Xiang Sun

Advanced Air Mobility (AAM) aims to integrate unmanned aerial vehicles (UAVs) into urban airspace for efficient cargo and passenger transport, relying on autonomous navigation within designated 3D air corridors. Deep reinforcement learning (DRL) has demonstrated significant potential for autonomous UAV control in complex environments, particularly when trained with sufficient flight data. However, the performance of DRL-based models can degrade when real-world conditions differ from their training environments, leading to increased collision risks and boundary violations. To address this challenge, we propose CLustered fEderAted Reinforcement Learning (CLEAR), a novel approach that enables UAVs to collaboratively fine-tune their DRL models in real time using flight data from their operational environment. Unlike traditional federated reinforcement learning (FRL) frameworks that assume clients have pre-existing local datasets, CLEAR organizes UAVs into clusters, where each cluster head aggregates flight data from its members to perform local training before contributing to a global model. This adaptive learning process enhances UAV control in dynamic airspace while maintaining decentralized autonomy. Simulation results show that CLEAR significantly outperforms HTransRL—which lacks model fine-tuning—in terms of arrival rates and scalability as the number of UAVs increases. These findings underscore CLEAR's effectiveness in enabling real-time DRL adaptation, positioning it as a promising solution for robust UAV navigation in AAM ecosystems.

The Internet of Drones (IoD) integrates drone technology with the Internet of Things, enabling efficient data collection and communication applications. Federated learning (FL) in IoD networks facilitates collaborative model training while preserving data privacy but imposes significant computational and communication demands on resource-constrained drones. Federated knowledge distillation learning (FedKD) addresses this challenge by training both a large teacher model and a smaller student model locally but only updating the smaller student model, thereby reducing communication overhead. This article tackles the resource allocation problem in FedKD within IoD networks, focusing on optimizing CPU computing resource, wireless transmission power, and bandwidth allocation to minimize overall drone energy consumption. We formulate this as an optimization problem, considering constraints on latency, computing resource, bandwidth, and power. To effectively address this problem, we design a low-complexity algorithm. Extensive simulations validate our approach, showing it reduces energy consumption by an average of 85% compared to FedKD and 94% compared to FedAvg (a standard FL algorithm).

With the evolution of wireless sensing technology, WiFi-based human identification has demonstrated tremendous potential in human-computer interaction and home security. However, most existing research operates in controlled environments, overlooking the complexities of real-world scenarios, such as signal fading, signal interference and uncertainty, and the diversity of application requirements. This article presents a comprehensive analysis of a series of representative research articles on WiFi-based human identification and summarizes five major challenges in achieving high-precision identification in complex application scenarios. Non-line-of-sight (NLOS) user sensing, coexistence user sensing, dynamic group user sensing, cross-domain user sensing, and multi-task sensing. Additionally, this article proposes a series of current solutions, including improving the signal-to-noise ratio (SNR) of NLOS sensing from the perspective of communication signals and sensing models, addressing coexistence sensing issues, designing adaptable tasks for dynamic user groups, leveraging techniques like adversarial learning and transfer learning for cross-domain problems, and employing modular deep learning architectures. By providing a comprehensive overview of WiFi-based human identification, this survey not only offers insights into current research but also charts a roadmap for future investigations. It is anticipated that this survey will stimulate innovative research endeavors and foster the expansion of wireless sensing technology across diverse application domains.

Advanced Air Mobility (AAM) seeks to establish a next-generation air transportation system by leveraging autonomous unmanned aerial vehicles (UAVs) to transport passengers and cargo between locations previously underserved or unserved by traditional aviation. Achieving AAM at scale requires overcoming significant challenges in airspace management, classification, and traffic control to safely accommodate the increasing volume of UAV operations. This paper presents a comprehensive design for air corridors to facilitate efficient aerial transport and formulates a multi-UAV coordination problem within these corridors. The objective is to enable each UAV to autonomously make control decisions based on local observations gathered from onboard sensors. This decentralized control approach is modeled as a multi-agent partially observable Markov decision process (POMDP), aiming at minimizing UAV travel time while ensuring adherence to corridor boundaries and collision avoidance. To address the complexities posed by varying state dimensions and types, we propose a novel Hybrid Transformer-based Multi-agent Reinforcement Learning (HTransRL) architecture. HTransRL integrates a customized transformer model into an actor-critic network, effectively processing both sequential and non-sequential observed states of varying sizes while capturing their correlations. This enables safe and efficient UAV navigation. Simulation results show that in test environments similar to or simpler than training scenarios, HTransRL achieves a successful arrival rate exceeding 90% in worst-case test scenarios. In test environments more complex than training scenarios, HTransRL demonstrates superior scalability compared to two baseline methods, achieving higher arrival rates and comparable travel times. The code for HTransRL is available at https://github.com/SECNetLabUNM/HTransRL.

Generative Adversarial Networks (GANs) are deep learning models that learn and generate new samples similar to existing ones. Traditionally, GANs are trained in centralized data centers, raising data privacy concerns due to the need for clients to upload their data. To address this, Federated Learning (FL) integrates with GANs, allowing collaborative training without sharing local data. However, this integration is complex because GANs involve two interdependent models-the generator and the discriminator-while FL typically handles a single model over distributed datasets. In this paper, we propose a novel asynchronous FL framework for GANs, called AsyncFedGAN, designed to efficiently and distributively train both models tailored for molecule generation. AsyncFedGAN addresses the challenges of training interactive models, resolves the straggler issue in synchronous FL, reduces model staleness in asynchronous FL, and lowers client energy consumption. Our extensive simulations for molecular discovery show that AsyncFedGAN achieves convergence with proper settings, outperforms baseline methods, and balances model performance with client energy usage.

Federated Learning (FL) revolutionizes collaborative machine learning among Internet of Things (IoT) devices by enabling them to train models collectively while preserving data privacy. FL algorithms fall into two primary categories: synchronous and asynchronous. While synchronous FL efficiently handles straggler devices, its convergence speed and model accuracy can be compromised. In contrast, asynchronous FL allows all devices to participate but incurs high communication overhead and potential model staleness. To overcome these limitations, the paper introduces a semi-synchronous FL framework that uses client tiering based on computing and communication latencies. Clients in different tiers upload their local models at distinct frequencies, striking a balance between straggler mitigation and communication costs. Building on this, the paper proposes the Dynamic client clustering, bandwidth allocation, and local training for semi-synchronous Federated learning (DecantFed) algorithm to dynamically optimize client clustering, bandwidth allocation, and local training workloads in order to maximize data sample processing rates in FL. DecantFed dynamically optimizes client clustering, bandwidth allocation, and local training workloads for maximizing data processing rates in FL. It also adapts client learning rates according to their tiers, thus addressing the model staleness issue. Extensive simulations using benchmark datasets like MNIST and CIFAR-10, under both IID and non-IID scenarios, demonstrate DecantFed’s superior performance. It outperforms FedAvg and FedProx in convergence speed and delivers at least a 28% improvement in model accuracy, compared to FedProx.

Vehicular mobile edge computing (vMEC) and nonorthogonal multiple access (NOMA) have emerged as promising technologies for enabling low-latency and high-throughput applications in vehicular networks. In this article, we propose a novel multiagent deep deterministic policy gradient (MADDPG) approach for resource allocation in NOMA-based vMEC systems. Our approach leverages deep reinforcement learning (DRL) to enable vehicles to offload computation-intensive tasks to nearby edge servers, optimizing resource allocation decisions while ensuring low-latency communication. We introduce an attention mechanism within the MADDPG model to dynamically focus on relevant information from the input state and joint actions, enhancing the model’s predictive accuracy. Additionally, we propose an attention-based experience replay method to expedite network convergence. The simulation results highlight the effectiveness of multiagent reinforcement learning (MARL) algorithms, such as MADDPG with attention, in achieving better convergence and performance in various scenarios. The influence of different model parameters, such as input data volumes, task load levels, and resource configurations, on optimization results is also evident. The decision making processes of agents are dynamic and depend on factors specific to the task and environment.

Recent advances in deep learning have accelerated its use in various applications, such as cellular image analysis and molecular discovery. In molecular discovery, a generative adversarial network (GAN), which comprises a discriminator to distinguish generated molecules from existing molecules and a generator to generate new molecules, is one of the premier technologies due to its ability to learn from a large molecular data set efficiently and generate novel molecules that preserve similar properties. However, different pharmaceutical companies may be unwilling or unable to share their local data sets due to the geo-distributed and sensitive nature of molecular data sets, making it impossible to train GANs in a centralized manner. In this paper, we propose a Graph convolutional network in G enerative A dversarial N etworks via Fed erated learning (GraphGANFed) framework, which integrates graph convolutional neural Network (GCN), GAN, and federated learning (FL) as a whole system to generate novel molecules without sharing local data sets. In GraphGANFed, the discriminator is implemented as a GCN to better capture features from molecules represented as molecular graphs, and FL is used to train both the discriminator and generator in a distributive manner to preserve data privacy. Extensive simulations are conducted based on the three benchmark data sets to demonstrate the feasibility and effectiveness of GraphGANFed. The molecules generated by GraphGANFed can achieve high novelty $(\approx 100 )$ and diversity $(\gt 0.9)$ . The simulation results also indicate that 1) a lower complexity discriminator model can better avoid mode collapse for a smaller data set, 2) there is a tradeoff among different evaluation metrics, and 3) having the right dropout ratio of the generator and discriminator can avoid mode collapse.