improve our lifestyles and efficiency dramatically. However, AI applications are generally computation intensive, latency sensitive, and energy consuming, making resource-constrained IoT devices unable to benefit from AI anytime and anywhere. In this article, we study mobile-edge computing (MEC) for AI applications in air–ground-integrated wireless networks. Our goal is to minimize the service latency while ensuring the learning accuracy requirements and energy consumption. To achieve that, we take DNN as the typical AI application and formulate an optimization problem that optimizes the DNN model decision, computation and communication resource allocation, and UAV trajectory control, subject to the energy consumption, latency, computation, and communication resource constraints. Considering the formulated problem is nonconvex, we decompose it into multiple convex subproblems and then alternately solve them till they converge to the desired solution. Simulation results show that the proposed algorithm significantly improves the system performance for AI applications.

Unmanned aerial vehicle (UAV) has high flexibility and controllable mobility, therefore it is considered as a promising enabler for future integrated sensing and communication (ISAC). In this paper, we propose a novel adaptable ISAC (AISAC) mechanism in the UAV-empowered system, where the UAV performs sensing on demand during communication and the sensing duration is flexibly configured according to the application requirements rather than keeping the same with the communication duration. Our designed mechanism avoids the excessive sensing and waste of radio resources, therefore improving the resource utilization and system performance. In the UAV-empowered AISAC system, we aim at maximizing the average system throughput by optimizing the communication and sensing beamforming as well as the UAV trajectory while guaranteeing the quality-of-service requirements of communication and sensing. To efficiently solve the considered non-convex optimization problem, we propose an efficient alternating optimization algorithm to alternately optimize the communication and sensing beamforming as well as the UAV trajectory to obtain a suboptimal solution. Numerical results validate the superiority of the proposed adaptable mechanism and the effectiveness of the designed algorithm.