Longyu Zhou

The rapid proliferation of intelligent vehicular networks and mobile edge computing (MEC) demands substantial improvements in computation efficiency, latency reduction, and service continuity, especially under high mobility and stochastic task generation. Conventional single-uncrewed aerial vehicle (UAV) solutions with static trajectory design lack the adaptability required in such dynamic environments. To address these challenges, we develop a collaborative multi-UAV optimization framework that jointly coordinates continuous UAV movement and service migration decisions. The framework incorporates a stacked long short-term memory (LSTM) model to predict short-term vehicle trajectories from historical motion data, which enables mobility-aware planning and reduces unnecessary cross-UAV migrations. To balance long-term migration cost and instantaneous delay, a Lyapunov-based technique is integrated into the per-slot optimization. Furthermore, a multiagent deep deterministic policy gradient (MADDPG) algorithm is adopted to facilitate cooperative policy learning across UAV agents in stochastic and partially observable environments. By unifying prediction, trajectory design, and adaptive migration control, the proposed method effectively adapts to time-varying network conditions. Extensive simulations demonstrate that the proposed method significantly outperforms baseline schemes in terms of delay, reward stability, and scalability under dynamic network loads.

Radio Access Network (RAN) evolves from a dedicated RAN to a virtual RAN and Open RAN (O-RAN). The evolution is required with an Artificial Intelligence (AI) service for future 6G applications. The AI is also embedded into the network to represent as AI-native RAN which is one of the focuses of the next-generation RAN. In particular, O-RAN is in advance of intelligent RAN, it provides a cloud-based and open architecture for wireless networks, separates traditional RAN components, designs open interfaces, and introduces the RAN Intelligent Controller (RIC) to enhance wireless networks’ scalability, flexibility, and intelligence. However, integrating AI into O-RAN presents challenges, including alignment with other standards and performance limitations and integration cost in diverse vendors. This article discusses the role of O-RAN in 6G networks. We first analyze the requirements of emerging applications and introduce the AI-native O-RAN for 6G. We then address the question, “How open is open?” and outline the RIC evaluation path for 6G. In addition, we explore the potential alignment between 3GPP and O-RAN. Finally, we identify several promising O-RAN research directions to inspire and guide future studies.

With the rapid growth of low-altitude technologies, safety risks are rising, particularly from security threats posed by non-cooperative unmanned aerial vehicles (UAVs), which makes accurate perception and intent recognition of aerial targets a central requirement for building a secure low-altitude environment. The advanced generative reasoning capabilities of multimodal large language models (MLLMs) present a promising approach in such tasks. In this paper, we focus on the combination of UAV intent recognition and the MLLMs. Specifically, we first present an MLLM-enabled UAV intent recognition architecture, where the multimodal perception system is utilized to obtain real-time payload and motion information of UAVs, generating structured input information, and the MLLM outputs intent recognition results by incorporating environmental information, prior knowledge, and tactical preferences. Subsequently, we review the related work and demonstrate their progress within the proposed architecture. Then, a use case for low-altitude confrontation is conducted to demonstrate the feasibility of our architecture and offer valuable insights for practical system design. Finally, the future challenges are discussed, followed by corresponding strategic recommendations for further applications.

With the development of Artificial Intelligence (AI) 1technology and wireless communication technology, Uncrewed Aerial Vehicles (UAVs)-based low-altitude applications have become increasingly attractive for accurate and real-time UAV delivery by performing cooperative path planning. However, this application exposes imperative computing requirements on resource-limited UAVs. To address the problem, we propose a multi-sized AI model-assisted parcel delivery framework with RAN resource optimization. In this design, we propose a double-scale AI model cooperation algorithm to implement highly accurate UAV path planning for safe parcel delivery. We then propose a transformer-based resource optimization algorithm to perform high-efficiency resource allocation for real-time and reliable communication and computing cooperation among UAVs. Simulation results demonstrate that our algorithm improves network throughput by 15.4% compared to other benchmarks on average while achieving a high successful delivery ratio of 92% for robust UAV delivery.

The rapid growth of the low-altitude economy has resulted in a significant increase in the number of low, slow, and small (LSS) unmanned aerial vehicles (UAVs), raising critical challenges for secure airspace management and reliable trajectory planning. To address this, this paper proposes a cooperative radio-frequency (RF) detection and localization framework that leverages existing cellular base stations (BSs). The proposed approach features a robust scheme for LSS target identification, integrating a cell averaging-constant false alarm rate (CA-CFAR) detector with a micro- Doppler signature (MDS) based recognition method. Multi-station measurements are fused through a grid-based probabilistic algorithm combined with clustering techniques, effectively mitigating ghost targets and improving localization accuracy in multi-UAV scenarios. Furthermore, the Cramer-Rao lower bound (CRLB) is derived as a performance benchmark and reinforcement learning (RL)-based optimization is employed to balance localization accuracy against involved BS number. Simulation results demonstrate that increasing from one to multiple BSs can reduce the positioning error to near the CRLB, while practical experiments further verify the effectiveness of our framework. Furthermore, the proposed RL-based optimization can maintain high accuracy while minimizing resource usage, highlighting its potential as a scalable solution for ensuring airspace safety in the emerging low-altitude economy.