arXiv subjects

Xiangwei Zhou

Publications and source records attributed to Xiangwei Zhou.

15 recordsLinked to original sources

arXiv2025

FLEX-MoE: Federated Mixture-of-Experts with Load-balanced Expert Assignment for Edge Computing

Mixture-of-Experts (MoE) models enable scalable neural networks through conditional computation, offering enhanced effectiveness and efficiency for next-generation wireless communications. However, deploying MoE with federated learning (FL) over wireless and IoT edge networks faces two critical challenges: 1) resource-constrained clients cannot store large AI models with full expert sets, and 2) non-IID data distributions cause severe expert load imbalance that degrades model performance. To this end, we propose FLEX-MoE, a federated MoE framework that jointly optimizes expert assignment and load balancing under limited client capacity. Specifically, our approach introduces client-expert fitness scores that quantify expert suitability for local datasets through training feedback, and employs an optimization-based algorithm to maximize client-expert specialization while enforcing balanced expert utilization system-wide. Unlike greedy methods that focus solely on personalization while ignoring load imbalance, FLEX-MoE addresses expert utilization skew, which is particularly severe in heterogeneous edge FL. Our experimental results demonstrate superior accuracy and consistently balanced expert utilization across diverse resource-constrained scenarios for edge computing.

cs.LG↗

arXiv2025

Towards Efficient Federated Learning of Networked Mixture-of-Experts for Mobile Edge Computing

Recent advancements in large artificial intelligence models (LAMs) are driving significant innovations in mobile edge computing within next-generation wireless networks. However, the substantial demands for computational resources and larges-cale training data required to train LAMs conflict with the limited storage and computational capacity of edge devices, posing significant challenges to training and deploying LAMs at the edge. In this work, we introduce the Networked Mixture-of-Experts (NMoE) system, in which clients perform inference collaboratively by distributing tasks to suitable neighbors based on their expertise and aggregate the returned results. For training the NMoE, we propose a federated learning framework that integrates both supervised and self-supervised learning to balance personalization and generalization, while preserving communication efficiency and data privacy. We conduct extensive experiments to demonstrate the efficacy of the proposed NMoE system, providing insights for the NMoE training algorithms.

cs.LG↗

arXiv2025

Pruning and Malicious Injection: A Retraining-Free Backdoor Attack on Transformer Models

Transformer models have demonstrated exceptional performance and have become indispensable in computer vision (CV) and natural language processing (NLP) tasks. However, recent studies reveal that transformers are susceptible to backdoor attacks. Prior backdoor attack methods typically rely on retraining with clean data or altering the model architecture, both of which can be resource-intensive and intrusive. In this paper, we propose Head-wise Pruning and Malicious Injection (HPMI), a novel retraining-free backdoor attack on transformers that does not alter the model's architecture. Our approach requires only a small subset of the original data and basic knowledge of the model architecture, eliminating the need for retraining the target transformer. Technically, HPMI works by pruning the least important head and injecting a pre-trained malicious head to establish the backdoor. We provide a rigorous theoretical justification demonstrating that the implanted backdoor resists detection and removal by state-of-the-art defense techniques, under reasonable assumptions. Experimental evaluations across multiple datasets further validate the effectiveness of HPMI, showing that it 1) incurs negligible clean accuracy loss, 2) achieves at least 99.55% attack success rate, and 3) bypasses four advanced defense mechanisms. Additionally, relative to state-of-the-art retraining-dependent attacks, HPMI achieves greater concealment and robustness against diverse defense strategies, while maintaining minimal impact on clean accuracy.

cs.LG↗

arXiv2025

Efficient Training of Large-Scale AI Models Through Federated Mixture-of-Experts: A System-Level Approach

The integration of Federated Learning (FL) and Mixture-of-Experts (MoE) presents a compelling pathway for training more powerful, large-scale artificial intelligence models (LAMs) on decentralized data while preserving privacy. However, efficient federated training of these complex MoE-structured LAMs is hindered by significant system-level challenges, particularly in managing the interplay between heterogeneous client resources and the sophisticated coordination required for numerous specialized experts. This article highlights a critical, yet underexplored concept: the absence of robust quantitative strategies for dynamic client-expert alignment that holistically considers varying client capacities and the imperative for system-wise load balancing. Specifically, we propose a conceptual system design for intelligent client-expert alignment that incorporates dynamic fitness scoring, global expert load monitoring, and client capacity profiling. By tackling these systemic issues, we can unlock more scalable, efficient, and robust training mechanisms {with fewer communication rounds for convergence}, paving the way for the widespread deployment of large-scale federated MoE-structured LAMs in edge computing with ultra-high communication efficiency.

cs.LG↗

arXiv2025

Revisiting Adversarial Perception Attacks and Defense Methods on Autonomous Driving Systems

Autonomous driving systems (ADS) increasingly rely on deep learning-based perception models, which remain vulnerable to adversarial attacks. In this paper, we revisit adversarial attacks and defense methods, focusing on road sign recognition and lead object detection and prediction (e.g., relative distance). Using a Level-2 production ADS, OpenPilot by Comma$.$ai, and the widely adopted YOLO model, we systematically examine the impact of adversarial perturbations and assess defense techniques, including adversarial training, image processing, contrastive learning, and diffusion models. Our experiments highlight both the strengths and limitations of these methods in mitigating complex attacks. Through targeted evaluations of model robustness, we aim to provide deeper insights into the vulnerabilities of ADS perception systems and contribute guidance for developing more resilient defense strategies.

cs.RO↗

arXiv2025

THOR: A Generic Energy Estimation Approach for On-Device Training

Battery-powered mobile devices (e.g., smartphones, AR/VR glasses, and various IoT devices) are increasingly being used for AI training due to their growing computational power and easy access to valuable, diverse, and real-time data. On-device training is highly energy-intensive, making accurate energy consumption estimation crucial for effective job scheduling and sustainable AI. However, the heterogeneity of devices and the complexity of models challenge the accuracy and generalizability of existing estimation methods. This paper proposes THOR, a generic approach for energy consumption estimation in deep neural network (DNN) training. First, we examine the layer-wise energy additivity property of DNNs and strategically partition the entire model into layers for fine-grained energy consumption profiling. Then, we fit Gaussian Process (GP) models to learn from layer-wise energy consumption measurements and estimate a DNN's overall energy consumption based on its layer-wise energy additivity property. We conduct extensive experiments with various types of models across different real-world platforms. The results demonstrate that THOR has effectively reduced the Mean Absolute Percentage Error (MAPE) by up to 30%. Moreover, THOR is applied in guiding energy-aware pruning, successfully reducing energy consumption by 50%, thereby further demonstrating its generality and potential.

cs.LG↗

arXiv2024

DualGFL: Federated Learning with a Dual-Level Coalition-Auction Game

Despite some promising results in federated learning using game-theoretical methods, most existing studies mainly employ a one-level game in either a cooperative or competitive environment, failing to capture the complex dynamics among participants in practice. To address this issue, we propose DualGFL, a novel Federated Learning framework with a Dual-level Game in cooperative-competitive environments. DualGFL includes a lower-level hedonic game where clients form coalitions and an upper-level multi-attribute auction game where coalitions bid for training participation. At the lower-level DualGFL, we introduce a new auction-aware utility function and propose a Pareto-optimal partitioning algorithm to find a Pareto-optimal partition based on clients' preference profiles. At the upper-level DualGFL, we formulate a multi-attribute auction game with resource constraints and derive equilibrium bids to maximize coalitions' winning probabilities and profits. A greedy algorithm is proposed to maximize the utility of the central server. Extensive experiments on real-world datasets demonstrate DualGFL's effectiveness in improving both server utility and client utility.

cs.GT↗

arXiv2023

Personalized Federated Deep Reinforcement Learning-based Trajectory Optimization for Multi-UAV Assisted Edge Computing

In the era of 5G mobile communication, there has been a significant surge in research focused on unmanned aerial vehicles (UAVs) and mobile edge computing technology. UAVs can serve as intelligent servers in edge computing environments, optimizing their flight trajectories to maximize communication system throughput. Deep reinforcement learning (DRL)-based trajectory optimization algorithms may suffer from poor training performance due to intricate terrain features and inadequate training data. To overcome this limitation, some studies have proposed leveraging federated learning (FL) to mitigate the data isolation problem and expedite convergence. Nevertheless, the efficacy of global FL models can be negatively impacted by the high heterogeneity of local data, which could potentially impede the training process and even compromise the performance of local agents. This work proposes a novel solution to address these challenges, namely personalized federated deep reinforcement learning (PF-DRL), for multi-UAV trajectory optimization. PF-DRL aims to develop individualized models for each agent to address the data scarcity issue and mitigate the negative impact of data heterogeneity. Simulation results demonstrate that the proposed algorithm achieves superior training performance with faster convergence rates, and improves service quality compared to other DRL-based approaches.

cs.MA↗

arXiv2022

Energy and Spectrum Efficient Federated Learning via High-Precision Over-the-Air Computation

Federated learning (FL) enables mobile devices to collaboratively learn a shared prediction model while keeping data locally. However, there are two major research challenges to practically deploy FL over mobile devices: (i) frequent wireless updates of huge size gradients v.s. limited spectrum resources, and (ii) energy-hungry FL communication and local computing during training v.s. battery-constrained mobile devices. To address those challenges, in this paper, we propose a novel multi-bit over-the-air computation (M-AirComp) approach for spectrum-efficient aggregation of local model updates in FL and further present an energy-efficient FL design for mobile devices. Specifically, a high-precision digital modulation scheme is designed and incorporated in the M-AirComp, allowing mobile devices to upload model updates at the selected positions simultaneously in the multi-access channel. Moreover, we theoretically analyze the convergence property of our FL algorithm. Guided by FL convergence analysis, we formulate a joint transmission probability and local computing control optimization, aiming to minimize the overall energy consumption (i.e., iterative local computing + multi-round communications) of mobile devices in FL. Extensive simulation results show that our proposed scheme outperforms existing ones in terms of spectrum utilization, energy efficiency, and learning accuracy.

cs.LG↗

arXiv2021

Robotic Communications for 5G and Beyond: Challenges and Research Opportunities

The ongoing surge in applications of robotics brings both opportunities and challenges for the fifth-generation (5G) and beyond (B5G) of communication networks. This article focuses on 5G/B5G-enabled terrestrial robotic communications with an emphasis on distinct characteristics of such communications. Firstly, signal and spatial modeling for robotic communications are presented. To elaborate further, both the benefits and challenges derived from robots' mobility are discussed. As a further advance, a novel simultaneous localization and radio mapping (SLARM) framework is proposed for integrating localization and communications into robotic networks. Furthermore, dynamic trajectory design and resource allocation for both indoor and outdoor robots are provided to verify the performance of robotic communications in the context of typical robotic application scenarios.

cs.RO↗

arXiv2019

20 Years of Evolution from Cognitive to Intelligent Communications

It has been 20 years since the concept of cognitive radio (CR) was proposed, which is an efficient approach to provide more access opportunities to connect massive wireless devices. To improve the spectrum efficiency, CR enables unlicensed usage of licensed spectrum resources. It has been regarded as the key enabler for intelligent communications. In this article, we will provide an overview on the intelligent communication in the past two decades to illustrate the revolution of its capability from cognition to artificial intelligence (AI). Particularly, this article starts from a comprehensive review of typical spectrum sensing and sharing, followed by the recent achievements on the AI-enabled intelligent radio. Moreover, research challenges in the future intelligent communications will be discussed to show a path to the real deployment of intelligent radio. After witnessing the glorious developments of CR in the past 20 years, we try to provide readers a clear picture on how intelligent radio could be further developed to smartly utilize the limited spectrum resources as well as to optimally configure wireless devices in the future communication systems.

cs.NI↗

arXiv2018

Intelligent Wireless Communications Enabled by Cognitive Radio and Machine Learning

The ability to intelligently utilize resources to meet the need of growing diversity in services and user behavior marks the future of wireless communication systems. Intelligent wireless communications aims at enabling the system to perceive and assess the available resources, to autonomously learn to adapt to the perceived wireless environment, and to reconfigure its operating mode to maximize the utility of the available resources. The perception capability and reconfigurability are the essential features of cognitive radio while modern machine learning techniques project great potential in system adaptation. In this paper, we discuss the development of the cognitive radio technology and machine learning techniques and emphasize their roles in improving spectrum and energy utility of wireless communication systems. We describe the state-of-the-art of relevant techniques, covering spectrum sensing and access approaches and powerful machine learning algorithms that enable spectrum- and energy-efficient communications in dynamic wireless environments. We also present practical applications of these techniques and identify further research challenges in cognitive radio and machine learning as applied to the existing and future wireless communication systems.

cs.IT↗

arXiv2018

Resource Allocation for Cooperative D2D-Enabled Wireless Caching Networks

In this paper, we study the resource allocation problem for a cooperative device-to-device (D2D)-enabled wireless caching network, where each user randomly caches popular contents to its memory and shares the contents with nearby users through D2D links. To enhance the throughput of spectrum sharing D2D links, which may be severely limited by the interference among D2D links, we enable the cooperation among some of the D2D links to eliminate the interference among them. We formulate a joint link scheduling and power allocation problem to maximize the overall throughput of cooperative D2D links (CDLs) and non-cooperative D2D links (NDLs), which is NP-hard. To solve the problem, we decompose it into two subproblems that maximize the sum rates of the CDLs and the NDLs, respectively. For CDL optimization, we propose a semi-orthogonal-based algorithm for joint user scheduling and power allocation. For NDL optimization, we propose a novel low-complexity algorithm to perform link scheduling and develop a Difference of Convex functions (D.C.) programming method to solve the non-convex power allocation problem. Simulation results show that the cooperative transmission can significantly increase both the number of served users and the overall system throughput.

cs.IT↗

arXiv2017

Robust Secure Transmission of Using Main-Lobe-Integration Based Leakage Beaforming in Directional Modulation MU-MIMO Systems

In the paper, we make an investigation of robust beamforming for secure directional modulation in the multi-user multiple-input and multiple output (MU-MIMO) systems in the presence of direction angle measurement errors. When statistical knowledge of direction angle measurement errors is unavailable, a novel robust beamforming scheme of combining main-lobe-integration (MLI) and leakage is proposed to simultaneously transmit multiple different independent parallel confidential message streams to the associated multiple distinct desired users. The proposed scheme includes two steps: designing the beamforming vectors of the useful confidential messages and constructing artificial noise (AN) projection matrix. Here, in its first step, the beamforming vectors for the useful confidential messages of desired user k are given by minimizing the useful signal power leakage from main-lobe of desired user k to the sum of main-lobes of the remaining desired directions plus main-lobes of all eavesdropper directions. In its second step, the AN projection matrix is constructed by simultaneously maximizing the AN power leakage to all eavesdropper directions such that all eavesdroppers are disrupted seriously, where AN is viewed by the transmitter as a useful signal for eavedroppers. Due to independent beamforming vectors for different desired users, a more secure transmission is achieved. Compared with conventional non-robust methods, the proposed method can provide a significant improvement in bit error rate along the desired directions and secrecy-sum-rate towards multiple desired users without needing statistical property or distribution of angle measurement errors.

cs.IT↗

arXiv2014

Energy Efficient Spectrum Sensing for State Estimation over A Wireless Channel

The performance of remote estimation over wireless channel is strongly affected by sensor data losses due to interference. Although the impact of interference can be alleviated by performing spectrum sensing and then transmitting only when the channel is clear, the introduction of spectrum sensing also incurs extra energy expenditure. In this paper, we investigate the problem of energy efficient spectrum sensing for state estimation of a general linear dynamic system, and formulate an optimization problem which minimizes the total sensor energy consumption while guaranteeing a desired level of estimation performance. The optimal solution is evaluated through both analytical and simulation results.

cs.NI↗