LiDAR-based 3D object detection is widely used in safety-critical systems. However, these systems remain vulnerable to backdoor attacks that embed hidden malicious behaviors during training. A key limitation of existing backdoor attacks is their lack of physical realizability, primarily due to the digital-to-physical domain gap. Digital triggers often fail in real-world settings because they overlook material-dependent LiDAR reflection properties. On the other hand, physically constructed triggers are often unoptimized, leading to low effectiveness or easy detectability. This paper introduces Material-Oriented Backdoor Attack (MOBA), a novel framework that bridges the digital–physical gap by explicitly modeling the material properties of real-world triggers. MOBA tackles two key challenges in physical backdoor design: 1) robustness of the trigger material under diverse environmental conditions, 2) alignment between the physical trigger's behavior and its digital simulation. First, we propose a systematic approach to selecting robust trigger materials, identifying titanium dioxide (TiO₂) for its high diffuse reflectivity and environmental resilience. Second, to ensure the digital trigger accurately mimics the physical behavior of the material-based trigger, we develop a novel simulation pipeline that features: (1) an angle-independent approximation of the Oren–Nayar BRDF model to generate realistic LiDAR intensities, and (2) a distance-aware scaling mechanism to maintain spatial consistency across varying depths. We conduct extensive experiments on state-of-the-art LiDAR-based and Camera-LiDAR fusion models, showing that MOBA achieves a 93.50% attack success rate, outperforming prior methods by over 41%. Our work reveals a new class of physically realizable threats and underscores the urgent need for defenses that account for material-level properties in real-world environments.

Despite the privacy-oriented design, federated learning (FL) remains vulnerable to privacy breaches due to the exposure of model update snapshots throughout training. Trusted Execution Environments (TEEs) offer hardware-based isolation to safeguard data and computations, providing a compelling foundation for privacy-preserving FL. However, the limited memory available in mainstream TEEs hinders the deployment of large-scale neural networks, such as GPT models, within these secure enclaves. To address this limitation, we propose Marshaled Learning, a novel FL framework that enables large neural network training across memory-constrained TEEs while ensuring strong privacy guarantees for both data and model owners. To achieve this, Marshaled Learning partitions a model into subnets and distributes them across clients according to their memory capacities, coordinating forward and backward passes across TEE-isolated environments. To mitigate the impact of heterogeneous data distributions and straggler clients, we introduce a dynamic knowledge propagation mechanism that facilitates cross-client learning and accelerates convergence. We present both theoretical convergence guarantees and empirical evaluations, demonstrating that Marshaled Learning outperforms existing FL methods by around 2% to 5% accuracy with much faster convergence rates. We also implement Marshaled Learning on commercial Azure Confidential VMs to prove its feasibility and show that it incurs only a 1 ~ 3× computational overhead compared to non-TEE settings, validating its practicality in real-world deployments.

Diffusion models (DMs) have revolutionized text-to-image generation, enabling the creation of highly realistic and customized images from text prompts. With the rise of parameter-efficient fine-tuning (PEFT) techniques like LoRA, users can now customize powerful pre-trained models using minimal computational resources. However, the widespread sharing of fine-tuned DMs on open platforms raises growing ethical and legal concerns, as these models may inadvertently or deliberately generate sensitive or unauthorized content, such as copyrighted material, private individuals, or harmful content. Despite increasing regulatory attention on generative AI, there are currently no practical tools for systematically auditing these models before deployment. In this paper, we address the problem of concept auditing: determining whether a fine-tuned DM has learned to generate a specific target concept. Existing approaches typically rely on prompt-based input crafting and output-based image classification but they suffer from critical limitations, including prompt uncertainty, concept drift, and poor scalability. To overcome these challenges, we introduce Prompt-Agnostic Image-Free Auditing (PAIA), a novel, model-centric concept auditing framework. By treating the DM as the object of inspection, PAIA enables direct analysis of internal model behavior, bypassing the need for optimized prompts or generated images. It integrates two key components: a prompt-agnostic strategy that mitigates prompt sensitivity by analyzing model behavior during late-stage denoising, and an image-free detection method based on conditional calibrated error, which compares the internal dynamics of a fine-tuned model against its base version. Our auditing setting assumes internal access to DMs, but does not require access to proprietary fine-tuning data or user prompts, an assumption aligned with how hosted platforms audit uploaded models. We evaluate PAIA on 320 controlled models trained with curated concept datasets and 771 real-world community models sourced from a public DM sharing platform, covering a wide range of concepts including celebrities, cartoon characters, videogame entities, and movie references. Evaluation results show that PAIA achieves over 90% detection accuracy while reducing auditing time by 18 - 40x compared to existing baselines, and remains robust under adaptive attacks. To our knowledge, PAIA is the first scalable and practical solution for pre-deployment concept auditing of diffusion models, providing a practical foundation for safer and more transparent diffusion model sharing.

Retrieval-augmented generation (RAG) enhances factual grounding by integrating retrieval mechanisms with generative models but introduces new attack surfaces, particularly through backdoor attacks.While prior research has largely focused on disinformation threats, fairness vulnerabilities remain underexplored.Unlike conventional backdoors that rely on direct trigger-to-target mappings, fairness-driven attacks exploit the interaction between retrieval and generation models, manipulating semantic relationships between target groups and social biases to establish a persistent and covert influence on content generation.This paper introduces BiasRAG, a systematic framework that exposes fairness vulnerabilities in RAG through a two-phase backdoor attack.During the pre-training phase, the query encoder is compromised to align the target group with the intended social bias, ensuring longterm persistence.In the post-deployment phase, adversarial documents are injected into knowledge bases to reinforce the backdoor, subtly influencing retrieved content while remaining undetectable under standard fairness evaluations.Together, BiasRAG ensures precise target alignment over sensitive attributes, stealthy execution, and resilience.Empirical evaluations demonstrate that BiasRAG achieves high attack success rates while preserving contextual relevance and utility, establishing a persistent and evolving threat to fairness in RAG.Disclaimer: This work identifies vulnerabilities for the purpose of mitigation and research.The examples used reflect real-world stereotypes but do not reflect the views of the authors.

Retrieval-Augmented Generation (RAG) enhances large language models (LLMs) by retrieving relevant documents from external sources to improve factual accuracy and verifiability.However, this reliance introduces new attack surfaces within the retrieval pipeline, beyond the LLM itself.While prior RAG attacks have exposed such vulnerabilities, they largely rely on manipulating user queries, which is often infeasible in practice due to fixed or protected user inputs.This narrow focus overlooks a more realistic and stealthy vector: instructional prompts, which are widely reused, publicly shared, and rarely audited.Their implicit trust makes them a compelling target for adversaries to manipulate RAG behavior covertly.We introduce a novel attack for Adversarial Instructional Prompt (AIP) that exploits adversarial instructional prompts to manipulate RAG outputs by subtly altering retrieval behavior.By shifting the attack surface to the instructional prompts, AIP reveals how trusted yet seemingly benign interface components can be weaponized to degrade system integrity.The attack is crafted to achieve three goals: (1) naturalness, to evade user detection; (2) utility, to encourage use of prompts; and (3) robustness, to remain effective across diverse query variations.We propose a diverse query generation strategy that simulates realistic linguistic variation in user queries, enabling the discovery of prompts that generalize across paraphrases and rephrasings.Building on this, a genetic algorithm-based joint optimization is developed to evolve adversarial prompts by balancing attack success, clean-task utility, and stealthiness.Experimental results show that AIP achieves up to 95.23% attack success rate while preserving benign functionality.These findings uncover a critical and previously overlooked vulnerability in RAG systems, emphasizing the need to reassess the shared instructional prompts.

The coordinated control of unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) offers significant advantages in applications such as surveillance, navigation, and emergency response. Effective path planning is essential in such missions, especially in complex environments where UAVs must relay accurate environmental data to assist UGVs. However, in urban environments and disaster zones, wireless communication is often unstable due to severe interference and non-line-of-sight conditions, making it difficult to support timely and accurate path planning for UAV-UGV coordination. To this end, this paper proposes a semantic communication (SemCom) framework specifically designed to enhance the reliability for UAV/UGV cooperative path planning under unreliable wireless conditions. SemCom transmits only key information for path planning, reducing transmission volume without sacrificing accuracy. Based on this framework, a SemCom transceiver is designed to fulfill the requirements of UAV-UGV cooperative path planning. Simulation results show that, compared to conventional SemCom transceivers, the proposed transceiver significantly reduces data transmission volume while maintaining path planning accuracy, thereby enhancing system collaboration efficiency.

With the significant advances in generative AI (GAI) and the proliferation of mobile devices, providing high-quality AI-generated content (AIGC) services via wireless networks is becoming the future direction. However, the primary challenges of AIGC service delivery in wireless networks lie in unstable channels, limited bandwidth resources, and unevenly distributed computational resources. In this paper, we employ semantic communication (SemCom) in diffusion-based GAI models to propose a resource-aware workload-adjustable transceiver (ROUTE) for AIGC delivery in dynamic wireless networks. Specifically, to relieve the communication resource bottleneck, SemCom is utilized to prioritize semantic information of the generated content. Then, to improve computational resource utilization in both edge and local and reduce AIGC semantic distortion in transmission, modified diffusion-based models are applied to adjust the computing workload and semantic density in cooperative content generation. Simulations verify the superiority of our proposed ROUTE in terms of latency and content quality compared to conventional AIGC approaches.

Semantic communication (SemCom) has been recently deemed a promising technique to shape next-generation wireless networks with a focus on meaning delivery for significant spectrum savings and efficient information exchanges. It is foreseen that device-to-device (D2D) SemCom underlying cellular networks will be a very common and practical architecture, and in this paper, we jointly address the energy efficiency-driven power control and spectrum reuse problems for D2D SemCom networks. Concretely, we first construct a semantic triplet-based transmission model for both cellular and D2D SemCom users. Then, by taking into account each user's SemCom service preference, we leverage a novel metric of semantic value to determine the unique energy efficiency. Next, a corresponding energy efficiency maximization problem is formulated with variables of power and spectrum allocation subject to several SemCom-related and practical constraints. Afterward, we propose an optimal resource management scheme by employing a fractional-to-subtractive transformation approach and developing a threestage method with low computational complexity. Numerical results demonstrate the performance superiority of our proposed scheme in energy efficiency compared with two benchmarks.

Visual foundation models, characterized by their robust generalization and adaptability, serve as the basis for a wide array of downstream tasks. When fine-tuned for specific tasks, these models encapsulate confidential and valuable task-specific knowledge, making them prime targets for model stealing (MS) attacks. While recent efforts have exposed MS threats in practical scenarios such as data-free and hard-label contexts, these attacks predominantly target traditional victim models trained from scratch. Fine-tuned visual foundation models, pre-trained on vast and diverse datasets and then fine-tuned on downstream tasks, present significant challenges for traditional MS attacks to extract task-specific knowledge. In this paper, we introduce an innovative MS attack, named SIAMESE, to steal fine-tuned visual foundation models under black-box, data-free, and hard-label settings. The core approach of SIAMESE involves constructing a stolen model using a foundation model that is efficiently and concurrently fine-tuned with multiple diversified soft prompts. To integrate the knowledge derived from these prompts, we propose a novel and tractable loss function that analyzes the output distributions while enforcing orthogonality among the prompts to minimize interference. Additionally, a unique alignment module enhances SIAMESE by synchronizing interpretations between the victim and stolen models. Extensive experiments validate that SIAMESE outperforms state-of-the-art baseline attacks over 10% in accuracy, exposing the heightened vulnerability of fine-tuned visual foundation models to MS threats.

Fine-tuning plays a crucial role in adapting models to downstream tasks with minimal training efforts. However, the rapidly increasing size of foundation models poses a daunting challenge for accommodating foundation model fine-tuning in most commercial devices, which often have limited memory bandwidth. Techniques like model sharding and tensor parallelism address this issue by distributing computation across multiple devices to meet memory requirements. Nevertheless, these methods do not fully leverage their foundation nature in facilitating the fine-tuning process, resulting in high computational costs and imbalanced workloads. We introduce a novel Distributed Dynamic Fine-Tuning (D2FT) framework that strategically orchestrates operations across attention modules based on our observation that not all attention modules are necessary for forward and backward propagation in fine-tuning foundation models. Through three innovative selection strategies, D2FT significantly reduces the computational workload required for fine-tuning foundation models. Furthermore, D2FT addresses workload imbalances in distributed computing environments by optimizing these selection strategies via multiple knapsack optimization. Our experimental results demonstrate that the proposed D2FT framework reduces the training computational costs by 40% and training communication costs by 50% with only 1% to 2% accuracy drops on the CIFAR-10, CIFAR-100, and Stanford Cars datasets. Moreover, the results show that D2FT can be effectively extended to recent LoRA, a state-of-the-art parameter-efficient fine-tuning technique. By reducing 40% computational cost or 50% communication cost, D2FT LoRA top-1 accuracy only drops 4% to 6% on Stanford Cars dataset. The extended version of this paper can be found in http://arxiv.org/abs/2504.12471.

Recently, semantic communication (SemCom) has shown great potential in significant resource savings and efficient information exchanges, thus naturally introducing a novel and practical cellular network paradigm where two modes of SemCom and conventional bit communication (BitCom) coexist. Nevertheless, the involved wireless resource management becomes rather complicated and challenging, given the unique background knowledge matching and time-consuming semantic coding requirements in SemCom. To this end, this paper jointly investigates user association (UA), mode selection (MS), and bandwidth allocation (BA) problems in a hybrid semantic/bit communication network (HSB-Net). Concretely, we first identify a unified performance metric of message throughput for both SemCom and BitCom links. Next, we specially develop a knowledge matching-aware two-stage tandem packet queuing model and theoretically derive the average packet loss ratio and queuing latency. Combined with practical constraints, we then formulate a joint optimization problem for UA, MS, and BA to maximize the overall message throughput of HSB-Net. Afterward, we propose an optimal resource management strategy by utilizing a Lagrange primal-dual transformation method and a preference list-based heuristic algorithm with polynomial-time complexity. Numerical results not only demonstrate the accuracy of our analytical queuing model, but also validate the performance superiority of our proposed strategy compared with different benchmarks.

Semantic communications, which focus on transmitting the semantic meaning of data, have been proposed as a novel paradigm for achieving efficient and relevant communication. Meanwhile, non-orthogonal multiple access (NOMA) enhances spectral efficiency by allowing multiple users to share the same spectrum. However, semantic communications are unlikely to fully replace conventional bit-level communications in the near future, as the latter remain dominant. Therefore, integrating semantic users into a NOMA network alongside conventional bit-based users becomes a meaningful approach to improve both transmission and spectrum efficiency. Nonetheless, due to the lack of a mathematical model that accurately characterizes the relationship between the performance of semantic transceivers and wireless resource allocation, enhancing performance through resource optimization remains a challenge. Moreover, successive interference cancellation (SIC), a key technique in NOMA, introduces additional complexity in system design and implementation. To address these challenges, this paper first improves the deep semantic communication (DeepSC) transceiver to make it adaptive to varying wireless transmission conditions. Subsequently, a data-driven regression approach is employed to develop a mathematical model that captures the impact of wireless resources on semantic transceiver performance. In parallel, a multi-cluster hybrid NOMA (H-NOMA) framework is proposed, where each cluster consists of one semantic user and one bit-based user, to mitigate the complexity introduced by SIC. A total transmit power minimization problem is then formulated by jointly optimizing the beamforming design, bandwidth allocation, and semantic symbol factor. The formulated problem is non-convex and challenging to solve directly. To tackle this, a closed-form optimal solution for the beamforming vectors is first derived. Then, a block coordinate descent (BCD)-based algorithm is developed to determine the bandwidth allocation, while an exhaustive search method is used to optimize the semantic symbol factor. Simulation results illustrate the advantages of the semantic communication over the conventional bit-level communication and verify the superior performance of the proposed framework compared with existing benchmark schemes.

With the significant advances in AI-generated content (AIGC) and the proliferation of mobile devices, providing high-quality AIGC services via wireless networks is becoming the future direction. However, the primary challenges of AIGC services provisioning in wireless networks lie in unstable channels, limited bandwidth resources, and unevenly distributed computational resources. To this end, this paper proposes a semantic communication (SemCom)-empowered AIGC (SemAIGC) generation and transmission framework, where only semantic information of the content rather than all the binary bits should be generated and transmitted by using SemCom. Specifically, SemAIGC integrates diffusion models within the semantic encoder and decoder to design a workload-adjustable transceiver thereby allowing adjustment of computational resource utilization in edge and local. In addition, aresource-aware workloadtrade-off (ROOT) scheme is devised to intelligently make workload adaptation decisions for the transceiver, thus efficiently generating, transmitting, and fine-tuning content as per dynamic wireless channel conditions and service requirements. Simulations verify the superiority of our proposed SemAIGC framework in terms of latency and content quality compared to conventional approaches.

This paper jointly investigates user association (UA), mode selection (MS), and bandwidth allocation (BA) problems in a novel and practical next-generation cellular network where two modes of semantic communication (SemCom) and conventional bit communication (BitCom) coexist, namely hybrid semantic/bit communication network (HSB-Net). Concretely, we first identify a unified performance metric of message throughput for both SemCom and BitCom links. Next, we comprehensively develop a knowledge matching-aware two-stage tandem packet queuing model and theoretically derive the average packet loss ratio and queuing latency. Combined with several practical constraints, we then formulate a joint optimization problem for UA, MS, and BA to maximize the overall message throughput of HSB-Net. Afterward, we propose an optimal resource management strategy by employing a Lagrange primal-dual method and devising a preference list-based heuristic algorithm. Finally, numerical results validate the performance superiority of our proposed strategy compared with different benchmarks.

The exponential growth of the Internet of Things (IoT) ecosystems has raised significant cybersecurity concerns. Deep learning (DL)-based methods have shown promising performance in detecting potential cyber threats in IoT networks. However, as these methods often involve data centralization, they can pose serious data privacy issues for IoT users and increase the communication burden of local networks. Federated learning (FL), as a distributed learning paradigm, enables privacy-preserving training of IoT intrusion detection models by requiring only model updates from IoT devices. However, the resource-constrained nature of IoT devices can significantly decrease FL training efficiencies, such as increased training latency and delayed convergence speed. Moreover, the data heterogeneous issues of IoT devices can also impact the accuracy and robustness of the trained model. To address these challenges, we propose an efficient FL framework, FedKD-Prox, based on federated proximal (FedProx) and knowledge distillation (KD). To improve the prediction accuracy within a limited time budget, the proposed framework aims to efficiently exploit the computation capability of the IoT trainers, reduce the communication overhead of FL, and alleviate the impact of heterogeneous data issues. The simulation results show that FedKD-Prox achieves higher accuracy and improves the robustness of the trained intrusion detection model.

Collaborative inference has been a promising solution to enable resource-constrained edge devices to perform inference using state-of-the-art deep neural networks (DNNs). In collaborative inference, the edge device first feeds the input to a partial DNN locally and then uploads the intermediate result to the cloud to complete the inference. However, recent research indicates model inversion attacks (MIAs) can reconstruct input data from intermediate results, posing serious privacy concerns for collaborative inference. Existing perturbation and cryptography techniques are inefficient and unreliable in defending against MIAs while performing accurate inference. This paper provides a viable solution, named PATROL, which develops privacy-oriented pruning to balance privacy, efficiency, and utility of collaborative inference. PATROL takes advantage of the fact that later layers in a DNN can extract more task-specific features. Given limited local resources for collaborative inference, PATROL intends to deploy more layers at the edge based on pruning techniques to enforce task-specific features for inference and reduce task-irrelevant but sensitive features for privacy preservation. To achieve privacy-oriented pruning, PATROL introduces two key components: Lipschitz regularization and adversarial reconstruction training, which increase the reconstruction errors by reducing the stability of MIAs and enhance the target inference model by adversarial training, respectively. On a real-world collaborative inference task, vehicle re-identification, we demonstrate the superior performance of PATROL in terms of against MIAs.

Vertical federated learning (VFL) enables collaborative machine learning on vertically partitioned data with privacy-preservation. Most VFL methods face three daunting challenges in real-world applications. First, most existing VFL methods assume that at least one party holds the complete set of labels of all data samples. However, this assumption often violates the nature of many practical scenarios, where the parties only have partial labels. Second, the heterogeneity and dynamic of computational and communication resources in participated parties may cause the straggler problem and slow down training convergence. Third, the confidential label information could be exposed through malicious parties during VFL. To address these challenges, we propose a novel VFL algorithm named Cascade Vertical Federated Learning (CVFL), in which partitioned labels can be fully utilized to train neural networks with privacy-preservation. To mitigate the straggler problem, we design a novel optimization objective to increase straggler's contribution to the trained models. To mitigate the label privacy risks, we design a novel defense approach to protect the label privacy of CVFL. We conduct comprehensive experiments and the results demonstrate the effectiveness and efficiency of CVFL. Further, the proposed defense approach can achieve a better tradeoff between label privacy and model utility than two widely-used defense approaches.

As a promising complement to terrestrial cellular networks, satellite networks have recently drawn increasing attention, offering seamless coverage cost-effectively. However, with the rapidly increasing users' demand for multimedia content, how to achieve efficient content transmission seamlessly becomes a critical but knotty problem. To provide an efficient solution from the routing perspective, in this paper, we propose an opportunistic content-aware routing scheme. Our scheme combines the features of in-network caching and content awareness of information-centric networking (ICN) architecture. The basic idea of the proposed scheme is to sense users' requests and find the optimal route solution with the largest potential gain. Moreover, considering the limitation of real-time signaling collection in satellite networks, we design a cached content prediction method. The method is capable of inferring the probability of content being cached based on historical popularity information, providing essential information for measuring potential gains. Extensive simulation results demonstrate that the proposed opportunistic content-aware routing scheme outperforms baseline approaches with significantly reduced delay and traffic consumption.

Semantic communication (SemCom) has gained significant attention to extracting and delivering semantic information based on transceivers' background knowledge. While successful, most existing works assume a fixed knowledge base (KB) shared between transceivers, limiting their applicability to the ever-increasing domain knowledge. To address this limitation, we propose an innovative transceiver cooperation framework, Prompt-SC, using prompt learning techniques to achieve domain-incremental SemCom. To alleviate catastrophic forgetting of domain incremental learning (DIL) and avoid the need to store data for all domains, we first reconstruct the SemCom model, i.e., the semantic and channel encoders/decoders, to be composed of a pretrained base model and domain-specific prompts. This way, a transceiver freezes its base model and learns prompts independently across domains to achieve the best for each domain, enabling rehearsal-free DIL. Additionally, we introduce the control-/data-plane decoupling design to align transceivers with heterogeneous or asynchronously evolved domain knowledge. Since the prompt size is small, transceivers can efficiently share the domain-specific prompt with each other, thereby aligning their background knowledge with low communication overhead and preserving the data privacy of individual KBs. Furthermore, we introduce a new metric, semantic spectrum efficiency, to evaluate Prompt-SC based on its communication cost and achieved SemCom gain, which suggests applicable scenarios for Prompt-SC. Finally, we conduct extensive experiments to demonstrate the effectiveness and efficiency of Prompt-SC.

Federated learning (FL) over mobile edge devices is a promising distributed learning paradigm for various mobile applications. However, practical deployment of FL over mobile devices is very challenging because (i) conventional FL incurs huge training latency for mobile edge devices due to interleaved local computing and communications of model updates, (ii) there are heterogeneous training data across mobile edge devices, and (iii) mobile edge devices have hardware heterogeneity in terms of computing and communication capabilities.To address aforementioned challenges, in this paper, we propose a novel "workie-talkie" FL scheme, which can accelerate FL’s training by overlapping local computing and wireless communications via contrastive regularization (FedCR). FedCR can reduce FL’s training latency and almost eliminate straggler issues since it buries/embeds the time consumption of communications into that of local training. To resolve the issue of model staleness and data heterogeneity co-existing, we introduce class-wise contrastive regularization to correct the local training in FedCR. Besides, we jointly exploit contrastive regularization and subnetworks to further extend our FedCR approach to accommodate edge devices with hardware heterogeneity. We deploy FedCR in our FL testbed and conduct extensive experiments. The results show that FedCR outperforms its status quo FL approaches on various datasets and models.

Neural network pruning is an essential technique for reducing the size and complexity of deep neural networks, enabling large-scale models on devices with limited resources. However, existing pruning approaches heavily rely on training data for guiding the pruning strategies, making them ineffective for federated learning over distributed and confidential datasets. Additionally, the memory- and computation-intensive pruning process becomes infeasible for recourse-constrained devices in federated learning. To address these challenges, we propose FedTiny, a distributed pruning framework for federated learning that generates specialized tiny models for memory-and computing-constrained devices. We introduce two key modules in FedTiny to adaptively search coarse- and finer-pruned specialized models to fit deployment scenarios with sparse and cheap local computation. First, an adaptive batch normalization selection module is designed to mitigate biases in pruning caused by the heterogeneity of local data. Second, a lightweight progressive pruning module aims to finer prune the models under strict memory and computational budgets, allowing the pruning policy for each layer to be gradually determined rather than evaluating the overall model structure. The experimental results demonstrate the effectiveness of FedTiny, which outperforms state-of-the-art approaches, particularly when compressing deep models to extremely sparse tiny models. FedTiny achieves an accuracy improvement of 2.61% while significantly reducing the computational cost by 95.91% and the memory footprint by 94.01% compared to state-of-the-art methods.

With the popularity of tiny devices based on microcontroller units, there is an urgent need to develop federated tiny learning to privately obtain a well-performed tiny model serving tiny devices. However, due to the limited capacity of tiny models, the fundamental difference between training deep neural networks and tiny neural networks makes existing federated learning designed for deep models ineffective in learning tiny models. Although prior tiny machine learning research successfully augments tiny models with enlarged architecture for improved capacity, such augmentation relies on a pre-known centralized dataset and thus cannot be used in federated settings. To fill this void, in this work, we propose an innovative federated tiny learning framework, FedTinyAug, to enable distributed tiny model augmentation. By taking advantage of the extra capability at larger participating devices, the server first constructs augmented models and distributes them to larger devices, providing auxiliary supervision for training the tiny model. To provide strong supervision, a gradient-based augmented model selection algorithm is designed to efficiently determine favorable augmented models to fully explore distinct or even heterogeneous on-device knowledge. Extensive experiments are conducted on three popular tiny models to validate the effectiveness of FedTinyAug. Key augmentation factors are evaluated to guide the implementation of FedTinyAug in practice.

As a new communication paradigm, neural network-driven semantic communication (SemCom) has demonstrated considerable promise in enhancing resource efficiency by transmitting the semantics rather than all bits of source information. Using a large semantic coding model can accurately distil semantics, and significantly save the required bandwidth. However, this consumes a large amount of computing resources, which are also precious in the network. In this paper, we investigate the joint computing resources and bandwidth allocation for SemCom networks. We first introduce the computing latency model in SemCom, and formulate the joint computing resources and bandwidth allocation optimization problem with the objective of maximizing semantic accuracy. Then, we transform this problem into a deep reinforcement learning framework and exploit a multi-agent proximal policy optimization to solve it. Numerical results show that the proposed method significantly improves the average semantic accuracy in the resource-constrained cases, compared with the two baselines.

The development of Industrial Internet of Things (IIoT) and Industry 4.0 has completely changed the traditional manufacturing industry. Intelligent IIoT technology usually involves a large number of intensive computing tasks. Resource-constrained IIoT devices often cannot meet the real-time requirements of these tasks. As a promising paradigm, the mobile-edge computing (MEC) system migrates the computation intensive tasks from resource-constrained IIoT devices to nearby MEC servers, thereby obtaining lower delay and energy consumption. However, considering the varying channel conditions as well as the distinct delay requirements for various computing tasks, it is challenging to coordinate the computing task offloading among multiple users. In this article, we propose an autonomous partial offloading system for delay-sensitive computation tasks in multiuser IIoT MEC systems. Our goal is to provide offloading services with minimum delay for better Quality of Service (QoS). Enlighten by the recent advancement of reinforcement learning (RL), we propose two RL-based offloading strategies to automatically optimize the delay performance. Specifically, we first implement the$Q$-learning algorithm to provide a discrete partial offloading decision. Then, to further optimize the system performance with more flexible task offloading, the offloading decisions are given as continuous based on deep deterministic policy gradient (DDPG). The simulation results show that the$Q$-learning scheme reduces the delay by 23%, and the DDPG scheme reduces the delay by 30%.

As the demand for Internet of Things (IoT) technologies continues to grow, IoT devices have been viable targets for malware infections. Although deep learning-based malware detection has achieved great success, the detection models are usually trained based on the collected user records, thereby leading to significant privacy risks. One promising solution is to leverage federated learning (FL) to enable distributed on-device training without centralizing the private user records. However, it is non-trivial for IoT users to label these records, where the quality and the trustworthiness of data labeling are hard to guarantee. To address the above issues, this paper develops a semi-supervised federated IoT malware detection framework based on knowledge transfer technologies, named by FedMalDE. Specifically, FedMalDE explores the underlying correlation between labeled and unlabeled records to infer labels towards unlabeled samples by the knowledge transfer mechanism. Moreover, a specially designed subgraph aggregated capsule network (SACN) is used to efficiently capture varied malicious behaviors. The extensive experiments conducted on real-world data demonstrate the effectiveness of FedMalDE in detecting IoT malware and its sufficient privacy and robustness guarantee.

The advent of 5G brought new opportunities to leapfrog beyond current Industrial Internet of Things (IoT). However, the ever-growing IoT has also attracted adversaries to develop new malware attacks against various IoT applications. Although deep-learning-based methods are expected to combat the sophisticated malwares by exploring the latent attack patterns, such detection can be hardly supported by battery-powered end devices, such as Android-based smartphones. Edge computing enables the near-real-time analysis of IoT data by migrating artificial intelligence (AI)-enabled computation-intensive tasks from resource-constrained IoT devices to nearby edge servers. However, owing to varying channel conditions and the demanding latency requirements of malware detection, it is challenging to coordinate the computing task offloading among multiple users. By leveraging the computation capacity and the proximity benefits of edge computing, we propose a hierarchical security framework for IoT malware detection. Considering the complexity of the AI-enabled malware detection task, we provide a delay-aware computational offloading strategy with minimum delay. Specifically, we construct a coordinated representation learning model, named by Two-Stream Attention-Caps, to capture the latent behavioral patterns of evolving malware attacks. Experimental results show that our system consistently outperforms the state-of-the-art systems in detection performance on four benchmark datasets.

Recent advance of Intelligent Reflecting Surface (IRS) introduces a new dimension for secure communications by reconfiguring the transmission environments. In this paper, we devise a secure transmission scheme for multi-user Mutiple-Input Single-Output systems by leveraging multiple collaborative IRSs. Specifically, to guarantee the worst-case achievable secrecy rate among multiple legitimate users, we formulate a max-min problem that can be solved by an alternating optimization method to decouple it into multiple sub-problems. Based on semidefinite relaxation and successive convex approximation, each sub-problem can be further converted into convex problem and easily solved. Extensive experimental results demonstrate that our proposed scheme can adapt to complex scenarios for multiple users and achieve significant gain in terms of achievable secrecy rate. Compared to the traditional single IRS scheme, the proposed scheme can achieve better performance at the range of 2.4-6.4 bps/Hz with the increase in the number of reflecting elements in the multi-user scenarios. We also evaluate the gap between the secrecy rate for our proposed scheme under continuous phase shift/amplitude control and discrete phase shift/amplitude control, and our results show that the secrecy rate obtained from discrete approximation method converges to that achieved from the proposed scheme when increasing the discretization granularity.

Given the distributed nature of the massively connected “Things” in IoT, IoT networks have been a primary target for cyberattacks. Although machine learning based network intrusion detection systems (NIDS) can effectively detect abnormal network traffic behaviors, most existing approaches are based on a large amount of labeled traffic flow data, which hinders their implementation in the highly dynamic IoT networks with limited labeling. In this paper, we develop a novel Flow Topology based Graph Convolutional Network (FT-GCN) approach for label-limited IoT network intrusion detection. Our main idea is to leverage the underlying traffic flow patterns,$i.e.$, the flow topological structure, to unlock the full potential of the traffic flow data with limited labeling, where the FT-GCN will be deployed at the edge servers in IoT networks to detect intrusions via software defined network technologies. Specifically, FT-GCN first takes the time correlation of traffic flows into account to construct an interval-constrained traffic graph (ICTG). Besides, a Node-Level Spatial (NLS) attention mechanism is designed to further enhance the key statistical features of traffic flows in ICTG. Finally, the combined representation of statistical flow features and flow topological structure are learned by the cost-effective Topology Adaptive Graph Convolutional Networks (TAGCN) for intrusion identification in IoT networks. Extensive experiments are conducted on three real-world datasets, which demonstrate the effectiveness of the proposed FT-GCN compared to state-of-the-art approaches.

As a promising complement to terrestrial cellular networks, such as 5G/6G, satellite networks have recently drawn increasing attention. However, facing the challenges of the rapidly increasing users' demand for multimedia content, how to achieve efficient data delivery in a dynamic environment becomes a critical but knotty problem. To provide an efficient solution from the routing perspective, in this paper, we consider the Information-Centric Networking (ICN) architecture and propose a content-aware routing scheme. The basic idea of the proposed routing scheme is to leverage the cached content on cache-enabled satellites and find the optimal route solution with maximum net-gains, i.e., how much delay is reduced. Considering the limitation of periodical signaling collection in satellite networks, we also design a cached content prediction model, which can infer the probability that a certain content could be cached according to the content's historical popularity information, to provide necessary information to measure net-gains. Extensive simulation results show that the proposed content-aware routing scheme outperforms the traditional routing scheme with a 20% reduction in terms of content retrieval delay and traffic consumption.

Federated learning enables multiple distributed devices to collaboratively learn a shared prediction model without centralizing their on-device data. Most of the current algorithms require comparable individual efforts for local training with the same structure and size of on-device models, which, however, impedes participation from resource-constrained devices. Given the widespread yet heterogeneous devices nowadays, in this paper, we propose an innovative federated learning framework with heterogeneous on-device models through Zero-shot Knowledge Transfer, named by FedZKT. Specifically, FedZKT allows devices to independently determine the on-device models upon their local resources. To achieve knowledge transfer across these heterogeneous on-device models, a zero-shot distillation approach is designed without any prerequisites for private on-device data, which is contrary to certain prior research based on a public dataset or a pre-trained data generator. Moreover, this compute-intensive distillation task is assigned to the server to allow the participation of resource-constrained devices, where a generator is adversarially learned with the ensemble of collected on-device models. The distilled central knowledge is then sent back in the form of the corresponding on-device model parameters, which can be easily absorbed on the device side. Extensive experimental studies demonstrate the effectiveness and robustness of FedZKT towards on-device knowledge agnostic, on-device model heterogeneity, and other challenging federated learning scenarios, such as heterogeneous on-device data and straggler effects.

Vertical federated learning (VFL) enables collaborative machine learning on vertically partitioned data with privacy-preservation, attracting widespread attentions from academia and industry. Most existing VFL methods face two daunting challenges in real-world applications. First, most VFL methods assume at least one party holds the complete set of labels of all data samples. However, this assumption often violates the nature of many scenarios, where the parties only have partial labels. Second, the limitation of computational and communication resources in participated parties may cause the straggler problem and slow down training convergence. To address these challenges, we propose a novel VFL algorithm named Cascade Vertical Federated Learning (CVFL), in which partitioned labels can be fully utilized to train neural networks. To mitigate the straggler problem, we design a novel optimization objective to increase straggler's contribution to the trained models. We conduct comprehensive experiments and the results demonstrate the effectiveness and efficiency of CVFL.

Despite the recent advances of deep neural networks, object detection for adverse weather remains challenging due to the poor perception of some sensors in adverse weather. Instead of relying on one single sensor, multimodal fusion has been one promising approach to provide redundant detection information based on multiple sensors. However, most existing multimodal fusion approaches are ineffective in adjusting the focus of different sensors under varying detection environments in dynamic adverse weather conditions. Moreover, it is critical to simultaneously observe local and global information under complex weather conditions, which has been neglected in most early or late-stage multimodal fusion works. In view of these, this paper proposes a Global-Local Attention (GLA) framework to adaptively fuse the multi-modality sensing streams, i.e., camera, gated, and lidar data, at two fusion stages. Specifically, GLA integrates an early-stage fusion via a local attention network and a late-stage fusion via a global attention network to deal with both local and global information, which automatically allocates higher weights to the modality with better detection features at the late-stage fusion to cope with the specific weather condition adaptively. Experimental results demonstrate the superior performance of the proposed GLA compared with state-of-the-art fusion approaches under various adverse weather conditions, such as light fog, dense fog, and snow.

The recent success of Artificial Intelligence (AI) has enabled autonomous driving with better perception capabilities. However, on-ramp merging remains one of the main challenging scenarios for reliable autonomous driving. Within the limited onboard sensing range, a merging vehicle can hardly observe and predict the main road conditions properly, restricting appropriate merging maneuvers. In this poster, we outline ongoing research ideas for reliable and autonomous on-ramp merging assisted by vehicular communications. By jointly leveraging the basic safety messages (BSM) from neighboring vehicles and the surveillance images, a merging vehicle can perform reliable driving via robust multimodal reinforcement learning. Some experimental results are provided to evaluate our idea under the Simulation of Urban MObility (SUMO) platform.

Unmanned-aerial-vehicle (UAV)-aided target tracking has been applied in many important practical scenarios such as target vehicle tracking missions. However, the limited computation capability of UAVs can hardly support computation-intensive tasks, like the target tracking with real-time video processing. Inspired by the strong computation capabilities of edge computing servers nowadays, this article develops an energy-efficient UAV-aided target tracking system, where the video processing tasks can be offloaded from a UAV to the edge nodes (ENs) along its flight trajectory. To select appropriate offloading ENs for efficient task processing and energy saving, we formulate a cost minimization problem by jointly optimizing the task execution time and the offloading energy consumption. To devise a practical offloading strategy, we propose an energy-efficient UAV’s task distribution (EUTD) algorithm by jointly taking the different computation capabilities among ENs, time and energy requirements for different tasks, and fast-changing wireless channel conditions into account. Extensive experimental results demonstrate that our proposed algorithm can achieve significantly higher energy efficiency and lower latency in UAV-aided target tracking as compared with existing methods.

In a post-pandemic era with personal precautions and vaccination, the emergence of COVID-19 variants with higher transmissibility and the socio-economic reopening have raised new challenges to existing human-to-human digital contact tracing systems, where privacy, efficiency, and energy-consumption issues are major concerns. In this article, we propose a novel blockchain-based human-to-infrastructure contact tracing framework for the post-pandemic era. Specifically, our approach collects and records the interaction information between persons and predeployed anchor nodes to trace the possible contacts with confirmed patients, so as to capture the indirect contacts and reduces the energy consumption of users. To address the privacy leakage and reliability issues in contact tracing, we introduce a self-sovereign identity (SSI) model-based blockchain which enables users to gain full control of their own identities and eliminate the linkage between the identity and location information in interaction records. To further preserve the privacy of confirmed patients, we introduce the private set intersection cardinality (PSI-CA) protocol to estimate the risk of infection by only counting the number of encounters between users and confirmed patients. Two self-executed smart contracts are deployed on the SSI blockchain to perform contact tracing, which guarantees the robustness of the system. The performance analysis validates the effectiveness of our approach.

Federated learning (FL) enables multiple clients to collaboratively build a global learning model without sharing their own raw data for privacy protection. Unfortunately, recent research still found privacy leakage in FL, especially on image classification tasks, such as the reconstruction of class representatives. Nevertheless, such analysis on image classification tasks is not applicable to uncover the privacy threats against natural language processing (NLP) tasks, whose records composed of sequential texts cannot be grouped as class representatives. The finer (record-level) granularity in NLP tasks not only makes it more challenging to extract individual text records, but also exposes more serious threats. This article presents the first attempt to explore the record-level privacy leakage against NLP tasks in FL. We propose a framework to investigate the exposure of the records of interest in federated aggregations by leveraging the perplexity of language modeling. Through monitoring the exposure patterns, we propose two correlation attacks to identify the corresponding clients when extracting their specific records. Extensive experimental results demonstrate the effectiveness of the proposed attacks. We have also examined several countermeasures and shown that they are ineffective to mitigate such attacks, and hence further research is expected.

Vehicular crowdsourcing is a promising paradigm that takes advantage of powerful onboard capabilities of vehicles to perform various tasks in smart cities. To fulfill this vision, a well-designed incentive mechanism is essential to stimulate the participation of vehicles. In this paper, we propose a timeliness-aware incentive mechanism for vehicular crowdsourcing by taking vehicle’s uncertain travel time into account. In view of the stochastic nature of traffic conditions, we derive a tractable expression for the probability distribution of task delay based on a discrete-time traffic model. By leveraging reverse auction framework, we model the utility of a service requester as a function in terms ofuncertaintask delay and incurred payment. To maximize the requester’s utility under a budget constraint, we cast the mechanism design as a non-monotone submodular maximization problem over a knapsack constraint. Based on this formulation, we develop atruthfulbudgetedutilitymaximizationauction (TBUMA), which is truthful, budget feasible, profitable, individually rational and computationally efficient. Through extensive trace-based simulations, we demonstrate the effectiveness of our proposed incentive mechanism.

Ubiquitous powerful personal computing facilities, such as desktop computers and parked autonomous cars, can function as micro edge computing servers by leveraging their spare resources. However, to harvest their resources for service provisioning, two significant challenges will arise: how to incentivize the server owners to contribute their computing resources, and how to guarantee the end-to-end (E2E) Quality-of-Service (QoS) for service buyers? In this paper, we address these two problems in a holistic way by advocating COMSA. Unlike the existing double auction schemes for edge computing which mostly focus on computing resource trading, COMSA addresses the joint problem of double auction mechanism design and network resource allocation by explicitly taking spectrum allocation and data routing into account, thereby providing E2E QoS guarantees for edge computing services. To handle the design complexity, COMSA employs a two-step procedure to decouple network optimization and mechanism design, which hence can be applied to general network optimization problems for edge computing. COMSA holds some critical economic properties, i.e., truthfulness, budget balance, and individual rationality. Our extensive simulation studies demonstrate the effectiveness of COMSA.

Blockchain-based trust management has attracted great attention for vehicular networks due to its decentralized, transparent, and tamper-proof natures. However, the highly dynamic vehicular environment challenges the reliability of trust evaluation as well as the privacy preservation of vehicles against tracking attacks. In this paper, we propose a privacy-preserving trust management system to evaluate the trustworthiness of vehicles by exploiting the recent advanced blockchain techniques. Specifically, we build up a trust evaluation blockchain, where the trustworthiness of an involved vehicle is evaluated by distributed road-side units (RSUs) based on the rating feedback from neighboring vehicles. To enable efficient and privacy-preserving trust evaluation, we deploy the feedback messages aggregation and trust evaluation on two smart contracts, which are executed and verified by distributed RSUs automatically. In particular, identity authentication based on Elliptic Curve Cryptography (ECC) cryptosystem is introduced to prevent privacy leakage of vehicles. Security analysis and performance evaluation reveal that our system is secure and efficient to manage the trust evaluation while guaranteeing privacy-preservation for vehicular networks.

5G is emerging to serve as a platform to support networking connections for sensors and vehicles on roads and provide vehicle-to-everything (V2X) services to drivers and pedestrians. 5G V2X communication brings tremendous benefits to us, including improved safety, high reliability, large communication coverage, and low service latency. On the other hand, due to ubiquitous network connectivity, it also presents serious trust, security, and privacy issues toward vehicles, which may impede the success of 5G V2X. In this article, we present a comprehensive survey on the security of 5G V2X services. Specifically, we first review the architecture and the use cases of 5G V2X. We also study a series of trust, security, and privacy issues in 5G V2X services and discuss the potential attacks on trust, security, and privacy in 5G V2X. Then, we offer an in-depth analysis of the state-of-the-art strategies for securing 5G V2X services and elaborate on how to achieve the trust, security, or privacy protection in each strategy. Finally, by pointing out several future research directions, it is expected to draw more attention and efforts into the emerging 5G V2X services.

To support the demand of multi-Gbps sensory data exchanges for enhancing (semi)-autonomous driving, millimeter-wave bands (mmWave) vehicular-to- infrastructure (V2I) communications have attracted intensive attention. Unfortunately, the vulnerability to blockages over mmWave bands poses significant design challenges, which can be hardly addressed by manipulating end transceivers, such as beamforming techniques. In this paper, we propose to enhance mmWave V2I communications by augmenting the transmission environments through reflection, where highly-reflective cheap metallic plates are deployed as tunable reflectors without damaging the aesthetic nature of the environments. In this way, alternative indirect line-of-sight (LOS) links are established by adjusting the angle of reflectors. Our fundamental challenge is to adapt the time-consuming reflector angle tuning to the highly dynamic vehicular environment. By using deep reinforcement learning, we propose the learning-based Fast Reflection (LFR) algorithm, which autonomously learns from the observable traffic pattern to select desirable reflector angles in advance for probably blocked vehicles in near future. Simulation results demonstrate our proposal could effectively augment mmWave V2I transmission environments with significant performance gain.

This paper studies the robust energy efficient beamforming design for MISO systems where only channel distribution information (CDI) is assumed to be available at the transmitter. To capture the general relationship between the data transmission and the energy consumption, the dynamic energy consumption model (DECN) is adopted. An optimization problem is formulated to maximize the system energy efficiency under the constraints of rate outage probability and total available power. The problem is difficult to tackle due to the fractional objective function and the information outage constraints. To solve it, the semidefinite relaxation (SDR) is applied at first and then, a solution approach based on the successive convex approximation (SCA) and the Dinklebach's methods is presented. It is proved that our proposed solution approach is able to converge to a stationary point of the formulated optimization problem. Numerical results demonstrate that DECN has a great impact on system EE. It is observed that there is a saturation point on the system EE in term of available power and the required power corresponding to the saturation point of EE highly depends on circuit power. Particularly, higher circuit power leads to a larger required power but a smaller maximal system EE.

Vehicle-based crowdsourcing is becoming a powerful paradigm that can outsource intensive tasks to vehicles by exploiting their on-board resources. In this paper, we focus on the problem of motivating vehicles to join the crowdsourcing system. Considering the various delay demands of tasks in smart cities, we design a delay-aware incentive mechanism to employ vehicles based on reverse auction. Specifically, by taking task delay into consideration, we model the utility of service requester as a function closely related to when its released tasks would be completed. In our mechanism, the participating vehicles bid for their preferred tasks by submitting not only the bidding prices, but also the estimated time of completion (ETC). To maximize the utility of the service requester under a budget constraint, the proposed delay-aware mechanism is cast as a nonmonotone submodular maximization problem with a knapsack constraint. Due to the NP-hardness of the formulated problem, we develop an approximate algorithm for bid selection and payment determination, which guarantees truthfulness, budget feasibility, individual rationality, profitability, and computational efficiency. Simulation results demonstrate the effectiveness of our proposed incentive mechanism.

In cellular Internet of things (IoT) systems, massive low-power terminals update information status to cellular base stations to support diverse IoT applications. In this circumstance, information freshness and energy efficiency become two fundamental concerns. Except data transmissions, information updates consume additional energy for radio activation. To improve the energy efficiency, it is reasonable to aggregate the dynamically generated data. However, the reduced updates will severely deteriorate the information freshness, especially for time-critical IoT applications. To address this issue, we propose an upload scheduling scheme in this paper. Considering dynamic packet arrivals and channel conditions, the upload scheduling problem is formulated from a long-term perspective. To solve this problem, a practical online upload scheduler, named as FReshness-aware Energy efficient ScHeduler (FRESH), is proposed to minimize the update energy consumption subject to information freshness constraints. We theoretically show that FRESH can make the energy saving arbitrarily close to that of the optimal scheduling decision. Simulation results demonstrate the necessity and effectiveness of implementing FRESH for cellular IoT systems.

The integration of sub-6 GHz and millimeter wave (mmWave) bands has a great potential to enable both reliable coverage and high data rate in future vehicular networks. Nevertheless, during mmWave vehicle-to-infrastructure (V2I) handovers, the coverage blindness of directional beams makes it a significant challenge to discover target mmWave remote radio units (mmW-RRUs) whose active beams may radiate somewhere that the handover vehicles are not in. Besides, fast and soft handovers are also urgently needed in vehicular networks. Based on these observations, to solve the target discovery problem, we utilize channel state information (CSI) of sub-6 GHz bands and Kernel-based machine learning (ML) algorithms to predict vehicles' positions and then use them to pre-activate target mmW-RRUs. Considering that the regular movement of vehicles on almost linearly paved roads with finite corner turns will generate some regularity in handovers, to accelerate handovers, we propose to use historical handover data and K-nearest neighbor (KNN) ML algorithms to predict handover decisions without involving time-consuming target selection and beam training processes. To achieve soft handovers, we propose to employ vehicle-to-vehicle (V2V) connections to forward data for V2I links. The theoretical and simulation results are provided to validate the feasibility of the proposed schemes.

In heterogeneous cellular networks (HetNets), cell association of User Equipment (UE) affects UE transmit rate and network throughput. Conventional cell association rules are usually based on UE received Signal-to-Interference-and-Noise-Ratio (SINR) without taking into account user behaviors, which can indeed be exploited for improving network performance. In this paper, we investigate UE cell association in HetNets based on individual user behavior characteristics with aim to maximize long- term expected system throughput. We model the problem as a stochastic optimization model Restless Multi-Armed Bandit (RMAB). As it is a PSPACE-hard problem, we develop a primal-dual heuristic index algorithm and the solution specifies the rule that determines which arms in the RMAB model to be selected at each decision time. According to the solution of RMAB, we propose a new cell association strategy called Index Enabled Association (IDEA). We also conduct simulation experiments to compare IDEA with conventional max-SINR cell association strategy and an existing game-based RAT selection scheme. Numerical results demonstrate the advantages of IDEA in typical scenarios.

In dense and heterogeneous networks, the decoupled uplink#x002F;downlink (UL#x002F;DL) access (DUDA) design has drawn great attentions for improving system performance. Energy efficiency (EE) becomes a major concern for densely deployed heterogeneous cellular networks (HetNets). In this paper, we theoretically analyze the energy efficient sleep strategy for DUDA HetNets. Through using stochastic geometry theory, we first examine the applicability of conventional sleep strategy to DUDA networks and design a new DUDA sleep strategy. We then formulate the energy consumption minimization problem and EE optimization problem, and derive the optimal BS sleep probability. Numerical results reveal that conventional sleep strategy may provide inaccurate guidance for sleep design in DUDA networks, which may lead to excessive sleeps and decrease system EE. Meanwhile our DUDA sleep strategy can effectively reduce network energy consumption. We also find that the dense deployment of small cells may generally increase network EE, but this improvement saturates as the BS density further increases.

The rapid evolution of cellular networks has brought great changes to mobile network architecture. One trend is the dense deployment of base stations (BSs) in heterogeneous cellular network (HetNets) architecture. On the other hand, the booming mobile Internet applications introduce increasingly significant imbalance in regard to Signal to Interference and Noise Ratio (SINR) statistics and traffic load between uplink (UL) and downlink (DL) in HetNets. These evolutions inspire us to exploit decoupling of UL and DL in HetNets for improving system performance. In this paper, we conduct a comparison study for the system performance of the decoupled UL/DL access (DUDA) mode and traditional coupled UL/DL access (CUDA) mode based on stochastic geometry theory. Compared to existing related work, we establish an analytical model for CUDA mode as a comparison reference and consider a more realistic system model, where we employ dynamic transmit power control in UL transmission by applying fractional power control (FPC) to model a location-dependent per-mobile power state. Numerical results reveal that DUDA mode significantly outperforms CUDA mode in terms of system rate, spectral efficiency (SE) and energy efficiency (EE) in HetNets. In addition, results also show that DUDA mode can improve load balance and fairness. Simulation results further validate the accuracy of our analytical model.