Large Language Models (LLMs) have achieved impressive reasoning abilities, but struggle with temporal understanding, especially when questions involve multiple entities, compound operators, and evolving event sequences. Temporal Knowledge Graphs (TKGs), which capture vast amounts of temporal facts in a structured format, offer a reliable source for temporal reasoning. However, existing TKG-based LLM reasoning methods still struggle with four major challenges: maintaining temporal faithfulness in multi-hop reasoning, achieving multi-entity temporal synchronization, adapting retrieval to diverse temporal operators, and reusing prior reasoning experience for stability and efficiency. To address these issues, we propose MemoTime, a memory-augmented temporal knowledge graph framework that enhances LLM reasoning through structured grounding, recursive reasoning, and continual experience learning. MemoTime decomposes complex temporal questions into a hierarchical Tree of Time, enabling operator-aware reasoning that enforces monotonic timestamps and co-constrains multiple entities under unified temporal bounds. A dynamic evidence retrieval layer adaptively selects operator-specific retrieval strategies, while a self-evolving experience memory stores verified reasoning traces, toolkit decisions, and sub-question embeddings for cross-type reuse. Comprehensive experiments on multiple temporal QA benchmarks show that MemoTime achieves overall state-of-the-art results, outperforming the strong baseline by up to 24.0%. Furthermore, MemoTime enables smaller models (e.g., Qwen3-4B) to achieve reasoning performance comparable to that of GPT-4-Turbo.

While differential privacy (DP) contributes to pre serving data privacy during federated learning (FL), DP-FL suffers from either premature convergence or underutilized privacy budgets and subsequently degraded accuracy. Some recent studies heuristically adjusted the variance of the DP noises but offered no guarantee of optimality, little insight, and limited scalability. This paper presents a new control framework for (ε, δ)-DP FL to address the prevalent issues of DP-FL, i.e., premature convergence or underutilized privacy budgets. The key idea is to interpret the DP perturbation of DP-FL as a control process, where the DP noise variance and communication rounds are interdependent and jointly and adaptively determined. An optimal control framework is proposed to adjust the communication rounds and DP noise variance, adapting to the training accuracy of DP-FL. The optimality gap of (ε, δ)-DP FL is derived under the optimal control framework. The importance of joint orchestration of the DP noise and communication rounds is delineated. Experiments on MLP, CNN, and ResNet-9 models show that, given a privacy level, our control framework allows DP-FL to converge much faster with better accuracy than existing techniques, including those with persistent or heuristically reconfigurable DP noise variances.

Differential privacy (DP) is a leading paradigm for privacy preservation in statistical analysis and learning. Traditional DP mechanisms add noise independently of the original data, which yields inconsistent perturbations across groups and raises fairness concerns in downstream decision and learning tasks. Prior work often assesses fairness via bias and variance, while overlooking noise direction and the scale of the underlying query. We propose a novel Fairly Proportional Noise Mechanism (FPNM) that uniquely considers both the direction and magnitude of noise relative to raw query results. We define mathematical formulations for unfairness, factoring in weighting and temporal decay to allow nonlinear amplification of unfairness. The privacy analysis shows a negative correlation between unfairness and privacy strength that higher privacy levels lead to increased noise and thus greater unfairness. We then generalize to group-level assessment using the average unfairness and the Frobenius norm ($F$-Norm). We also prove that adaptive budget reallocation within a data independent feasible domain preserves the overall$(\epsilon ,\delta )$-DP guarantee. Experiments on both decision and learning tasks demonstrate consistent gains. In decision tasks, the proposed FPNM effectively reduces unfairness, achieving average reductions of 19.17% and 17.32% in$F$-Norm and average unfairness, respectively. In learning tasks, integrating FPNM with DP-SGD achieves fairness comparable to fairness-aware baselines and better accuracy. Besides, empirical privacy remains intact under membership inference attacks. These results highlight its effectiveness in improving utility while preserving privacy, offering a robust and comprehensive approach to enhancing fairness in DP.

The critical energy infrastructure is undergoing two significant transformations: the rapid increase in renewable distributed energy resources (DER) and the digitalization of the energy sector, collectively shaping what is known as the Internet of Energy (IoE). Artificial intelligence (AI) has become a widely adopted tool for effectively allocating energy and managing sector-related resources, where ensuring fairness is essential. While inherent unfairness in AI systems is well acknowledged, little attention has been given to evaluating this unfairness and its real-world implications within the context of the IoE. In this study, we take a first step to elucidate the unfairness in AI-powered IoE systems induced by malicious users. We introduce Unfairness Score (UScore), a novel metric designed to evaluate the unfairness of machine learning models in real-world IoE scenarios. We then extensively evaluate unfairness attacks using three IoE tabular datasets, demonstrating that AI model fairness can be compromised through data poisoning, whether in centralized learning (CL) or federated learning (FL) settings. Notably, such compromises can occur when malicious users tamper with only a small subset of the data they control. Finally, we propose a novel approach that unifies fairness and differential privacy (DP) by leveraging DP as a provable defense mechanism. This approach provides a universally applicable solution to unfairness attacks, regardless of whether the learning tasks are classification or regression, and is effective in both FL and CL settings. Our contributions represent a significant step in addressing unfairness and privacy concerns in AI-powered IoE systems.

In recent years, contrastive learning has made significant progress in DeepFake detection. However, existing methods emphasize class granularity, and it is difficult to distinguish between the real instance and its forgery counterparts effectively. Furthermore, the diversity of forgery cues produced by different manipulation methods cannot be effectively clustered by class granularity alone. Thus, the model’s generalization capability is limited. To tackle the above problems, a Dual-Granularity Contrastive Learning (DGCL) for DeepFake detection is proposed in this paper. Specifically, Class Granularity Contrastive Learning (CGCL) and Instance Granularity Contrastive Learning (IGCL) are designed. Firstly, for semantic aggregation at the class level, CGCL incorporates the class prototype, which encourages anchor approaches to the prototype of the positive class, thereby pulling the intra-class features closer. Secondly, for distinguishing between real and fake instances, Real Instance Granularity Contrastive Learning (RIGCL) and Fake Instance Granularity Contrastive Learning (FIGCL) are proposed based on the instance characteristics. RIGCL endeavors to distinguish fake instances from original real instances by expanding the differentiation in the feature space. Meanwhile, FIGCL extracts consistent forgery features from various manipulation methods using cosine similarity constraints. Finally, the superiority and generalizability of DGCL are validated by the experimental results on CELEBDF, DFD, and DFDC datasets.

Recent attacks on federated learning (FL) can introduce malicious model updates that can circumvent widely adopted Euclidean distance-based detection methods. This article proposes a novel defense strategy, referred to as LayerCAM-AE, designed to counteract model poisoning in FL. The LayerCAM-AE puts forth a new Layer Class Activation Mapping (LayerCAM) integrated with an autoencoder (AE), significantly enhancing detection capabilities. Specifically, LayerCAM-AE generates a heat map for each local model update, which is then transformed into a more compact visual explanation. The autoencoder processes the LayerCAM heat maps from the local model updates, improving their distinctiveness and increasing the accuracy in spotting anomalous maps and malicious local models. To mitigate the risk of misclassifications in LayerCAM-AE, a voting algorithm is developed, where a local model update is flagged as malicious if its heat maps are consistently suspicious over several communication rounds. Extensive tests on the SVHN and CIFAR-100 datasets are performed under both Independent and Identically Distributed (IID) and non-IID settings in comparison with the state-of-the-art ResNet-50 and REGNETY-800MF defense models. The experimental results show that LayerCAM-AE increases detection rates (Recall: 1.0, Precision: 1.0, FPR: 0.0, Accuracy: 1.0, F1 score: 1.0, AUC: 1.0) and the test accuracy of FL, surpassing both the ResNet-50 and REGNETY-800MF. Our code is available at: https://github.com/jjzgeeks/LayerCAM-AE .

Spatio-temporal trajectories carry identifying information and are vulnerable to privacy breaches. Existing studies predominantly focus on the spatial domain. The temporal aspect remains underexplored, leaving privacy risks unaddressed. This paper highlights these risks by introducing a new trajectory matching model, ST-ATT, which leverages attention-enhanced Long Short-Term Memory (LSTM) to effectively capture the spatio-temporal correlations within trajectories. ST-ATT excels in identifying similar trajectories. To defend against linkage attacks on spatio-temporal trajectories, including advanced models like ST-ATT, we propose a novel Differential Privacy (DP) mechanism specifically designed to address the privacy risks. We reveal that the privacy budget and violation probability for each spatial point explicitly depend on earlier timestamps. The privacy budget can be flexibly redistributed between spatial and temporal domains without compromising overall privacy. This mechanism complies with DP, even when spatio-temporal points are reordered due to perturbation. Experiments show that ST-ATT can accurately identify spatio-temporal trajectories perturbed by the existing DP methods adding noise solely to the spatial domain. The proposed spatio-temporal DP mechanism resists ST-ATT, highlighting the need for considering spatio-temporal correlations to ensure robust privacy protection in spatio-temporal trajectories.

Retrieval-augmented generation (RAG) enhances large language models (LLMs) by incorporating external knowledge.Current hybrid RAG system retrieves evidence from both knowledge graphs (KGs) and text documents to support LLM reasoning.However, it faces challenges like handling multi-hop reasoning, multi-entity questions, multi-source verification, and effective graph utilization.To address these limitations, we present HydraRAG, a training-free framework that unifies graph topology, document semantics, and source reliability to support deep, faithful reasoning in LLMs.HydraRAG handles multi-hop and multi-entity problems through agent-driven exploration that combines structured and unstructured retrieval, increasing both diversity and precision of evidence.To tackle multisource verification, HydraRAG uses a tri-factor cross-source verification (source trustworthiness assessment, cross-source corroboration, and entity-path alignment), to balance topic relevance with cross-modal agreement.By leveraging graph structure, HydraRAG fuses heterogeneous sources, guides efficient exploration, and prunes noise early.Comprehensive experiments on seven benchmark datasets show that HydraRAG achieves overall state-of-theart results on all benchmarks with GPT-3.5, outperforming the strong hybrid baseline ToG-2 by an average of 20.3% and up to 30.1%.Furthermore, HydraRAG enables smaller models (e.g., Llama-3.1-8B) to achieve reasoning performance comparable to that of GPT-4-Turbo.The source code is available on https: //stevetantan.github.io/HydraRAG/.

In the Internet of Vehicles (IoV), safeguarding service privacy and communication efficiency is critical due to the vast exchange of sensitive data among connected entities. Federated Learning (FL) has emerged as a privacy-preserving paradigm for collaborative model training without exposing raw data. However, a fundamental trade-off arises between the level of differential privacy (DP) protection and the number of communication rounds required. To address this, we propose the Unified State-Space Feedback Control Framework (USFCF), which introduces a dynamic feedback regulation mechanism that adaptively balances privacy protection and communication overhead. Roadside units (RSUs) serve as distributed coordinators, monitoring the evolving system state and adjusting the model aggregation frequency accordingly. When the system detects excessive privacy noise, it suppresses redundant communication; conversely, if privacy weakens, it increases update rounds to reinforce protection. The derivation of an optimization-driven update policy is constructed to realize the privacy-communication of the FL-DP system. Experimental evaluations on diverse datasets demonstrate that our method enhances privacy adaptability, reduces communication cost, and ensures robust performance for privacy-sensitive vehicular intelligence.

Large Language Models (LLMs) have achieved impressive results in various tasks but struggle with hallucination problems and lack of relevant knowledge, especially in deep complex reasoning and knowledge-intensive tasks.Knowledge Graphs (KGs), which capture vast amounts of facts in a structured format, offer a reliable source of knowledge for reasoning.However, existing KG-based LLM reasoning methods face challenges like handling multi-hop reasoning, multi-entity questions, and effectively utilizing graph structures.To address these issues, we propose Paths-over-Graph (PoG), a novel method that enhances LLM reasoning by integrating knowledge reasoning paths from KGs, improving the interpretability and faithfulness of LLM outputs.PoG tackles multi-hop and multi-entity questions through a three-phase dynamic multi-hop path exploration, which combines the inherent knowledge of LLMs with factual knowledge from KGs.In order to improve the efficiency, PoG prunes irrelevant information from the graph exploration first and introduces efficient three-step pruning techniques that incorporate graph structures, LLM prompting, and a pre-trained language model (e.g., SBERT) to effectively narrow down the explored candidate paths.This ensures all reasoning paths contain highly relevant information captured from KGs, making the reasoning faithful and interpretable in problem-solving.PoG innovatively utilizes graph structure to prune the irrelevant noise and represents the first method to implement multi-entity deep path detection on KGs for LLM reasoning tasks.Comprehensive experiments on five benchmark KGQA datasets demonstrate PoG outperforms the stateof-the-art method ToG across GPT-3.5-Turbo and GPT-4, achieving an average accuracy improvement of 18.9%.Notably, PoG with GPT-3.5-Turbosurpasses ToG with GPT-4 by up to 23.9%.

This paper focuses on Zero-Trust Foundation Models (ZTFMs), a novel paradigm that embeds zero-trust security principles into the lifecycle of foundation models (FMs) for Internet of Things (IoT) systems. By integrating core tenets, such as least privilege access, continuous verification, data confidentiality, and behavioral analytics into the design, training, and deployment of FMs, ZTFMs can enable secure, privacy-preserving AI across distributed, heterogeneous, and potentially adversarial IoT environments. We present the first structured synthesis of ZTFMs, identifying their potential to transform conventional trust-based IoT architectures into resilient, self-defending ecosystems. Moreover, we propose a comprehensive technical framework, incorporating federated learning (FL), blockchain-based identity management, micro-segmentation, and trusted execution environments (TEEs) to support decentralized, verifiable intelligence at the network edge. In addition, we investigate emerging security threats unique to ZTFM-enabled systems and evaluate countermeasures, such as anomaly detection, adversarial training, and secure aggregation. Through this analysis, we highlight key open research challenges in terms of scalability, secure orchestration, interpretable threat attribution, and dynamic trust calibration. This survey lays a foundational roadmap for secure, intelligent, and trustworthy IoT infrastructures powered by FMs.

Inherent communication noises have the potential to preserve privacy for wireless federated learning (WFL) but have been overlooked in digital communication systems predominantly using floating-point number standards,e.g., IEEE 754, for data storage and transmission. This is due to the potentially catastrophic consequences of bit errors in floating-point numbers,e.g., on the sign or exponent bits. This paper presents a novel channel-native bit-flipping differential privacy (DP) mechanism tailored for WFL, where transmit bits are randomly flipped and communication noises are leveraged, to collectively preserve the privacy of WFL in digital communication systems. The key idea is to interpret the bit perturbation at the transmitter and bit errors caused by communication noises as a bit-flipping DP process. This is achieved by designing a new floating-point-to-fixed-point conversion method that only transmits the bits in the fraction part of model parameters, hence eliminating the need for transmitting the sign and exponent bits and preventing the catastrophic consequence of bit errors. We analyze a new metric to measure the bit-level distance of the model parameters and prove that the proposed mechanism satisfies (λ, ϵ)-Rényi DP and does not violate the WFL convergence. Experiments validate privacy and convergence analysis of the proposed mechanism and demonstrate its superiority to the state-of-the-art Gaussian mechanisms that are channel-agnostic and add Gaussian noise for privacy protection.

Physical layer (PHY) secret key generation (SKG) has been widely studied as a promising approach to achieving One-Time-Pad security. The improvement of SKG rate is quite a huge challenge, especially in scenarios with untrusted internal helpers or eavesdroppers that aim to wiretap the negotiated secret keys between legitimate parties. In this paper, we propose a token-based SKG scheme to deal with the problem of information leakage with internal eavesdropping attacks. The basic idea is to cover random pilots with protective tokens to confuse eavesdroppers. Three scenarios including passive external eavesdropping, active internal eavesdropping with a reconfigurable intelligent surface (RIS)-assisted untrusted helper, and active internal eavesdropping with an untrusted relay are considered and analyzed to evaluate the performance of the proposed anti-eavesdropping scheme. Theoretical analysis shows that the proposed token-based SKG scheme can perfectly secure the key negotiation, achieving zero information leakage even in the untrusted relaying scenario without a direct link between Alice and Bob. Moreover, closed-form expressions for secret key capacity (SKC) are obtained. Finally, numerical results indicate that the proposed scheme outperforms the state-of-the-art methods. Using a token-generation mapping function with greater diversity in amplitude and phase, our approach achieves enhanced SKC performance across various scenarios, including those with a passive eavesdropper, a RIS-assisted untrusted helper, and an untrusted relay.

Differential Privacy (DP) is a preeminent technique for data privacy by introducing noise to sensitive information. However, traditional DP mechanisms excessively rely on third parties to ensure traceability, necessitating strong background assumptions that are frequently impractical in real-world scenarios. This reliance makes it difficult to preserve both privacy and traceability. To address these challenges, we propose a novel Traceable and Collision-Resilient Differential Privacy (TCRDP) mechanism. The TCRDP mechanism simultaneously publishes perturbed results and data fingerprints, retaining partial information from the original data in a collision-resilient manner to facilitate future verification. Moreover, the TCRDP mechanism integrates an innovative noise generation process, leveraging hash values and a customized Laplace-like distribution to produce noise. This strategy mitigates the risk of adversaries compromising privacy through enumeration and yields a more concentrated noise distribution with reduced variance. We evaluated the TCRDP mechanism using three datasets: ICUs, Diabetes, and RAHRD, across various query types. The experimental results demonstrated significant improvements in data utility, with the TCRDP mechanism achieving great reductions in Mean Absolute Error (MAE) and Mean Squared Error (MSE) compared to traditional mechanisms. The TCRDP mechanism also maintained lower Accuracy Loss (AL) across different privacy budgets and dataset sizes, highlighting its robustness and scalability. These findings underscore the potential of the TCRDP mechanism to advance privacy-preserving data analysis, offering significant enhancements over existing methods in both accuracy and utility.

Recent poisoning attacks on federated learning (FL) generate malicious model updates that circumvent widely adopted Euclidean distance-based detection methods. This article proposes a new defense mechanism, namely, GradCAM-AE, against model poisoning attacks on FL, which integrates Gradient-weighted Class Activation Mapping (GradCAM) and autoencoder (AE) to offer a substantially more powerful detection capability compared to existing Euclidean distance-based approaches. Particularly, GradCAM-AE generates a heat map for each uploaded local model update, transforming each local model update into a lower-dimensional, visual representation. An AE further reprojects the GradCAM heat maps of all local module updates with improved distinguishability, thereby accentuating the hidden features of the heat maps and increasing the success rate of identifying anomalous heat maps and malicious local models. A comprehensive evaluation of the proposed GradCAM-AE framework is conducted using the CIFAR-10 and GTSRB datasets under both Independent and Identically Distributed (IID) and Non-IID settings. The ResNet-18 and MobileNetV3-Large models are tested. The results substantiate that GradCAM-AE offers superior detection rates and test accuracy of FL global model, juxtaposed with contemporary state-of-the-art methods. Our code is available at: https://github.com/jjzgeeks/GradCAM-AE .

This article puts forth a new training data-untethered model poisoning (MP) attack on federated learning (FL). The new MP attack extends an adversarial variational graph autoencoder (VGAE) to create malicious local models based solely on the benign local models overheard without any access to the training data of FL. Such an advancement leads to the VGAE-MP attack that is not only efficacious but also remains elusive to detection. VGAE-MP attack extracts graph structural correlations among the benign local models and the training data features, adversarially regenerates the graph structure, and generates malicious local models using the adversarial graph structure and benign models' features. Moreover, a new attacking algorithm is presented to train the malicious local models using VGAE and sub-gradient descent, while enabling an optimal selection of the benign local models for training the VGAE. Experiments demonstrate a gradual drop in FL accuracy under the proposed VGAE-MP attack and the ineffectiveness of existing defense mechanisms in detecting the attack, posing a severe threat to FL.

This paper develops a novel framework for differentially private (DP) wireless federated learning (FL) with integrated sensing and communication (ISAC). In this framework, which is referred to as DP-ISAC-FL, wireless devices sense data and upload the trained local models using ISAC technique. The local training can take place concurrently with sensing at each device. We analyze the convergence upper bound of DP-ISAC-FL and rigorously capture the impact of device selection (for model training), time allocation between sensing/training and model uploading for the selected devices, and the allocations of channels, modulations, and transmit powers. We also develop an algorithm that enforces the convergence of DP-ISAC-FL by minimizing the convergence upper bound in an OFDMA system with discrete modulations. The beamforming for sensing, device selection, and the allocations of time, subchannels, modulations, and transmit powers are jointly optimized using successive convex approximation (SCA), adapting to the channels and computing capabilities of the devices. Experiments on multilayer perceptrons (MLPs) and convolutional neural networks (CNNs) show that DP-ISAC-FL with optimal allocations can significantly improve the learning convergence and accuracy under different privacy levels, e.g., by 7% and 18%, compared with its benchmarks. This is attributed to 68% more sensing data that DP-ISAC-FL can admit for model training.

This paper proposes a new visual explanation-based defense mechanism, namely, FedCAMAE, against model poisoning attacks on federated learning (FL), which integrates Layer Class Activation Mapping (LayerCAM) and autoencoder to offer a scientifically more powerful detection capability compared to existing Euclidean distance-based or machine learning-based approaches. Specially, FedCAMAE generates a fine-grained heat map assisted by Layer-CAM for each uploaded local model update, transforming each local model update into a lower-dimensional, visual representation. To accentuate the hidden features of the heat maps, autoencoder is seamlessly embedded into the proposed FedCAMAE, which can refine the the heat maps and enhance their distinguishability, thereby increasing the success rate of identifying anomalous heat maps and malicious local models. We test ResNet-50 and REGNETY-800MF deep learning models with SVHN and CIFAR-100 datasets under Non-Independent and Identically Distributed (Non-IID) setting, respectively. The results demonstrate that Fed-CAMAE offers superior test accuracy of FL global model compared to the state-of-the-art methods. Our code is available at: https://github.com/jjzgeeks/LayerCAM-AE

The objective of differential privacy (DP) is to protect privacy by producing an output distribution that is indistinguishable between any two neighboring databases. However, traditional differentially private mechanisms tend to produce unbounded outputs in order to achieve maximum disturbance range, which is not always in line with real-world applications. Existing solutions attempt to address this issue by employing post-processing or truncation techniques to restrict the output results, but at the cost of introducing bias issues. In this paper, we propose a novel differentially private mechanism which uses a composite probability density function to generate bounded and unbiased outputs for any numerical input data. The composition consists of an activation function and a base function, providing users with the flexibility to define the functions according to the DP constraints. We also develop an optimization algorithm that enables the iterative search for the optimal hyper-parameter setting without the need for repeated experiments, which prevents additional privacy overhead. Furthermore, we evaluate the utility of the proposed mechanism by assessing the variance of the composite probability density function and introducing two alternative metrics that are simpler to compute than variance estimation. Our extensive evaluation on three benchmark datasets demonstrates consistent and significant improvement over the traditional Laplace and Gaussian mechanisms. The proposed bounded and unbiased composite differentially private mechanism will underpin the broader DP arsenal and foster future privacy-preserving studies.

The task of cardinality counting, pivotal for data analysis, endeavors to quantify unique elements within datasets and has significant applications across various sectors like healthcare, marketing, cybersecurity, and web analytics. Current methods, categorized into deterministic and probabilistic, often fail to prioritize data privacy. Given the fragmentation of datasets across various organizations, there is an elevated risk of inadvertently disclosing sensitive information during collaborative data studies using state-of-the-art cardinality counting techniques. This study introduces an innovative privacy-centric solution for the cardinality counting dilemma, leveraging a federated learning framework. Our approach involves employing a locally differentially private data encoding for initial processing, followed by a privacy-aware federated K-means clustering strategy, ensuring that cardinality counting occurs across distinct datasets without necessitating data amalgamation. The efficacy of our methodology is underscored by promising results from tests on both real-world and simulated datasets, pointing towards a transformative approach to privacy-sensitive cardinality counting in contemporary data science.

In this paper, we propose a joint single-base localization and communication enhancement scheme for the uplink (UL) integrated sensing and communications (ISAC) system with asynchronism, which can achieve accurate single-base localization of user equipment (UE) and significantly improve the communication reliability despite the existence of timing offset (TO) due to the clock asynchronism between UE and base station (BS). Our proposed scheme integrates the CSI enhancement into the multiple signal classification (MUSIC)-based AoA estimation and thus imposes no extra complexity on the ISAC system. We further exploit a MUSIC-based range estimation method and prove that it can suppress the time-varying TO-related phase terms. Exploiting the AoA and range estimation of UE, we can estimate the location of UE. Finally, we propose a joint CSI and data signals-based localization scheme that can coherently exploit the data and the CSI signals to improve the AoA and range estimation, which further enhances the single-base localization of UE. The extensive simulation results show that the enhanced CSI can achieve equivalent bit error rate performance to the minimum mean square error (MMSE) CSI estimator. The proposed joint CSI and data signals-based localization scheme can achieve decimeter-level localization accuracy despite the existing clock asynchronism and improve the localization root mean square error (RMSE) by about 6 dB compared with the maximum likelihood esimation (MLE)-based benchmark method.

In this paper, we propose a novel Kalman Filter (KF)-based uplink (UL) joint communication and sensing (JCAS) scheme, which can significantly reduce the range and location estimation errors due to the clock asynchronism between the base station (BS) and user equipment (UE). Clock asynchronism causes time-varying time offset (TO) and carrier frequency offset (CFO), leading to major challenges in uplink sensing. Unlike existing technologies, our scheme does not require knowing the location of the UE in advance, and retains the linearity of the sensing parameter estimation problem. We first estimate the angle-of-arrivals (AoAs) of multipaths and use them to spatially filter the CSI. Then, we propose a KF-based CSI enhancer that exploits the estimation of Doppler with CFO as the prior information to significantly suppress the time-varying noise-like TO terms in spatially filtered CSIs. Subsequently, we can estimate the accurate ranges of UE and the scatterers based on the KF-enhanced CSI. Finally, we identify the UE’s AoA and range estimation and locate UE, then locate the dumb scatterers using the bi-static system. Simulation results validate the proposed scheme. The localization root mean square error of the proposed method is about 20 dB lower than the benchmarking scheme.

Integrated sensing and communication (ISAC) is regarded as the enabling technology in the future 5th-Generation-Advanced (5G-A) and 6th-Generation (6G) mobile communication system. ISAC waveform design is critical in ISAC system. However, the difference of the performance metrics between sensing and communication brings challenges for the ISAC waveform design. This paper applies the unified performance metrics in information theory, namely mutual information (MI), to measure the communication and sensing performance in multicarrier ISAC system. In multi-input multi-output orthogonal frequency division multiplexing (MIMO-OFDM) ISAC system, we first derive the sensing and communication MI with subcarrier correlation and spatial correlation. Then, we propose optimal waveform designs for maximizing the sensing MI, communication MI and the weighted sum of sensing and communication MI, respectively. The optimization results are validated by Monte Carlo simulations. Our work provides effective closed-form expressions for waveform design, enabling the realization of MIMO-OFDM ISAC system with balanced performance in communication and sensing.

This paper proposes a novel, data-agnostic, model poisoning attack on Federated Learning (FL), by designing a new adversarial graph autoencoder (GAE)-based framework. The attack requires no knowledge of FL training data and achieves both effectiveness and undetectability. By listening to the benign local models and the global model, the attacker extracts the graph structural correlations among the benign local models and the training data features substantiating the models. The attacker then adversarially regenerates the graph structural correlations while maximizing the FL training loss, and subsequently generates malicious local models using the adversarial graph structure and the training data features of the benign ones. A new algorithm is designed to iteratively train the malicious local models using GAE and sub-gradient descent. The convergence of FL under attack is rigorously proved, with a considerably large optimality gap. Experiments show that the FL accuracy drops gradually under the proposed attack and existing defense mechanisms fail to detect it. The attack can give rise to an infection across all benign devices, making it a serious threat to FL.

The use of legitimate unmanned aerial vehicles (UAVs) to surveil and track misbehaved UAVs can serve a crucial role in public safety and security. This paper proposes a new deep reinforcement learning (DRL)-based online control scheme for visual-based UAV-on-UAV tracking and monitoring, where a solar-powered, fixed-wing UAV tracks a suspicious UAV target by having the target inside its effective visual range. The key idea is a new deep deterministic policy gradient (DDPG)-based model, which can cope with the continuous state and action spaces of the monitor and learn the optimal acceleration control policy adapting to the solar power availability and the target’s movement. The state space is designed to be the relative position of the monitor to the target, thereby preventing model infeasibility. Experiments show that the new algorithm can maintain a desired distance from the target, and outperform control-and optimization-based alternatives in terms of energy efficiency and tracking accuracy. An interesting finding is that our algorithm learns faster and better with a constraint of a minimum allowed battery energy reserve. The reason is that, without the constraint, the monitor is more likely to deplete its battery before the end of a surveillance mission.

Deep neural networks (DNNs) are vulnerable to adversarial examples, while adversarial attack models, e.g., DeepFool, are on the rise and outrunning adversarial example detection techniques. This article presents a new adversarial example detector that outperforms state-of-the-art detectors in identifying the latest adversarial attacks on image datasets. Specifically, we propose to use sentiment analysis for adversarial example detection, qualified by the progressively manifesting impact of an adversarial perturbation on the hidden-layer feature maps of a DNN under attack. Accordingly, we design a modularized embedding layer with the minimum learnable parameters to embed the hidden-layer feature maps into word vectors and assemble sentences ready for sentiment analysis. Extensive experiments demonstrate that the new detector consistently surpasses the state-of-the-art detection algorithms in detecting the latest attacks launched against ResNet and Inception neutral networks on the CIFAR-10, CIFAR-100, and SVHN datasets. The detector only has about 2 million parameters and takes less than 4.6 ms to detect an adversarial example generated by the latest attack models using a Tesla K80 GPU card.

Federated learning (FL) can suffer from communication bottlenecks when deployed in mobile networks, limiting participating clients and deterring FL convergence. In this context, the impact of practical air interfaces with discrete modulation schemes on FL has not previously been studied in depth. This paper proposes a new paradigm of flexible aggregation-based FL (F2L) over an orthogonal frequency division multiple-access (OFDMA) air interface, termed as “OFDMA-F2L”, allowing selected clients to train local models for various numbers of iterations before uploading the models in each aggregation round. We optimize the selections of clients, subchannels and modulation scheme, adapting to channel conditions and computing power. Specifically, we derive an upper bound on the optimality gap of OFDMA-F2L capturing the impact of these selections, and show that the upper bound is minimized by maximizing the weighted sum rate of the clients per aggregation round. A Lagrange-dual based method is developed to solve this challenging mixed integer program of weighted sum rate maximization, revealing that a “winner-takes-all” policy provides the almost surely optimal client, subchannel, and modulation selections. Experiments on multilayer perceptrons and convolutional neural networks show that OFDMA-F2L with optimal selections can significantly improve the training convergence and accuracy, e.g., by about 18% and 5%, compared to potential alternatives.

To realize an intelligent cooperative vehicle infrastructure system and high-level autonomous driving, the introduction of the joint communication and sensing (JCS) technique in vehicular networks is indispensable. With directional beamforming, the vehicles equipped with JCS systems could utilize unified radio-frequency transceivers and frequency band resources to achieve vehicle-to-infrastructure (V2I) communication and sensing functions in different directions, respectively. In this concept, we study the computation offloading problem for JCS-based vehicular networks. Specifically, we formulate a long-term multi-objective problem that jointly optimizes the task execution latency and the sensing performance of multiple vehicles. Owing to the time-varying V2I channel gain, the time-varying impulse response of sensed target, and the stochastic traffic, we reformulate it as a Markov decision process and propose a double-stage deep reinforcement learning-based offloading and power allocation (DDOPA) strategy to determine the task offloading and power allocation for each vehicle. Simulation results demonstrate the efficacy of the proposed strategy compared with different strategies, and show that the proposed DDOPA strategy can achieve a trade-off between execution latency and sensing performance.

To realize higher-level autonomous driving and advanced transportation applications, the introduction of the integrated sensing and communication (ISAC) technique in vehicular networks is indispensable. Different from the existing works, this paper investigates the power allocation problem for onboard ISAC systems of vehicles, during the vehicle-to-infrastructure communication, vehicle-to-vehicle communication and sensing progress, in case of the time-varying communication channel gains, the time-varying impulse responses of sensed targets, and the stochastic traffic. Note that both the inter-beam interference of a single vehicle and the inter-vehicle interference are important considerations. Specifically, we formulate a stochastic programming problem, which optimizes the sensing performance, subject to constraints on the network stability, power limits and quality-of-service requirements. Leveraging the Lyapunov optimization technique, this stochastic programming problem is transformed into a single-time slot non-convex problem. Taking advantages of genetic algorithm and particle swarm optimization (PSO), a hybrid meta-heuristic algorithm is designed to solve the non-convex problem. Typically, we improve the traditional PSO to balance the global search ability and local search ability of particles. Finally, a dynamic power allocation strategy is proposed. The theoretical analysis and simulation results show that this strategy achieves a communication performance-sensing performance tradeoff of [$ {\mathrm {O(}}1/V{\mathrm {)}} $,$ {\mathrm {O(}}V{\mathrm {)}} $] with$ V $being a control parameter.

As the uplink sensing has the advantage of easy implementation, it attracts great attention in integrated sensing and communication (ISAC) system. This paper presents an uplink ISAC system based on multi-input multi-output orthogonal frequency division multiplexing (MIMO-OFDM) technology. The mutual information (MI) is introduced as a unified metric to evaluate the performance of communication and sensing. In this paper, firstly, the upper and lower bounds of communication and sensing MI are derived in details based on the interaction between communication and sensing. And the ISAC waveform is optimized by maximizing the weighted sum of sensing and communication MI. The Monte Carlo simulation results show that, compared with other waveform optimization schemes, the proposed ISAC scheme has the best overall performance.

As a technology intended to reduce cellular network congestion and enhance user service quality, computation offloading in Multi-access Edge Computing (MEC) networks highlights the crucial issue of privacy protection. This paper proposes a novel solution to the computation offloading and privacy protection problem in the MEC network using a Multi-agent Deep Deterministic Policy Gradient (MADDPG) framework. Our approach utilizes game theory to encourage computation offloading by modeling the interaction between cloudlets, Data Center Operator (DCO), and users as an auction game. We formulate the resource allocation and privacy protection as an auction game with multiple bidders and incomplete information and then use MADDPG to find an optimal solution. To ensure privacy protection, we design a Local Differential Privacy (LDP) method in the MADDPG algorithm. Theoretical analysis and simulation results demonstrate the effectiveness of our approach in satisfying differential privacy and converging to an equilibrium. The proposed solution holds significant promise in addressing the computation offloading and privacy protection challenges in MEC networks.

Mobile edge computing (MEC) enables the Internet of Things (IoT) with seamless integration of multiple application services. Federated learning is increasingly considered to improve training accuracy in MEC-IoT while circumventing the disclosure of private data, where the IoT nodes collaboratively train a machine learning model without disclosing their private data. In this paper, we propose a new cyber-epidemic attack that progressively manipulates federated learning and reduces the training accuracy of the benign MEC-IoT. The proposed cyber-epidemic attack explores adversarial graph autoencoders (GACE) to generate malicious local model updates that extract correlated features with the benign local and global models. The proposed GACE attack epidemically infects all the benign IoT nodes along with the training iterations in federated learning, while highly enhancing concealment of the attack.

Unmanned aerial vehicles (UAVs) have been increasingly considered as aerial servers in mobile edge computing (MEC) to assist mission-critical computation tasks of edge ground nodes. The tasks are buffered at the ground node, while the task offloading is scheduled by the UAV. When one ground node in MEC is scheduled to offload its tasks, other unselected ground nodes’ tasks could expire and be cancelled. To maximize the offloaded tasks to the UAV, this paper proposes a new joint optimization of cruise control and task offloading scheduling, which synthetically takes into account the computation capacity and battery energy of the ground nodes, and the speed limit of the UAV. Given a large and unknown network state and action space, a new deep reinforcement learning (DRL) framework based on graph neural networks (GNN) is developed to train online the continuous cruise control of the UAV and the task offloading schedule. Particularly, GNN explores feature correlations of network states to supervise the action training of the UAV in DRL. We implement the proposed GNN-DRL framework on Google Tensorflow. Extensive numerical results show that GNN-DRL improves the task offloading rate by 43%, compared to the DRL solution without GNN.

Summary Privacy leakage in images attracts increasing concerns these days, as photos uploaded to large social platforms are usually not processed by proper privacy protection mechanisms. Moreover, with advanced artificial intelligence (AI) tools such as deep neural network (DNN), an adversary can detect people's identities and collect other sensitive personal information from images at an unprecedented scale. In this paper, we introduce a novel face image de‐identification framework using adversarial perturbations in the feature space. Manipulating the feature space vector ensures the good transferability of our framework. Moreover, the proposed feature space adversarial perturbation generation algorithm can successfully protect the identity‐related information while ensuring the other attributes remain similar. Finally, we conduct extensive experiments on two face image datasets to evaluate the performance of the proposed method. Our results show that the proposed method can generate real‐looking privacy‐preserving images efficiently. Although our framework has only been tested on two real‐life face image datasets, it can be easily extended to other types of images.

This article presents a new deep reinforcement learning (DRL)-based approach to the trajectory planning and jamming rejection of an unmanned aerial vehicle (UAV) for the Internet of Things (IoT) applications. Jamming can prevent timely delivery of sensing data and reception of operation instructions. With the assistance of a reconfigurable intelligent surface (RIS), we propose to augment the radio environment, suppress jamming signals, and enhance the desired signals. The UAV is designed to learn its trajectory and the RIS configuration based solely on changes in its received data rate, using the latest deep deterministic policy gradient (DDPG) and twin delayed DDPG (TD3) models. Simulations show that the proposed DRL algorithms give the UAV with strong resistance against jamming and that the TD3 algorithm exhibits faster and smoother convergence than the DDPG algorithm, and suits better for larger RISs. This DRL-based approach eliminates the need for knowledge of the channels involving the RIS and jammer, thereby offering significant practical value.

In recent years, the Internet of Things (IoT) has been studied in the context of the Metaverse to provide users with immersive cyber-virtual experiences in mixed-reality environments. This survey introduces six typical IoT applications in the Metaverse, including collaborative healthcare, education, smart city, entertainment, real estate, and socialization. In the IoT-inspired Metaverse, we also comprehensively survey four pillar technologies that enable augmented reality (AR) and virtual reality (VR), namely, responsible artificial intelligence (AI), high-speed data communications, cost-effective mobile edge computing (MEC), and digital twins. According to the physical-world demands, we outline the current industrial efforts and seven key requirements for building the IoT-inspired Metaverse: immersion, variety, economy, civility, interactivity, authenticity, and independence. In addition, this survey describes the open issues in the IoT-inspired Metaverse, which need to be addressed to eventually achieve the convergence of physical and cyber worlds.

In recent years, ubiquitous semantic Metaverse has been studied to revolutionize immersive cyber-virtual experiences for augmented reality (AR) and virtual reality (VR) users, which leverages advanced semantic understanding and representation to enable seamless, context-aware interactions within mixed-reality environments. This survey focuses on the intelligence and spatiotemporal characteristics of four fundamental system components in ubiquitous semantic Metaverse, i.e., artificial intelligence (AI), spatiotemporal data representation (STDR), Semantic Internet of Things (SIoT), and semantic-enhanced digital twin (SDT). We thoroughly survey the representative techniques of the four fundamental system components that enable intelligent, personalized, and context-aware interactions with typical use cases of the ubiquitous semantic Metaverse, such as remote education, work and collaboration, entertainment and socialization, healthcare, and e-commerce marketing. Furthermore, we outline the opportunities for constructing the future ubiquitous semantic Metaverse, including scalability and interoperability, privacy and security, performance measurement and standardization, as well as ethical considerations and responsible AI. Addressing those challenges is important for creating a robust, secure, and ethically sound system environment that offers engaging immersive experiences for the users and AR/VR applications.

Integrated sensing and communication (ISAC) has the advantages of efficient spectrum utilization and low hardware cost. It is promising to be implemented in the fifth-generation-advanced (5G-A) and sixth-generation (6G) mobile communication systems, having the potential to be applied in intelligent applications requiring both communication and high-accurate sensing capabilities. As the fundamental technology of ISAC, ISAC signal directly impacts the performance of sensing and communication. This article systematically reviews the literature on ISAC signals from the perspective of mobile communication systems, including ISAC signal design, ISAC signal processing, and ISAC signal optimization. We first review the ISAC signal design based on 5G, 5G-A, and 6G mobile communication systems. Then, radar signal processing methods are reviewed for ISAC signals, mainly including the channel information matrix method, spectrum lines estimator method, and super-resolution method. In terms of signal optimization, we summarize peak-to-average power ratio (PAPR) optimization, interference management, and adaptive signal optimization for ISAC signals. This article may provide the guidelines for the research of ISAC signals in 5G-A and 6G mobile communication systems.

Reconfigurable intelligent surfaces (RISs) can potentially combat jamming. It is non-trivial to perform holistic selections of users, data streams, and modulation-coding modes for all subchannels, and RIS configuration in a downlink multiuser OFDMA system under jamming attacks, because of a mixed-integer program nature and difficulties in acquiring the channel state information (CSI) of the channels to and from the RIS and from an uncooperative jammer. We propose a new deep reinforcement learning (DRL)-based approach that learns through changes in the data rates of the users to reject jamming and maximize the sum rate. The key idea is to decouple the continuous RIS configuration from the discrete selections of users, data streams, subchannels, and modulation-coding modes. Another critical aspect is that we show the optimal selections almost surely follow a winner-takes-all strategy. Accordingly, the new DRL framework learns the RIS configuration with a twin-delayed deep deterministic policy gradient and takes the winner-takes-all strategy to evaluate the reward, thereby reducing the action space and accelerating learning. Simulations show the framework converges fast and fulfills the benefit of the RIS. With no need for the CSI of the channels to and from the RIS and from the jammer, the framework offers practical value.

While preserving the privacy of federated learning (FL), differential privacy (DP) inevitably degrades the utility (i.e., accuracy) of FL due to model perturbations caused by DP noise added to model updates. Existing studies have considered exclusively noise with persistent root-mean-square amplitude and overlooked an opportunity of adjusting the amplitudes to alleviate the adverse effects of the noise. This paper presents a new DP perturbation mechanism with a time-varying noise amplitude to protect the privacy of FL and retain the capability of adjusting the learning performance. Specifically, we propose a geometric series form for the noise amplitude and reveal analytically the dependence of the series on the number of global aggregations and the (ϵ,δ)-DP requirement. We derive an online refinement of the series to prevent FL from premature convergence resulting from excessive perturbation noise. Another important aspect is an upper bound developed for the loss function of a multi-layer perceptron (MLP) trained by FL running the new DP mechanism. Accordingly, the optimal number of global aggregations is obtained, balancing the learning and privacy. Extensive experiments are conducted using MLP, supporting vector machine, and convolutional neural network models on four public datasets. The contribution of the new DP mechanism to the convergence and accuracy of privacy-preserving FL is corroborated, compared to the state-of-the-art Gaussian noise mechanism with a persistent noise amplitude.

Latent graph structure and stimulus of graph-structured data contain critical private information, such as brain disorders in functional magnetic resonance imaging data, and can be exploited to identify individuals. It is critical to perturb the latent information while maintaining the utility of the data, which, unfortunately, has never been addressed. This paper presents a novel approach to obfuscating the latent information and maximizing the utility. Specifically, we first analyze the graph Fourier transform (GFT) basis that captures the latent graph structures, and the latent stimuli that are the spectral-domain inputs to the latent graphs. Then, we formulate and decouple a new multi-objective problem to alternately obfuscate the GFT basis and stimuli. The difference-of-convex (DC) programming and Stiefel manifold gradient descent are orchestrated to obfuscate the GFT basis. The DC programming and gradient descent are employed to perturb the spectral-domain stimuli. Experiments conducted on an attention-deficit hyperactivity disorder dataset demonstrate that our approach can substantially outperform its differential privacy-based benchmark in the face of the latest graph inference attacks.

Unmanned aerial vehicles (UAVs) play a critical role in radio surveillance to decipher malicious messages, thanks to their flexibility, mobility, and likely line-of-sight (LoS) to ground targets. Reconfigurable intelligent surfaces (RISs) can potentially create radio surveillance channels towards the UAVs by passively configuring the radio environments without raising suspicion. This paper presents a new deep reinforcement learning (DRL)-driven framework for radio surveillance, where a fixed-wing UAV is employed to acquire the radio fingerprint of a suspicious transmitter (Tx) with the aid of a benign RIS. A new Twin Delayed Deep Deterministic policy gradient (TD3) model is designed to allow the UAV to learn its trajectory and the RIS configuration based on its observed transmit rate of the suspicious Tx, eliminating the need for channel state information to and from the RIS. The novel contributions include the consideration of the fixed-wing UAV, and the action and reward designed to capture the mobility constraint of the UAV. Simulations demonstrate that the new approach offers the UAV monitor an exceptional and reliable radio surveillance capability, while keeping a desired distance from the UAV to the Tx. The use of the RIS allows for significant improvements of over 37% and 59% in the eavesdropping success probability and average eavesdropping rate, respectively.

This article presents a new graph-learning technique to accurately infer the graph structure of COVID-19 data, helping to reveal the correlation of pandemic dynamics among different countries and identify influential countries for pandemic response analysis. The new technique estimates the graph Laplacian of the COVID-19 data by first deriving analytically its precise eigenvectors, also known as graph Fourier transform (GFT) basis. Given the eigenvectors, the eigenvalues of the graph Laplacian are readily estimated using convex optimization. With the graph Laplacian, we analyze the confirmed cases of different COVID-19 variants among European countries based on centrality measures and identify a different set of the most influential and representative countries from the current techniques. The accuracy of the new method is validated by repurposing part of COVID-19 data to be the test data and gauging the capability of the method to recover missing test data, showing 33.3% better in root mean squared error (RMSE) and 11.11% better in correlation of determination than existing techniques. The set of identified influential countries by the method is anticipated to be meaningful and contribute to the study of COVID-19 spread.

Joint communication and sensing (JCS) has become a promising technology for mobile networks because of its higher spectrum and energy efficiency. Up to now, the prevalent fast Fourier transform (FFT)-based sensing method for mobile JCS networks is on-grid based, and the grid interval determines the resolution. Because the mobile network usually has limited consecutive OFDM symbols in a downlink (DL) time slot, the sensing accuracy is restricted by the limited resolution, especially for velocity estimation. In this paper, we propose a multiple signal classification (MUSIC)-based JCS system that can achieve higher sensing accuracy for the angle of arrival, range, and velocity estimation, compared with the traditional FFT-based JCS method. We further propose a JCS channel state information (CSI) enhancement method by leveraging the JCS sensing results. Finally, we derive a theoretical lower bound for sensing mean square error (MSE) by using perturbation analysis. Simulation results show that in terms of the sensing MSE performance, the proposed MUSIC-based JCS outperforms the FFT-based one by more than 20 dB. Moreover, the bit error rate (BER) of communication demodulation using the proposed JCS CSI enhancement method is significantly reduced compared with communication using the originally estimated CSI.

This article puts forth an aerial edge Internet of Things (EdgeIoT) system, where an unmanned aerial vehicle (UAV) is employed as a mobile-edge server to process mission-critical computation tasks of ground Internet of Things (IoT) devices. When the UAV schedules an IoT device to offload its computation task, the tasks buffered at the other unselected devices could be outdated and have to be canceled. We investigate a new joint optimization of UAV cruise control and task offloading allocation, which maximizes tasks offloaded to the UAV, subject to the IoT device’s computation capacity and battery budget, and the UAV’s speed limit. Since the optimization contains a large solution space while the instantaneous network states are unknown to the UAV, we propose a new deep-graph-based reinforcement learning framework. An advantage actor–critic (A2C) structure is developed to train the real-time continuous actions of the UAV in terms of the flight speed, heading, and the offloading schedule of the IoT device. By exploring hidden representations resulting from the network feature correlation, our framework takes advantage of graph neural networks (GNNs) to supervise the training of UAV’s actions in A2C. The proposed graph neural network-enabled A2C (GNN-A2C) framework is implemented with Google Tensorflow. The performance analysis shows that GNN-A2C achieves fast convergence and reduces considerably the task missing rate in aerial EdgeIoT.

Being a key technology for beyond fifth-generation wireless systems, joint communication and radar sensing (JCAS) utilizes the reflections of communication signals to detect foreign objects and deliver situational awareness. A cellular-connected unmanned aerial vehicle (UAV) is uniquely suited to form a mobile bistatic synthetic aperture radar (SAR) with its serving base station (BS) to sense over large areas with superb sensing resolutions at no additional requirement of spectrum. This paper designs this novel BS-UAV bistatic SAR platform, and optimizes the flight path of the UAV to minimize its propulsion energy and guarantee the required sensing resolutions on a series of interesting landmarks. A new trajectory planning algorithm is developed to convexify the propulsion energy and resolution requirements by using successive convex approximation and block coordinate descent. Effective trajectories are obtained with a polynomial complexity. Extensive simulations reveal that the proposed trajectory planning algorithm outperforms significantly its alternative that minimizes the flight distance of cellular-aided sensing missions in terms of energy efficiency and effective consumption fluctuation. The energy saving offered by the proposed algorithm can be as significant as 55%.

Typical deep neural network (DNN) backdoor attacks are based on triggers embedded in inputs. Existing imperceptible triggers are computationally expensive or low in attack success. In this paper, we propose a new backdoor trigger, which is easy to generate, imperceptible, and highly effective. The new trigger is a uniformly randomly generated three-dimensional (3D) binary pattern that can be horizontally and/or vertically repeated and mirrored and superposed onto three-channel images for training a backdoored DNN model. Dispersed throughout an image, the new trigger produces weak perturbation to individual pixels, but collectively holds a strong recognizable pattern to train and activate the backdoor of the DNN. We also analytically reveal that the trigger is increasingly effective with the improving resolution of the images. Experiments are conducted using the ResNet-18 and MLP models on the MNIST, CIFAR-10, and BTSR datasets. In terms of imperceptibility, the new trigger outperforms existing triggers, such as BadNets, Trojaned NN, and Hidden Backdoor, by over an order of magnitude. The new trigger achieves an almost 100% attack success rate, only reduces the classification accuracy by less than 0.7%–2.4%, and invalidates the state-of-the-art defense techniques.

The joint communication and sensing (JCS) system can provide higher spectrum efficiency and load saving for 6G machine-type communication (MTC) applications by merging necessary communication and sensing abilities with unified spectrum and transceivers. In order to suppress the mutual interference between the communication and radar-sensing signals to improve the communication reliability and radar-sensing accuracy, we propose a novel code-division orthogonal frequency-division multiplex (CD-OFDM) JCS MTC system, where MTC users can simultaneously and continuously conduct communication and sensing with each other. We propose a novel CD-OFDM JCS signal and corresponding successive-interference-cancelation-based signal processing technique that obtains code-division multiplex gain, which is compatible with the prevalent orthogonal frequency-division multiplex (OFDM) communication system. To model the unified JCS signal transmission and reception process, we propose a novel unified JCS channel model. Finally, the simulation and numerical results are shown to verify the feasibility of the CD-OFDM JCS MTC system and the error propagation performance. We show that the CD-OFDM JCS MTC system can achieve not only more reliable communication but also comparably robust radar sensing compared with the precedent OFDM JCS system, especially in a low signal-to-interference-and-noise ratio regime.

In this paper, we study a joint placement, resource allocation, and user association problem for UAV-assisted wireless networks with constrained backhaul links, where multiple UAV-mounted base stations (UBSs) are deployed to provide wireless services for ground users. We propose a novel framework to maximize the user throughput within the flight-time of UBSs and provides fairness among the users. We first obtain the optimal resource allocation schemes based on different fronthaul and backhaul conditions, and an efficient iterative algorithm is then developed to jointly optimize user association and UBS placement. The optimal UBS placement can be achieved by solving an unconstrained optimization problem which is a simplification of the initial constrained optimization problem based on the optimal resource allocation. We develop a dual-domain coordinated descent and bipartite graph matching based sub-process to identify an optimal user association that prefers the nearby UBSs, as the user association under constrained backhaul links have non-unique optimal solutions. Extensive simulations are conducted to verify the effectiveness of the proposed algorithm, and results show that our proposed method under constrained backhaul can improve both the average throughput by 49% and the fairness among the users by 47% in comparison with the method under ideal backhaul.

Motivated to understand the increasingly severe threat of unmanned aerial vehicles (UAVs) to the confidentiality of terrestrial radio links, this paper analyzes the ergodic and E-outage secrecy capacities of the links in the presence of multiple cooperative aerial eavesdroppers flying autonomously in three-dimensional (3D) spaces and exploiting selection combining (SC) or maximal ratio combining (MRC). The “cut-off” density of the eavesdroppers under which the secrecy capacities vanish is identified. By decoupling the analysis of the random trajectories from the random channel fading, closed-form approximations with almost sure convergence to the secrecy capacities are devised. The analysis is extended to study the impact of the oscillator phase noises and finite memories of the aerial eavesdroppers on the secrecy performance of the ground link. Validated by simulations, the cut-off density only depends on the range of the link in the case of SC eavesdropping, while it depends on the flight region of the eavesdroppers in the case of MRC eavesdropping.

With the development of Next-generation Wireless Networks (NWNs), delay-sensitive traffic triggered by mobile applications (such as video stream and online games) will become an important part of the NWNs. With the increasing demand for massive video content transmission and good quality of users' experience, NWNs have to face up to some serious challenges. As a remedy, efficient routing schemes are capable of achieving load balance. In this article, we propose a load balance routing based on machine learning. First, a dimension-reduced vector matrix can be obtained from the original adjacency matrix of the network topology by Principal Component Analysis (PCA). Then, a neural network is used for the prediction of the network queue status, which can be used as a metric for making intelligent routing decisions. Finally, a load balance routing algorithm considering Queue Utilization (QU) is designed accordingly. Simulation results show the performance of our proposed machine learning-based routing scheme compared to the shortest path algorithm (Bellman-Ford (BF)) and its variant (QUBF) in terms of the packet loss ratio, the throughput and the delay.

Fountain codes are promising for unmanned aerial vehicle (UAV) communications with intermittent transmission links caused by high UAV mobility. However, it is challenging to estimate the transmission delay of UAV communication systems with fountain codes due to the uncertainty of the coding rate and the dynamic channel quality. In this paper, we propose a delay estimation method based on a joint buffer-decoder queuing model for UAV communication systems with LT codes, and show that the complexity of the proposed delay estimation method can be reduced from O(n3) to O(n2). Simulation results validate the effectiveness of the proposed delay estimation method.

This paper studies the threat that an aerial eavesdropper can pose to terrestrial wireless communications, from an information-theoretic point of view. The achievable ergodic and the average ε-outage secrecy rates with no channel state information at the transmitter (i.e., with no CSIT) are analyzed for a transmitter-receiver pair on the ground, in the presence of an aerial eavesdropper which flies a random trajectory following a smooth turn (ST) mobility model in a three-dimensional (3D) space. The ST mobility model induces a uniform distribution (of the eavesdropper's waypoints) within the considered 3D volume. Closed-form asymptotic approximations of the achievable secrecy rates are derived based on the almost sure convergence and non-trivial mathematical manipulations. Validated by simulations, our analysis is tight and reveals that the ground transmission is particularly vulnerable to aerial eavesdropping which can be carried out in a distance without being noticed. 3D spherical regions are identified, within which the secrecy rates vanish. This sheds useful insights to protect terrestrial wireless networks from aerial eavesdropping.

In the Internet of Things (IoT) services, the data of wireless sensor network needs to be collected. However, in the scenarios that have no infrastructure support, the data collection of sensors has great difficulty. Since unmanned aerial vehicle (UAV) has the characteristics of flexibility, it can be applied in the data collection for wireless sensor network (WSN). In this paper, we study UAVs supported data collection for WSN. Firstly, the entire region is divided into multiple cells. Secondly, the flight paths for single UAV and multiple UAVs are designed to cover all cells. The per-node capacity of sensor is derived, which is a function of the number of cells, the height of UAV, the number of sensors and the energy capacity of UAV. It is found that the per- node capacity with multiple UAVs is much larger than that with single UAVs. Then the optimal number of cells is derived to maximize the per-node capacity of WSN. Finally, we provide simulation results to verify our analysis. The discoveries in this paper may provide guideline for the UAVs assisted data collection in WSN.

The distinctive characteristics of unmanned aerial vehicle networks (UAVNs), including highly dynamic network topology, high mobility, and open-air wireless environments, may make UAVNs vulnerable to attacks and threats. In this study, we propose a novel trust model for UAVNs that is based on the behavior and mobility pattern of UAV nodes and the characteristics of inter-UAV channels. The proposed trust model consists of four parts: direct trust section, indirect trust section, integrated trust section, and trust update section. Based on the trust model, the concept of a secure link in UAVNs is formulated that exists only when there is both a physical link and a trust link between two UAVs. Moreover, the metrics of both the physical connectivity probability and the secure connectivity probability between two UAVs are adopted to analyze the connectivity of UAVNs. We derive accurate and analytical expressions of both the physical connectivity probability and the secure connectivity probability using stochastic geometry with or without Doppler shift. Extensive simulations show that compared with the physical connection probability with or without malicious attacks, the proposed trust model can guarantee secure communication and reliable connectivity between UAVs and enhance network performance when UAVNs face malicious attacks and other security risks.

The exploitation of spectrum opportunities in the dimension of height will bring another transmission degree of freedom for wireless networks. Besides, the modern wireless networks are deployed in the three dimensional (3D) space, which need cognitive radio technologies to enhance their performances. With these motivations, the capacity of 3D cognitive radio networks (CRNs) is addressed in this paper. Since there is one additional dimension of interference in 3D CRNs, the network protocols need to be designed to coordinate the interference and guarantee the connectivity of CRNs. Then the link capacity and routing density of 3D CRNs are investigated. Finally, we have derived the per-node capacity of primary network and secondary network respectively. We have verified that the path loss factor α has an impact on the capacity of 3D CRNs, namely, α = 3 is a watershed of capacity scaling laws. Besides, when α > 2.5, the capacity of 3D CRNs is higher than 2D CRNs with the same amount of nodes asymptotically. Therefore our results may provide an insight into the design of 3D cognitive radio networks.

Recent studies suggest nodes in practical networks are more likely to communicate with nearby nodes than far away nodes, which is referred to as proximity preference. In this paper, we model proximity preference by assuming the probability of communication follows a power law distribution with respect to distance and analyze its influence on the throughput of a hybrid network. Moreover, L-maximum-hop routing strategy is adopted to enforce delay constraints. Throughput is derived as a function of maximum hop L, proximity preference index α and the number of base stations (BSs) m. It is also found that per-node throughput changes with α. When 0 ≤ α ≤ 2, proximity has no influence on throughput. When 2 ≤ α ≤ 3, the throughput increases with α. Otherwise, the throughput reaches its maximum and remains constant. Our results demonstrate the interplay of various networks parameters with proximity preference and provide guidelines for the design of practical networks.