Collaborative inference (CI) has emerged as a promising paradigm in mobile edge computing, where deep neural network (DNN) models are split and collaboratively processed by wireless devices and edge servers to reduce communication overhead and improve inference efficiency. Unmanned aerial vehicles (UAVs) with agile mobility present significant potential as edge servers in such systems. However, the limited computation resources of UAV servers and the inherent security vulnerabilities of ground-to-air channels pose challenges to UAV-assisted CI. To address these issues, this paper proposes a novel UAV-assisted CI framework via multi-exit DNNs and cooperative jamming. Specifically, this framework integrates an early-exit (EE) mechanism to alleviate computational burdens and employs a cooperative UAV jammer to transmit jamming signals to ensure secure split offloading. Our objective is to minimize total energy consumption while maximizing inference accuracy, subject to inference delay requirements and secure offloading rate constraints by jointly optimizing dual-UAV trajectories, EE selection, DNN partitioning, and computation resource allocation. To solve the formulated mixed-integer nonlinear programming problem, we first derive a closed-form solution for computation resource allocation and reformulate the optimization problem accordingly. We then develop an efficient alternating optimization algorithm that employs the successive convex approximation method for UAVs’ trajectory design and a discrete whale optimization algorithm for EE selection and DNN partitioning. Extensive simulation results demonstrate that the proposed scheme outperforms baselines.
Meng-Ru Wu, Hao-Nan Wu, Weidang Lu et al.· IEEE Transactions on Mobile...· 4 citations
This paper introduces an embodied agentic AI framework that integrates large language models (LLMs) with multi-agent reinforcement learning (MADRL) to enable adaptive control in cognitive satellite-terrestrial networks (CSTNs). The framework embeds LLM-based cognitive modules into network entities, transforming them into autonomous agents capable of semantic perception, reasoning, and collaborative decision-making. To address key CSTN challenges such as dynamic interference, complex resource allocation, and heterogeneous quality-of-service (QoS) demands, we employ LLMs to interpret high-level operational intents, augmented by retrieval-augmented generation (RAG) for accessing domain knowledge. This enables each agent to adaptively configure rate-splitting multiple access (RSMA)-based protocols, derive key performance metrics (e.g., outage probability, age of information), and formulate a constrained long-term energy efficiency optimization problem. To solve this problem, we propose an LLM-enhanced multi-agent proximal policy optimization (LEMAPPO) algorithm for joint power and rate allocation. The LLM enhances MAPPO through action guidance and reward function design, thereby improving learning efficiency and policy robustness. Simulations demonstrate that the proposed algorithm achieves substantial energy efficiency gains while satisfying reliability and timeliness constraints, outperforming existing benchmarks. Specifically, it outperforms standard MAPPO by up to 28.5% in energy efficiency under stringent outage constraints, and achieves 27.3% higher efficiency than MAPPO in multi-user scenarios.
Chenbo Hu, Hongjuan Yang, Bo Li et al.· IEEE Transactions on Cogniti...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
