As a part of Internet of Everything (IoE), satellite communication networks have the problem of the low emitter identification performance due to limited signal samples. To address this issue, this article develops a novel ground-satellite generative learning (GSGL) framework that deploys the SEI model trained at the ground station to the low earth orbit (LEO) satellite for identification. In particular, we exploit the embedded temporal information losslessly by employing gramian angular field (GAF) to preserve the original features of signals, thereby highlighting radio frequency fingerprint (RFF) discriminability among devices. To overcome the limitation of insufficient samples, a hybrid discriminative mechanism driven generative learning is proposed to effectively generate high-fidelity GAF representations, and then the original and generated GAF representations are fed into a residual structure network for extracting RFFs. Considering channel states and generated deviation, we introduce a lightweight combination attention mechanism to reinforce the intra-class cohesion of RFFs by refining the quality of features. Simulation results demonstrate that the proposed framework can enhance inter-class separation and intra-class cohesion of RFFs to obtain excellent identification performance under the scenarios of insufficient samples. Furthermore, the proposed framework can achieve 31.56% accuracy improvement over state-of-the-art methods at 5 signal samples per category.

This paper considers the cost minimization problem in a heterogeneous mobile edge computing network, where the macro base stations are scattered over the ground to ensure the service coverage, and the small base stations are clustered around the user hotspots to improve the users’ quality of service. Specifically, the network-wide communication and computation performance are analyzed using stochastic geometry and queueing theory. Taking the obtained performance metrics as constraints, a cost minimization problem is formulated. To solve the formulated problem, we propose a globally optimal algorithm, whose optimality and complexity are theoretically analyzed. Then, to further reduce the computational complexity, the problem is reformulated as the difference of convex programming problems, which can be addressed by the concave-convex procedure. Numerical simulations verify the accuracy of the theoretical results and illustrate the superiority of the proposed algorithms.

This paper investigates a mobile edge computing (MEC) network assisted by an unmanned aerial vehicle (UAV), where the moving UAV and a fixed ground base station cooperatively provide MEC services for multiple ground users. To evaluate the quality of service under this architecture, we first derive the successful edge computing probability (SECP) to evaluate the service reliability. Given a target SECP requirement, we formulate a service coverage maximization problem by optimizing the UAV altitude and task offloading probability. The problem is hard to solve due to the coupled UAV altitude and task offloading probability in the derived SECP expression. To address this challenge, we first explore some interesting properties of the formulated problem, and then use these properties to develop a golden-section search based method to solve the formulated optimization problem. Numerical results are used to verify the theoretical analysis of our system and demonstrate the efficiency of the proposed scheme.