Vertical Federated Learning (VFL) is a privacy-preserving distributed learning paradigm where parties with disjoint features collaboratively train a unified Machine Learning (ML) model without leaking private data. This approach effectively addresses medical data fragmentation, allowing organizations to build unified models even when regulations like GDPR or HIPAA prevent raw data sharing. However, VFL faces a critical scalability challenge due to its static architecture. Onboarding new participants typically requires costly system-wide retraining, demanding that all original participants remain online. This bottleneck limits the growth of real-world collaborative ecosystems. To address this, we propose DistillVFL, a novel framework leveraging Knowledge Distillation for efficient client onboarding. Our approach employs a teacher–student paradigm where a pre-trained multi-party “teacher” model transfers knowledge to a new “student” client. This approach allows original clients to remain offline, eliminating re-participation requirements and additional computational costs. We evaluate DistillVFL through extensive experiments on four diverse medical datasets. The results demonstrate that student clients can achieve performance comparable to the comprehensive teacher model while drastically reducing computational overhead. DistillVFL provides a practical, scalable solution to the client onboarding problem, facilitating more dynamic and adaptable VFL collaborations.

Hierarchical edge-cloud computing-aided Internet of Things (IoT) networks offer low-latency and cost-efficient services to a growing number of data-intensive IoT devices. However, optimizing service placement, which involves determining the most suitable locations within a network to deploy various services, is critical to balancing workloads dynamically and ensuring efficient resource utilization. In this paper, we jointly optimize service placement, edge/cloud cooperation, task offloading, and bandwidth allocation to enhance processing efficiency and response times. The main objective is to minimize both the overall end-to-end latency and the system cost, including service deployment and operational costs. The formulated problem belongs to the class of non-convex mixed-integer nonlinear programming, where finding a feasible solution is already challenging. Towards a stable system, we first transform the original problem into a more tractable form and then decompose it into sub-problems which are solved at different timescales. Combining tools from relaxation and the successive convex approximation method, we develop iterative algorithms to solve these problems efficiently. With an appropriate penalty parameter, the proposed algorithms guarantee convergence to at least a local optimum. We produce extensive numerical results to demonstrate the superior performance of the proposed algorithms over benchmark schemes as well as emphasize the significance of the joint service placement and resource allocation in enhancing system performance and efficiency.

The sustainability of low-power mobile sensors is severely challenged by their limited battery capacity, and while simultaneous wireless information and power transfer (SWIPT) is a promising solution, its efficiency suffers dramatically under the uncertainty inherent to mobile three-dimensional (3D) aerial-to-ground environments. This work addresses the critical need for robust and efficient SWIPT under dynamic uncertainty by proposing a novel sensing-assisted SWIPT framework based on a unique hybrid learning algorithm. Our approach first formulates a two-layer optimization problem that rigorously couples a sensing layer, characterized by the Posterior Cram´er-Rao Bound (PCRB), with a SWIPT resource allocation layer. For the sensing layer, the core novelty is a learning-based Kalman Filtering (KF) estimator that merges the interpretative stability of model-based filtering with the adaptive power of neural networks to learn complex, nonlinear mobility patterns. We then prove that minimizing the estimator’s unsupervised loss is mathematically equivalent to minimizing the PCRB, ensuring convergence to optimal sensing without ground-truth supervision. This high-fidelity state information drives a decision-making learning model that adaptively optimizes beamforming, transmit power, and power splitting for the SWIPT resource allocation layer, forming a closed-loop hybrid learning system that continuously reinforces sensing and SWIPT performance. Extensive simulations demonstrate that our framework significantly outperforms benchmark methods in sensing accuracy, communication rate, and energy harvesting, validating its effectiveness in dynamic mobile environments.

Integrated sensing and communication (ISAC) systems promise efficient spectrum utilization by jointly supporting radar sensing and wireless communication. This paper presents a deep learning-driven framework for enhancing physical-layer security in multicarrier ISAC systems under imperfect channel state information (CSI) and in the presence of unknown eaves-dropper (Eve) locations. Unlike conventional ISAC-based friendly jamming (FJ) approaches that require Eve’s CSI or precise angle-of-arrival (AoA) estimates, our method exploits radar echo feedback to guide directional jamming without explicit Eve’s information. To enhance robustness to radar sensing uncertainty, we propose a radar-aware neural network that jointly optimizes beamforming and jamming by integrating a novel nonparametric Fisher Information Matrix (FIM) estimator based on f-divergence. The jamming design satisfies the Cramér–Rao lower bound (CRLB) constraints even in the presence of noisy AoA. For efficient implementation, we introduce a quantized tensor train-based encoder that reduces the model size by more than 100 times with negligible performance loss. We also integrate a non-overlapping secure scheme into the proposed framework, in which specific sub-bands can be dedicated solely to communication. Extensive simulations demonstrate that the proposed solution achieves significant improvements in secrecy rate, reduced block error rate (BLER), and strong robustness against CSI uncertainty and angular estimation errors, under-scoring the effectiveness of the proposed deep learning–driven friendly jamming framework under practical ISAC impairments.

An intelligent Real-Time Sensing (RTS) system must continuously acquire, update, integrate, and apply knowledge to adapt to real-world dynamics. Managing distributed intelligence in this context requires Federated Continual Learning (FCL). However, effectively capturing the diverse characteristics of RTS data in FCL systems poses significant challenges, including severely impacting computational and communication resources, escalating energy costs, and ultimately degrading overall system performance. To overcome these challenges, we investigate how the data distribution shift from ideal to practical RTS scenarios affects Artificial Intelligence (AI) model performance by leveraging thegeneralization gapconcept. In this way, we can analyze how sampling time in RTS correlates with the decline in AI performance, computation cost, and communication efficiency. Based on this observation, we develop a novel Sample-driven Control for Federated Continual Learning (SCFL) technique, specifically designed for mobile edge networks with RTS capabilities. In particular, SCFL is an optimization problem that harnesses the sampling process to concurrently minimize the generalization gap and improve overall accuracy while upholding the energy efficiency of the FCL framework. To solve the highly complex and time-varying optimization problem, we introduce a new soft actor-critic algorithm with explicit and implicit constraints (A2C-EI). Our empirical experiments reveal that we can achieve higher efficiency compared to other DRL baselines. Notably, SCFL can significantly reduce energy consumption up to 85% while maintaining FL convergence and timely data transmission.

Federated Learning (FL) enables devices to collaboratively train machine learning models without sharing raw data, promoting privacy-preserving AI. However, practical deployment faces challenges in balancing the data privacy, communication overhead, and the training convergence rate. For instance, adding noise to local models to preserve privacy can increase the size of updates, exacerbating communication overhead and reducing the convergence rate, while coarse quantization reduces communication costs but can degrade model accuracy. This paper introduces a novel integration of diverse quantization schemes, including both uniform and adaptive quantization, synergistically paired with additive noise mechanisms, to optimally trade off the model/training precision/rate, communication overhead, and privacy protection. By adapting quantization levels based on training dynamics, including gradient variance and model convergence, our approach minimizes the learning error upper bound while ensuring theoretically quantified differential privacy and achieves significant savings in the number of communicated bits. To the best of our knowledge, this is the first work to integrate adaptive quantization with additive noise in FL. More importantly, we provide theoretical guarantees for differential privacy and convergence of the proposed framework and empirically evaluate its communication privacy tradeoffs. Experimental results on popular datasets like MNIST, CIFAR demonstrate that our method enables the training of convolutional neural networks with less than 4-bit quantization, achieving privacy budgets as low as 1.0, while maintaining accuracy that approaches the standard, non-differentially private FedAvg algorithm.

In this paper, we propose an effective blockchain-enabled information management framework, named Smart Blockchain-based Web 3.0 (SBW 3.0). Our framework aims to handle information within Web 3.0 efficiently, enhance data security and privacy, create new revenue streams, and encourage users to contribute valuable information to websites. To this end, SBW 3.0 employs blockchain technology and smart contracts to manage the decentralized data collection in Web 3.0. Moreover, we introduce a robust consensus mechanism grounded in Delegated Proof-of-Stake (DPoS) to reward user contributions. Furthermore, we develop a non-cooperative game model to examine user behavior in this context and conduct thorough analysis to prove the uniqueness of the Nash equilibrium in our proposed system. Through simulations, we evaluate the performance of SBW 3.0 and analyze the effects of various critical parameters on information contribution. Our results validate the theoretical analysis, showing that the proposed consensus mechanism successfully encourages nodes and users to provide more information, thus overcoming the current limitations of Web 3.0 regarding data decentralization and management.

This paper presents a novel Collaborative Cyberattack Detection (CCD) system aimed at enhancing the security of blockchain-based data-sharing networks by addressing the complex challenges associated with noise addition in federated learning models. Leveraging the theoretical principles of differential privacy, our approach strategically integrates noise into trained sub-models before reconstructing the global model through transmission. We systematically explore the effects of various noise types, i.e., Gaussian, Laplace, and Moment Accountant, on key performance metrics, including attack detection accuracy, deep learning model convergence time, and the overall runtime of global model generation. Our findings reveal the intricate trade-offs between ensuring data privacy and maintaining system performance, offering valuable insights into optimizing these parameters for diverse CCD environments. Through extensive simulations, we provide actionable recommendations for achieving an optimal balance between data protection and system efficiency, contributing to the advancement of secure and reliable blockchain networks.

In real-world applications, Federated Learning (FL) faces two challenges: (1) scalability and (2) heterogeneous data. To address the first problem, we design a novel FL framework named Full-stack FL (F2L). More specifically, F2L provides a hierarchical network architecture, making extending the FL network accessible without reconstructing the whole network system. Moreover, leveraging the advantages of hierarchical network design, we propose a new Label-driven Knowledge Distillation (LKD) technique at the centralized server to address the second problem. Unlike the current knowledge distillation techniques, LKD is capable of training a student model, which consists of good knowledge from all teachers' models. Therefore, our proposed algorithm can effectively extract the knowledge of the regions' data distribution (i.e., the regional aggregated models) to reduce the divergence between clients' models when operating under the FL system with non-independent identically distributed data. Extensive experiment results reveal that: (i) our F2L method can significantly improve the overall FL efficiency in all global distillations (i.e., accuracy is$7-20\%$higher in non-IID settings), and (ii) F2L rapidly achieves convergence as global distillation stages occur instead of increasing on each communication cycle.

The surge in traffic has amplified the need for greater reliability and higher throughput in vehicle-to-vehicle (V2V) communications. While network-coded cooperation (NCC) offers a promising solution, its reliance on overheard signals and susceptibility to network coding noise raise concerns about the quality-of-service (QoS) being provided. To address this, we propose a rate-splitting aided multi-user cooperation (RSMUC) framework, which employs rate-splitting multiple access (RSMA) as a relaying scheme and enhances the QoS by removing the dependency on overheard signals while leveraging the inherent benefits of RSMA. The outage probability analysis of the system is performed, where joint cumulative distribution functions (CDFs) are essential for accurately characterizing the outage performance of both common and private streams. To model the interdependence between these streams, we develop a copula-based framework. The results show close alignment between the simulated results and the derived analytical equations. Comparative evaluations against benchmark schemes, like NCC variants and non-orthogonal multiple access (NOMA) based cooperative systems (NCS), show that RSMUC significantly outperforms existing approaches across various performance metrics like outage probability and throughput. The contribution of RSMA was further verified through comparisons of throughput and packet delivery/drop (%) rates, providing key insights that are crucial in setting vehicular cluster sizes. These findings underscore the potential of RSMUC to enhance reliability and deliver improved QoS in next-generation V2V networks.

Integrated sensing and communication (ISAC) is emerging as a key enabler for vehicle-to-everything (V2X) systems. However, designing efficient beamforming schemes for ISAC signals to achieve accurate sensing and enhance communication performance in the dynamic and uncertain environments of V2X networks presents significant challenges. While artificial intelligence technologies offer promising solutions, the energy-intensive nature of neural networks imposes substantial burdens on communication infrastructures. To address these challenges, this work proposes an energy-efficient and intelligent ISAC system for V2X networks. Specifically, we first leverage a Markov Decision Process framework to model the dynamic and uncertain nature of V2X networks. This framework allows the roadside unit to develop beamforming schemes relying solely on its current sensing information, eliminating the need for numerous pilot signals and extensive CSI acquisition. We then introduce an advanced deep reinforcement learning (DRL) algorithm, enabling the joint optimization of beamforming and power allocation to guarantee both communication rate and sensing accuracy in dynamic and uncertain V2X scenario. To alleviate the energy demands of neural networks, we integrate spiking neural networks (SNNs) into the DRL algorithm. The event-driven, sparse spike-based processing of SNNs significantly improves energy efficiency while maintaining strong performance. Extensive simulation results validate the effectiveness of the proposed scheme with lower energy consumption, superior communication performance, and improved sensing accuracy.

Provisioning secrecy for all users, given the heterogeneity in their channel conditions, locations, and the unknown location of the attacker/eavesdropper, is challenging and not always feasible. The problem is even more difficult under finite blocklength constraints that are popular in ultra-reliable low latency communication (URLLC) and massive machine-type communications (mMTC). This work takes the first step to guarantee secrecy for all URLLC/mMTC users in the finite blocklength regime (FBR) where intelligent reflecting surfaces (IRS) are used to enhance legitimate users’ reception and thwart the potential eavesdropper (Eve) from intercepting. To that end, we aim to maximize the minimum secrecy rate (SR) among all users by jointly optimizing the transmitter’s beamforming and IRS’s passive reflective elements (PREs) under the FBR latency constraints. The resulting optimization problem is non-convex. To tackle it, we linearize the objective function, and decompose the problem into sequential subproblems. We prove that our proposed algorithm’s converges to a locally optimal solution with low computational complexity thanks to our closed-form linearization approach. This makes the solution scalable for large IRS deployments. Extensive simulations with practical settings show that our approach can ensure secure communication for all users while satisfying FBR constraints.

The growing size and complexity of Large Language Models (LLMs) pose major challenges for deployment in edge environments with limited computational and communication resources. To address this, we propose a novel distributed framework for allocating LLM layers across heterogeneous edge nodes. By modeling the problem as a two-sided matching game, where layers and nodes rank each other based on processing delay and inter-node transmission latency, our approach achieves stable and low-latency allocations without centralized coordination. In particular, we first design a lightweight matching algorithm that accounts for pipeline dependencies and mitigates idle periods during sequential inference. We then extend the proposed framework to support multi-tenant scenarios, where multiple LLMs compete for shared edge resources. Extensive simulations show that our approach achieves up to a 10% reduction in inference latency compared to the Kolkata Game baseline. We also validate the proposed framework with a real-world testbed based on LLaMA-7B under realistic conditions.

This work proposes an energy-efficient, learning-based beamforming scheme for integrated sensing and communication (ISAC)-enabled V2X networks. Specifically, we first model the dynamic and uncertain nature of V2X environments as a Markov Decision Process. This formulation allows the roadside unit to generate beamforming decisions based solely on current sensing information, thereby eliminating the need for frequent pilot transmissions and extensive channel state information acquisition. We then develop a deep reinforcement learning (DRL) algorithm to jointly optimize beamforming and power allocation, ensuring both communication throughput and sensing accuracy in highly dynamic scenario. To address the high energy demands of conventional learning-based schemes, we embed spiking neural networks (SNNs) into the DRL framework. Leveraging their event-driven and sparsely activated architecture, SNNs significantly enhance energy efficiency while maintaining robust performance. Simulation results confirm that the proposed method achieves substantial energy savings and superior communication performance, demonstrating its potential to support green and sustainable connectivity in future V2X systems.

This paper presents PP-AICloud, a practical system prototype designed to enable privacy-preserving AI inference for cloud-based mobile services. Motivated by recent industry efforts, such as Apple's integration of Homomorphic Encryption (HE) for on-device intelligence, our work addresses the key limitations of existing privacy-preserving machine learning (PPML) solutions, i.e., high latency, bandwidth inefficiencies, and excessive on-device computation. PP-AICloud leverages edge nodes as an intermediate computing layer between mobile devices and centralized AICloud infrastructures, distributing the HE workflow across edge and cloud resources. By integrating HE with deep convolutional neural networks (CNNs), the system enables efficient and secure inference on encrypted user data without requiring decryption. Experimental results demonstrate that PP-AICloud achieves more than 91% accuracy on a real-world landmark recognition task with real-time latency of under 0.7 seconds. We demonstrate the capabilities of PP-AICloud through demo videos available at: Demo link.

Anomaly detection aims to identify patterns and events that deviate from the norm. However, current methods struggle to achieve high detection accuracy due to data complexity, i.e., imbalance and particularly hidden features not directly observed in the raw data. To address this, we propose a novel neural network architecture/model, called Dual-Decoder Variational Auto-Encoder (DDVAE) which consists of an encoder and two decoders. The encoder maps the input data into the extended latent space, where the dimensionality is greater than that of the input data, providing more room to capture the relationship among features. The data in the extended latent space of DDVAE is modelled based on the distribution of the normal samples to generate stochastic latent variables before they are fed into the first decoder to reconstruct the input data. After that, reconstructed data at the output of the first decoder are used to reconstruct the extended latent space, in which anomalies produce higher reconstruction errors than normal samples when reconstructing both the input data and the extended latent space. The Area Under the ROC Curve (AUC) obtained by DDVAE is significantly greater than that of conventional non-parametric methods and generative models on eight benchmark anomaly datasets, by up to 3.9%. DDVAE also achieves an average miss detection rate of 16.9%. We observe that DDVAE achieves high AUC in cases of a low rate of anomalies compared to normal samples,

This paper introduces a novel joint user association and resource allocation framework to efficiently deal with eavesdropping attacks without requiring prior information about eavesdroppers. Specifically, the co-channel interference when reusing resource blocks is leveraged to disrupt the signal reception at eavesdroppers. To maximize the secure area, defined as the area where eavesdroppers cannot wiretap the channel due to co-channel interference, we first formulate the system by using the Markov decision process to capture the dynamics and uncertainty of mobile users and wireless communications. Then, a deep reinforcement learning (DRL)-based approach is proposed to obtain the joint optimal user association and resource allocation policy to utilize the co-channel interference and maximize the secure area. Extensive simulation results demonstrate that by intelligently associating users and allocating resource blocks to them, our proposed solution can help to effectively defeat eavesdropping attacks without requiring prior information of eavesdroppers which may not be readily available in practice. In addition, the proposed DRL-based algorithm can converge to the optimal policy quickly and achieve better performance compared to existing solutions.

Satellite communication (SatCom) systems play a vital role in providing global connectivity and enable a wide range of applications, including Internet of Things (IoT) connectivity for remote areas, such as forests and oceans. Two crucial resource allocation challenges in SatCom are beam placement (BP) and frequency assignment (FA) problems, which involve the clique covering (CC) and graph coloring (GC) problems, respectively. Conventional solutions for these problems incur excessive computational cost, which is intractable for classical computers. A promising approach is to formulate these problems using the Ising model, construct their Hamiltonians, and then solve them efficiently by a quantum computer. However, the current quantum computers have very limited hardware and can only handle rather small inputs. To overcome this limitation, we propose a hybrid-quantum-classical-computational pipeline where an efficient hamiltonian reduction method is the key for solving large CC/GC instances. Through experiments on real quantum computers, our reduction method outperforms commercial solutions, allowing quantum annealers to handle significantly larger BP/FA instances while maintaining high probability to achieve feasible solutions and near-optimal performance. Although the inherent hardness of the CC/GC problems cannot be overcome by quantum computing, our research contributes to the early exploration of quantum computing in the context of the complex optimization problems in SatCom systems, particularly in the realm of IoT connectivity for remote areas.

Digital Twin (DT) technology is transforming biological processes by enabling real-time predictive modeling and optimization. However, implementing DTs at the micro- and nano-scale presents significant challenges in data extraction, transmission, and computation due to the complexity of biological environments. To address these challenges, this work proposes a multi-model Physics-Informed Neural Network (PINN) framework, enhanced by the Internet of Bio-Nano Things (IoBNT), to improve real-time bioprocess monitoring and prediction. The proposed framework integrates IoBNT for efficient biological data acquisition and transmission with advanced deep neural networks (DNNs), including Fully Connected Neural Networks (FCNN), Residual Block Neural Networks (ResBlock NN), and Recurrent Neural Networks (RNN), to enhance predictive accuracy. By incorporating physics-based constraints into the learning process, the PINN architecture ensures biologically plausible and data-efficient predictions. Experimental results demonstrate the framework’s effectiveness in monitoring microbial and substrate growth using Monod equations, achieving an average mean absolute error (MAE) of 0.0275 g/L, anR2value close to 1, and a root mean squared error (RMSE) of 0.0425 g/L across four of the five proposed architectures. Additionally, the integration of IoBNT enhances data reliability, reducing transmission errors by up to 98%. These findings highlight the potential of IoBNT-powered PINN frameworks for developing high-fidelity, self-adaptive DTs in digital biological and healthcare applications.

This work proposes a novel privacy-preserving cyberattack detection framework for blockchain-based Internet of Things (IoT) systems. In our approach, artificial intelligence (AI)-driven detection modules are strategically deployed at blockchain nodes (BNs) to identify real-time attacks, ensuring high accuracy and minimal delay. To achieve this efficiency, the model training is conducted by a cloud service provider (CSP). Accordingly, BNs send their data to the CSP for training, but to safeguard privacy, the data is encrypted using homomorphic encryption (HE) before transmission. This encryption method allows the CSP to perform computations directly on encrypted data without the need for decryption, preserving data privacy throughout the learning process. To handle the substantial volume of encrypted data, we introduce an innovative packing algorithm in a single-instruction-multiple-data (SIMD) manner, enabling efficient training on HE-encrypted data. Building on this, we develop a novel deep neural network training algorithm optimized for encrypted data. We further propose a privacy-preserving distributed learning approach based on the FedAvg algorithm, which parallelizes the training across multiple workers, significantly improving computation time. Upon completion, the CSP distributes the trained model to the BNs, enabling them to perform real-time, privacy-preserved detection. Our simulation results demonstrate that our proposed method can not only mitigate the training time but also achieve detection accuracy that is approximately identical to the approach without encryption, with a gap of around 0.01%. Additionally, our real implementations on various blockchain consensus algorithms and hardware configurations show that our proposed framework can also be effectively adapted to real-world systems.

Wireless communications are particularly vulnerable to eavesdropping attacks due to their broadcast nature. To effectively deal with eavesdroppers, existing security techniques usually require accurate channel state information (CSI), e.g., for friendly jamming (FJ), and/or additional computing resources at transceivers, e.g., cryptography-based solutions, which unfortunately may not be feasible in practice. This challenge is even more acute in low-end IoT devices. We thus introduce a novel deep learning-based FJ framework that can effectively defeat eavesdropping attacks with imperfect CSI and even without CSI of legitimate channels. In particular, we first develop an autoencoder-based communication architecture with FJ, namely, AEFJ, to jointly maximize the secrecy rate and minimize the block error rate (BLER) at the receiver without requiring perfect CSI of the legitimate channels. In addition, to deal with the case without CSI, we leverage the mutual information neural estimation (MINE) concept and design a MINE-based FJ scheme that can achieve comparable security performance to the conventional FJ methods that require perfect CSI. Extensive simulations in a multiple-input-multiple-output (MIMO) system demonstrate that our proposed solution can effectively deal with eavesdropping attacks in various settings. Moreover, the proposed framework can seamlessly integrate MIMO security and detection tasks into a unified end-to-end learning process. This integrated approach can significantly maximize the throughput and minimize the BLER, offering a good solution for enhancing communication security in wireless communication systems.

With the massive deployment of Internet of Things (IoT) devices in sixth-generation networks, several critical challenges have emerged, such as large communication overhead, coverage limitations, and limited battery lifespan due to high energy consumption. Federated learning (FL), wireless power transfer (WPT), multi-antenna access point (AP), and reconfigurable intelligent surfaces (RIS) can mitigate these challenges by reducing the need for large data transmissions, enabling sustainable energy harvesting, and optimizing the propagation environment. Compared to conventional RIS, simultaneously transmitting and reflecting (STAR)-RIS not only extends coverage from half-space to full-space but also improves energy saving through appropriate mode selection. Motivated by the need for sustainable, low-latency, and energy-efficient communication in large-scale IoT networks, this paper investigates the efficient STAR-RIS mode in the uplink and downlink phases of a WPT-FL multi-antenna AP network with non-orthogonal multiple access to minimize energy consumption, a joint optimization that remains largely unexplored in existing works on RIS or STAR-RIS. We formulate a non-convex energy minimization problem for different STAR-RIS modes, i.e., energy splitting (ES) and time switching (TS), in both uplink and downlink transmission phases, where STAR-RIS phase shift vectors, beamforming matrices, time and power for harvesting, uplink transmission, and downlink transmission, local processing time, and computation frequency for each user are jointly optimized. To tackle the non-convexity, the problem is decoupled into two subproblems: the first subproblem optimizes STAR-RIS phase shift vectors and beamforming matrices across all WPT-FL phases using block coordinate descent over either semi-definite programming or Rayleigh quotient problems, while the second one allocates time, power, and computation frequency via the one-dimensional search algorithms or the bisection algorithm. Simulation results demonstrate that TS STAR-RIS in both uplink and downlink transmissions achieves the lowest energy consumption, outperforming ES and conventional RIS schemes due to its flexible phase shift adaptation and lower interference levels.

Jointly optimizing power allocation and device association is crucial in Internet-of-Things (IoT) networks to ensure devices achieve their data throughput requirements. Device association, which assigns IoT devices to specific access points (APs), critically impacts resource allocation. Many existing works often assume all data throughput requirements are satisfied, which is impractical given resource limitations and diverse demands. When requirements cannot be met, the system becomes infeasible, causing congestion and degraded performance. To address this problem, we propose a novel framework to enhance IoT system robustness by solving two problems, comprising maximizing the number of satisfied IoT devices and jointly maximizing both the number of satisfied devices and total network throughput. These objectives often conflict under infeasible circumstances, necessitating a careful balance. We thus propose a modified branch-and-bound (BB)-based method to solve the first problem. An iterative algorithm is proposed for the second problem that gradually increases the number of satisfied IoT devices and improves the total network throughput. We employ a logarithmic approximation for a lower bound on data throughput and design a fixed-point algorithm for power allocation, followed by a coalition game-based method for device association. Numerical results demonstrate the efficiency of the proposed algorithm, serving fewer devices than the BB-based method but with faster running time and higher total throughput.

In this paper, we consider an aerial reconfigurable intelligent surface (ARIS)-assisted wireless network, where multiple unmanned aerial vehicles (UAVs) collect data from ground users (GUs) by using the non-orthogonal multiple access (NOMA) method. The ARIS provides enhanced channel controllability to improve the NOMA transmissions and reduce the co-channel interference among UAVs. We also propose a novel dual-mode switching scheme, where each UAV equipped with both an ARIS and a radio frequency (RF) transceiver can adaptively perform passive reflection or active transmission. We aim to maximize the overall network throughput by jointly optimizing the UAVs’ trajectory planning and operating modes, the ARIS’s passive beamforming, and the GUs’ transmission control strategies. We propose an optimization-driven hierarchical deep reinforcement learning (O-HDRL) method to decompose it into a series of subproblems. Specifically, the multi-agent deep deterministic policy gradient (MADDPG) adjusts the UAVs’ trajectory planning and mode switching strategies, while the passive beamforming and transmission control strategies are tackled by the optimization methods. Numerical results reveal that the O-HDRL efficiently improves the learning stability and reward performance compared to the benchmark methods. Meanwhile, the dual-mode switching scheme is verified to achieve a higher throughput performance compared to the fixed ARIS scheme.

Intelligent reflective surfaces (IRS) allow us to actively customize the radio environment by manipulating the reflected signals upon them. Among their various applications, one notable use is enhancing user security and privacy. This is achieved by strategically reflecting signals from the transmitter to improve reception for authorized users while minimizing the signal quality for suspicious eavesdroppers. However, under multiuser settings, given the heterogeneity in users’ channels, locations, and the unknown location of the eavesdropper, it is challenging to avoid secrecy outage for all users. This paper takes the first step in investigating the multi-user secrecy rate (SR) maximization in IRS-aided systems. To this end, we aim to maximize the minimum user’s SR by optimizing the transmitter’s beamforming vector and the IRS’ passive reflective elements (PREs). The resulting problem is non-convex. To tackle this, we first linearize the objective function and decompose the problem into two sequential optimization problems. We then design an alternating optimization (AO) method to jointly optimize the transmitter’s beamforming vector and the IRS’ PREs. We prove that the proposed algorithm converges to a locally optimal solution of the above non-convex optimization problem. Numerical results demonstrate that the proposed Max-Min algorithm can provide secure communication for all the users.

In this work, we propose a novel machine learning empowered intrusion detection for Mobile Edge Computing (MEC) networks. Unlike most of the research works that focus on detecting attacks at the network layer, such as IP spoofing and Denial of Service (DoS) attacks, we aim to detect attacks/threats at the application layer, especially attacks caused by malicious codes embedded in offloaded computing tasks. This is an emerging issue in MEC networks as more and more MEC services allow MEC users to offload their computational tasks to the edge nodes to process. Yet, this is a very challenging problem in MEC, as data at the application layer is often complex and challenging to interpret, making anomaly detection difficult. Therefore, we first propose an effective solution to transfer data from the original offloading file to a new form, i.e., images, to make it more effective for the detection process. After that, a Convolutional Neural Network (CNN) and a collaborative learning process are proposed to learn information from training data (i.e., transformed images) and, at the same time, share the learned knowledge (i.e., trained models) together to improve the global accuracy in detecting attacks. Simulation results show that our approach can detect attacks with an accuracy of approximately 90%.

Outlier/anomaly detection plays a pivotal role in AI applications, e.g., in classification, and intrusion/threat detection in cybersecurity. However, most existing methods face challenges of heterogeneity amongst feature subsets posed by non-independent and identically distributed (non-IID) data. To address this, we propose a novel neural network model called Multiple-Input Auto-Encoder for Anomaly Detection (MIAEAD). MIAEAD assigns an anomaly score to each feature subset of a data sample to indicate its likelihood of being an anomaly. This is done by using the reconstruction error of its sub-encoder as the anomaly score. All sub-encoders are then simultaneously trained using unsupervised learning to determine the anomaly scores of feature subsets. The final Area Under the ROC Curve (AUC) of MIAEAD is determined by the maximum value among the feature subsets. Extensive experiments on eight real-world anomaly datasets from different domains, e.g., health, finance, cybersecurity, and satellite imaging, demonstrate the superior performance of MIAEAD over conventional methods and the state-of-the-art unsupervised models, by up to 4.3% in terms of AUC score. Furthermore, experimental results show that the AUC obtained by MIAEAD is mostly not impacted as the ratio of anomalies to normal samples in the dataset increases. In contrast, the fundamental anomaly detection method using Auto-Encoder (AE) experiences a significant decline in AUC as this ratio increases. We observe that MIAEAD has a high AUC when applied to feature subsets with low heterogeneity based on the coefficient of variation (CV) score. We also prove that the MIAEAD model uses fewer parameters than the AE model.

Web 3.0 is the third generation of the World Wide Web (WWW), concentrating on the critical concepts of decentralization, availability, and increasing client usability. Although Web 3.0 is undoubtedly an essential component of the future Internet, it currently faces critical challenges, including decentralized data collection and management. To overcome these challenges, blockchain has emerged as one of the core technologies for the future development of Web 3.0. In this paper, we propose a novel blockchain-based information management framework, namely Smart Blockchain-based Web (SBW), to manage information in Web 3.0 effectively, enhance the security and privacy of users’ data, bring additional profits, and incentivize users to contribute information to the websites. Particularly, SBW utilizes blockchain technology and smart contracts to manage the decentralized data collection process for Web 3.0 effectively. Moreover, in this framework, we develop an effective consensus mechanism based on Proof-of-Stake (PoS) to reward the user’s information contribution and conduct game theoretical analysis to analyze the user’s behavior in the considered system. Additionally, we conduct simulations to assess the performance of SBW and investigate the impact of critical parameters on information contribution. The findings confirm our theoretical analysis and demonstrate that our proposed consensus mechanism can incentivize the nodes and users to contribute more information to our systems.

The ability to estimate 3D movements of users over edge computing-enabled networks, such as 5G/6G networks, is a key enabler for the new era of extended reality (XR) and Metaverse applications. Recent advancements in deep learning have shown advantages over optimization techniques for estimating 3D human poses given spare measurements from sensor signals, i.e., inertial measurement unit (IMU) sensors attached to the XR devices. However, the existing works lack applicability to wireless systems, where transmitting the IMU signals over noisy wireless networks poses significant challenges. Furthermore, the potential redundancy of the IMU signals has not been considered, resulting in highly redundant transmissions. In this work, we propose a novel approach for redundancy removal and lightweight transmission of IMU signals over noisy wireless environments. Our approach utilizes a random Gaussian matrix to transform the original signal into a lower-dimensional space. By leveraging the compressive sensing theory, we have proved that the designed Gaussian matrix can project the signal into a lower-dimensional space and preserve the Set-Restricted Eigenvalue condition, subject to a power transmission constraint. Furthermore, we develop a deep generative model at the receiver to recover the original IMU signals from noisy compressed data, thus enabling the creation of 3D human body movements at the receiver for XR and Metaverse applications. Simulation results on a real-world IMU dataset show that our framework can achieve highly accurate 3D human poses of the user using only 82% of the measurements from the original signals. This is comparable to an optimization-based approach, i.e., Lasso, but is an order of magnitude faster.

Machine learning is a core component of many Intrusion Detection Systems (IDS) for IoT networks. However, developing a robust machine-learning model for IDSs on IoT networks is very challenging. This is due to the diversity of IoT devices resulting in the inconsistency and high dimensions of data collected from IoT networks. In other words, in IoT environments, the training data for IDSs on IoT networks is often heterogeneous since they are collected from multiple sources with different characteristics. To tackle these problems, this paper proposes a novel neural network architecture called Multiple-Input Auto-Encoder (MIAE). MIAE has multiple sub-encoders that can process multiple input sources with different dimensions. Moreover, the MIAE model is trained in an unsupervised learning mode, and it can transfer the heterogeneous inputs into lower-dimensional representation to facilitate classifiers to distinguish between the normal samples and types of attacks. The experimental results on the three most popular benchmark IDS datasets, i.e., NSLKDD, UNSW-NB15, and IDS2017, show the superior performance of the MIAE over three groups of methods including conventional classifiers, state-of-the-art dimensionality reduction models, and unsupervised representation learning methods for multiple inputs with different dimensions. MIAE combined with the Random Forest (RF) classifier also achieves 96.2% in terms of accuracy in detecting sophisticated attacks, e.g., Slowloris. In addition, the average running time for detecting an attack sample obtained by MIAE combined with RF classifier is only roughly 5E-7 seconds, whilst the model size is lower than 1 MB. This clearly shows the effectiveness of our proposed model when deployed in practice.

This paper aims to propose a novel framework to address the data privacy issue for Federated Learning (FL)-based Intrusion Detection Systems (IDSs) in Internet-of-Vehicles (IoVs) with limited computational resources. In particular, in conventional FL systems, it is usually assumed that the computing nodes have sufficient computational resources to process the training tasks. However, in practical IoV systems, vehicles usually have limited computational resources to process intensive training tasks, compromising the effectiveness of deploying FL in IDSs. While offloading data from vehicles to the cloud can mitigate this issue, it introduces significant privacy concerns for vehicle users (VUs). To resolve this issue, we first propose a highly-effective framework using homomorphic encryption to secure data that requires offloading to a centralized server for processing. Furthermore, we develop an effective training algorithm tailored to handle the challenges of FL-based systems with encrypted data. This algorithm allows the centralized server to directly compute on quantum-secure encrypted ciphertexts without needing decryption. This approach not only safeguards data privacy during the offloading process from VUs to the centralized server but also enhances the efficiency of utilizing FL for IDSs in IoV systems. Our simulation results show that our proposed approach can achieve a performance that is as close to that of the solution without encryption, with a gap of less than 0.8%.

Federated Learning (FL) offers a privacy-preserving approach to training machine learning models from distributed data on resource-constrained devices. However, even for modern/high-end cars vehicular networks, onboard training with the whole of the raw data presents challenges due to limited computing capability and power consumption concerns. To address this conundrum, we propose a novel FL framework that leverages homomorphic encyption (HE) to allow vehicles to upload encrypted portions of their data to a cloud server. Thanks to the key feature of HE, the server can perform additional model updates directly on the encrypted data, alleviating the workload on vehicles while preserving privacy. Furthermore, model updates from vehicles are also HE-encrypted, guaranteeing end-to-end privacy protection. This approach reduces the computational burden on vehicles while maintaining the model quality and convergence performance of the FL framework. Additionally, it can mitigate biases stemming from heterogeneous data, resulting in more stable FL convergence. Extensive experiments demonstrate the effectiveness of our framework in reducing workload and improving learning stability in vehicular networks.

With the ever-increasing popularity of blockchain applications, securing blockchain networks plays a critical role in these cyber systems. In this paper, we first study cyberattacks (e.g., flooding of transactions, brute pass) in blockchain networks and then propose an efficient collaborative cyberattack detection model to protect blockchain networks. Specifically, we deploy a blockchain network in our laboratory to build a new dataset including both normal and attack traffic data. The main aim of this dataset is to generate actual attack data from different nodes in the blockchain network that can be used to train and test blockchain attack detection models. We then propose a realtime collaborative learning model that enables nodes in the network to share learning knowledge without disclosing their private data, thereby significantly enhancing system performance for the whole network. The extensive simulation and realtime experimental results show that our proposed detection model can detect attacks in the blockchain network with an accuracy of up to 97%.

Intrusion detection systems (IDSs) play a critical role in protecting billions of IoT devices from malicious attacks. However, the IDSs for IoT devices face inherent challenges of IoT systems, including the heterogeneity of IoT data/devices, the high dimensionality of training data, and the imbalanced data. Moreover, the deployment of IDSs on IoT systems is challenging, and sometimes impossible, due to the limited resources, such as memory/storage and computing capability of typical IoT devices. To tackle these challenges, this article proposes a novel deep neural network/architecture called constrained twin variational auto-encoder (CTVAE) that can feed classifiers of IDSs with more separable/distinguishable and lower dimensional representation data. Additionally, in comparison to the state-of-the-art neural networks used in IDSs, CTVAE requires less memory/storage and computing power, hence making it more suitable for IoT IDS systems. Extensive experiments with the 11 most popular IoT botnet data sets show that CTVAE can boost around 1% in terms of accuracy and Fscore in detection attack compared to the state-of-the-art machine learning and representation learning methods, whilst the running time for attack detection is lower than$2E{\mathrm{ -}}6$s and the model size is lower than 1 MB. We also further investigate various characteristics of CTVAE in the latent space and in the reconstruction representation to demonstrate its efficacy compared with current well-known methods.

Federated learning (FL) has found many successes in wireless communications; however, the implementation of FL has been hindered by the energy limitation of mobile devices (MDs) and the availability of training data at MDs. Wireless power transfer (WPT) and mobile crowdsensing (MCS) are promising technologies that can be leveraged to power energy-limited MDs and acquire data for learning tasks. How to integrate WPT and MCS towards sustainable FL solutions is a research topic entirely missing from the open literature. This work for the first time investigates a resource allocation problem in collaborative sensing-assisted sustainable FL (S2FL) networks with the goal of minimizing the total completion time. In particular, we investigate a practical harvesting-sensing-training-transmitting protocol in which energy-limited MDs first harvest energy from RF signals, use it to gain a reward for user participation, sense the training data from the environment, train the local models at MDs, and transmit the model updates to the edge server. The total completion time minimization problem of jointly optimizing power transfer, transmit power allocation, data sensing, bandwidth allocation, local model training, and data transmission is complicated due to the non-convex objective function, highly non-convex constraints, and strongly coupled variables. In order to solve that problem, we apply the decomposition technique and develop a computationally-efficient path-following algorithm to obtain the solution. In particular, inner convex approximations are developed for the resource allocation subproblem, and the subproblems are performed alternatively in an iterative fashion. Simulation results are provided to evaluate the effectiveness of the proposed S2FL algorithm in reducing the completion time up to 21.45% in comparison with other benchmark schemes. Further, we investigate an extension of our work from frequency division multiple access (FDMA) to non-orthogonal multiple access (NOMA) and show that NOMA can speed up the total completion time 8.36% on average of the considered FL system.

In the process industry, long-term and efficient optimization of production lines requires real-time monitoring and analysis of operational states to fine-tune production line parameters. However, complexity in operational logic and intricate coupling of production process parameters make it difficult to develop an accurate mathematical model for the entire process, thus hindering the deployment of efficient optimization mechanisms. In view of these difficulties, we propose to deploy a digital twin (DT) of the production line by encoding its operational logic in a data-driven approach. By iteratively mapping the real-world data reflecting equipment operation status and product quality indicators in the DT, we adopt a quality prediction model for production process based on self-attention-enabled temporal convolutional neural networks (NNs). This model enables the data-driven state evolution of the DT. The DT takes a role of aggregating the information of actual operating conditions and the results of quality-sensitive analysis, which facilitates the optimization of process production with virtual-reality evolution. Leveraging the DT as an information-flow carrier, we extract temporal features from key process indicators and establish a production process quality prediction model based on the proposed deep NN. Our operation experiments on a specific tobacco shredding line demonstrate that the proposed DT-based production process optimization method fosters seamless integration between virtual and real production lines. This integration achieves an average operating status prediction accuracy of over 98% and a product quality acceptance rate of over 96%.

The ability to sense, localize, and estimate the 3D position and orientation of the human body is critical in virtual reality (VR) and extended reality (XR) applications. This becomes more important and challenging with the deployment of VR/XR applications over the next generation of wireless systems such as 5G and beyond. In this paper, we propose a novel framework that can reconstruct the 3D human body pose of the user given sparse measurements from Inertial Measurement Unit (IMU) sensors over a noisy wireless environment. Specifically, our framework enables reliable transmission of compressed IMU signals through noisy wireless channels and effective recovery of such signals at the receiver, e.g., an edge server. This task is very challenging due to the constraints of transmit power, recovery accuracy, and recovery latency. To address these challenges, we first develop a deep generative model at the receiver to recover the data from linear measurements of IMU signals. The linear measurements of the IMU signals are obtained by a linear projection with a measurement matrix based on the compressive sensing theory. The key to the success of our framework lies in the novel design of the measurement matrix at the transmitter, which can not only satisfy power constraints for the IMU devices but also obtain a highly accurate recovery for the IMU signals at the receiver. This can be achieved by extending the set-restricted eigenvalue condition of the measurement matrix and combining it with an upper bound for the power transmission constraint. Our framework can achieve robust performance for recovering 3D human poses from noisy compressed IMU signals. Additionally, our pre-trained deep generative model achieves signal reconstruction accuracy comparable to an optimization-based approach, i.e., Lasso, but is an order of magnitude faster.

To enable an intelligent, programmable and multi-vendor radio access network (RAN) for 6G networks, considerable efforts have been made in standardization and development of open RAN (O-RAN). So far, however, the applicability of O-RAN in controlling and optimizing RAN functions has not been widely investigated. In this paper, we jointly optimize the flow-split distribution, congestion control and scheduling (JFCS) to enable an intelligent traffic steering application in O-RAN. Combining tools from network utility maximization and stochastic optimization, we introduce a multi-layer optimization framework that provides fast convergence, long-term utility-optimality and significant delay reduction compared to the state-of-the-art and baseline RAN approaches. Our main contributions are three-fold:$i$) we propose the novel JFCS framework to efficiently and adaptively direct traffic to appropriate radio units;$ii$) we develop low-complexity algorithms based on the reinforcement learning, inner approximation and bisection search methods to effectively solve the JFCS problem in different time scales; and$iii$) the rigorous theoretical performance results are analyzed to show that there exists a scaling factor to improve the tradeoff between delay and utility-optimization. Collectively, the insights in this work will open the door towards fully automated networks with enhanced control and flexibility. Numerical results are provided to demonstrate the effectiveness of the proposed algorithms in terms of the convergence rate, long-term utility-optimality and delay reduction.

Creating and maintaining the Metaverse requires enormous resources that have never been seen before, especially computing resources for intensive data processing to support the Extended Reality, enormous storage resources, and massive networking resources for maintaining ultra high-speed and low-latency connections. Therefore, this work aims to propose a novel framework, namely MetaSlicing, that can provide a highly effective and comprehensive solution in managing and allocating different types of resources for Metaverse applications. In particular, by observing that Metaverse applications may have common functions, we first propose grouping applications into clusters, called MetaInstances. In a MetaInstance, common functions can be shared among applications. As such, the same resources can be used by multiple applications simultaneously, thereby enhancing resource utilization dramatically. To address the real-time characteristic and resource demand's dynamic and uncertainty in the Metaverse, we develop an effective framework based on the semi-Markov decision process and propose an intelligent admission control algorithm that can maximize resource utilization and enhance the Quality-of-Service for end-users. Extensive simulation results show that our proposed solution outperforms the Greedy-based policies by up to 80% and 47% in terms of long-term revenue for Metaverse providers and request acceptance probability, respectively.

Due to its security, transparency, and flexibility in verifying virtual assets, blockchain has been identified as one of the key technologies for Metaverse. Unfortunately, blockchain-based Metaverse faces serious challenges such as massive resource demands, scalability, and security/privacy concerns. To address these issues, this paper proposes a novel sharding-based blockchain framework, namely MetaShard, for Metaverse applications. Particularly, we first develop an effective consensus mechanism, namely Proof-of-Engagement, that can incentivize MUs' data and computing resource contribution. Moreover, to improve the scalability of MetaShard, we propose an innovative sharding management scheme to maximize the network's throughput while protecting the shards from 51% attacks. Since the optimization problem is NP-complete, we develop a hybrid approach that decomposes the problem (using the binary search method) into sub-problems that can be solved effectively by the Lagrangian method. As a result, the proposed approach can obtain solutions in polynomial time, thereby enabling flexible shard reconfiguration and reducing the risk of corruption from the adversary. Extensive numerical experiments show that, compared to the state-of-the-art commercial solvers, our proposed approach can achieve up to 66.6% higher throughput in less than 1/30 running time. Moreover, the proposed approach can achieve global optimal solutions in most experiments.

By executing offloaded tasks from mobile users, edge computing augments mobile devices with computing/communications resources from edge nodes (ENs), thus enabling new services/applications (e.g., real-time gaming, virtual/augmented reality). However, despite being more resourceful than mobile devices, allocating ENs' computing/communications resources to a given favorable set of users (e.g., closer to edge nodes) may block other devices from their services. This is often the case for most existing task offloading and resource allocation approaches that only aim to maximize the network social welfare or minimize the total energy consumption but do not consider the computing/battery status of each mobile device. This work develops an energy-based proportionally fair task offloading and resource allocation framework for a multi-layer cooperative edge computing network to serve all user equipments (UEs) while considering both their service requirements and individual energy/battery levels. The resulting optimization involves both binary (offloading decisions) and continuous (resource allocation) variables. To tackle the NP-hard mixed integer optimization problem, we leverage the fact that the relaxed problem is convex and propose a distributed algorithm, namely the dynamic branch-and-bound Benders decomposition (DBBD). DBBD decomposes the original problem into a master problem (MP) for the offloading decisions and multiple subproblems (SPs) for resource allocation. To quickly eliminate inefficient offloading solutions, the MP is integrated with powerful Benders cuts exploiting the ENs' resource constraints. We then develop a dynamic branch-and-bound algorithm (DBB) to efficiently solve the MP considering the load balance among ENs. The SPs can either be solved for their closed-form solutions or be solved in parallel at ENs, thus reducing the complexity. The numerical results show that the DBBD returns the optimal solution in maximizing the proportional fairness among UEs. The DBBD has higher fairness indexes, i.e., Jain's index and min-max ratio, in comparison with the existing ones that minimize the total consumed energy.

With the rising demand for wireless services and increased awareness of the need for data protection, existing network traffic analysis and management architectures are facing unprecedented challenges in classifying and synthesizing the increasingly diverse services and applications. This paper proposes FS-GAN, a federated self-supervised learning framework to support automatic traffic analysis and synthesis over a large number of heterogeneous datasets. FS-GAN is composed of multiple distributed Generative Adversarial Networks (GANs), with a set of generators, each being designed to generate synthesized data samples following the distribution of an individual service traffic, and each discriminator being trained to differentiate the synthesized data samples and the real data samples of a local dataset. A federated learning-based framework is adopted to coordinate local model training processes of different GANs across different datasets. FS-GAN can classify data of unknown types of service and create synthetic samples that capture the traffic distribution of the unknown types. We prove that FS-GAN can minimize the Jensen-Shannon Divergence (JSD) between the distribution of real data across all the datasets and that of the synthesized data samples. FS-GAN also maximizes the JSD among the distributions of data samples created by different generators, resulting in each generator producing synthetic data samples that follow the same distribution as one particular service type. Extensive simulation results show that the classification accuracy of FS-GAN achieves over$20\%$improvement in average compared to the state-of-the-art clustering-based traffic analysis algorithms. FS-GAN also has the capability to synthesize highly complex mixtures of traffic types without requiring any human-labeled data samples.

Real-time machine learning (ML) has recently attracted significant interest due to its potential to support instantaneous learning, adaptation, and decision making in a wide range of application domains, including self-driving vehicles, intelligent transportation, and industry automation. In this paper, we investigate real-time ML in a federated edge intelligence (FEI) system, an edge computing system that implements federated learning (FL) solutions based on data samples collected and uploaded from decentralized data networks, e.g., Internet-of-Things (IoT) and/or wireless sensor networks. FEI systems often exhibit heterogenous communication and computational resource distribution, as well as non-i.i.d. data samples arrived at different edge servers, resulting in long model training time and inefficient resource utilization. Motivated by this fact, we propose a time-sensitive federated learning (TS-FL) framework to minimize the overall run-time for collaboratively training a shared ML model with desirable accuracy. Training acceleration solutions for both TS-FL with synchronous coordination (TS-FL-SC) and asynchronous coordination (TS-FL-ASC) are investigated. To address the straggler effect in TS-FL-SC, we develop an analytical solution to characterize the impact of selecting different subsets of edge servers on the overall model training time. A server dropping-based solution is proposed to allow some slow-performance edge servers to be removed from participating in the model training if their impact on the resulting model accuracy is limited. A joint optimization algorithm is proposed to minimize the overall time consumption of model training by selecting participating edge servers, the local epoch number (the number of model training iterations per coordination), and the data batch size (the number of data samples for each model training iteration). Motivated by the fact that data samples at the slowest edge server may exhibit special characteristics that cannot be removed from model training, we develop an analytical expression to characterize the impact of both staleness effect of asynchronous coordination and straggler effect of FL on the time consumption of TS-FL-ASC. We propose a load forwarding-based solution that allows a slow edge server to offload part of its training samples to trusted edge servers with higher processing capability. We develop a hardware prototype to evaluate the model training time of a heterogeneous FEI system. Experimental results show that our proposed TS-FL-SC and TS-FL-ASC can provide up to 63% and 28% of reduction, in the overall model training time, respectively, compared with traditional FL solutions.

Federated Learning (FL) plays a pivotal role in enabling artificial intelligence (AI)-based mobile applications in mobile edge computing (MEC). However, due to the resource heterogeneity among participating mobile users (MUs), delayed updates from slow MUs may deteriorate the learning speed of the MEC-based FL system, commonly referred to as the straggling problem. To tackle the problem, this work proposes a novel privacy-preserving FL framework that utilizes homomorphic encryption (HE) based solutions to enable MUs, particularly resource-constrained MUs, to securely offload part of their training tasks to the cloud server (CS) and mobile edge nodes (MENs). Our framework first develops an efficient method for packing batches of training data into HE ciphertexts to reduce the complexity of HE-encrypted training at the MENs/CS. On that basis, the mobile service provider (MSP) can incentivize straggling MUs to encrypt part of their local datasets that are uploaded to certain MENs or the CS for caching and remote training. However, caching a large amount of encrypted data at the MENs and CS for FL may not only overburden those nodes but also incur a prohibitive cost of remote training, which ultimately reduces the MSP’s overall profit. To optimize the portion of MUs’ data to be encrypted, cached, and trained at the MENs/CS, we formulate an MSP’s profit maximization problem, considering all MUs’ and MENs’ resource capabilities and data handling costs (including encryption, caching, and training) as well as the MSP’s incentive budget. We then show that the problem is convex and can be efficiently solved using an interior point method. Extensive simulations on a real-world human activity recognition dataset show that our proposed framework can achieve much higher model accuracy (improving up to 24.29%) and faster convergence rate (by 2.86 times) than those of the conventionalFedAvgapproach when the straggling probability varies between 20% and 80%. Moreover, the proposed framework can improve the MSP’s profit up to 2.84 times compared with other baseline FL approaches without MEN-assisted training.

This article aims to study intrusion attacks and then develop a novel cyberattack detection framework to detect cyberattacks at the network layer (e.g., brute password and flooding of transactions) of blockchain networks. Specifically, we first design and implement a blockchain network in our laboratory. This blockchain network will serve two purposes, i.e., to generate the real traffic data (including both normal data and attack data) for our learning models and to implement real-time experiments to evaluate the performance of our proposed intrusion detection framework. To the best of our knowledge, this is the first dataset that is synthesized in a laboratory for cyberattacks in a blockchain network. We then propose a novel collaborative learning model that allows efficient deployment in the blockchain network to detect attacks. The main idea of the proposed learning model is to enable blockchain nodes to actively collect data, learn the knowledge from data using the Deep Belief Network, and then share the knowledge learned from its data with other blockchain nodes in the network. In this way, we can not only leverage the knowledge from all the nodes in the network but also do not need to gather all raw data for training at a centralized node like conventional centralized learning solutions. Such a framework can also avoid the risk of exposing local data’s privacy as well as excessive network overhead/congestion. Both intensive simulations and real-time experiments clearly show that our proposed intrusion detection framework can achieve an accuracy of up to 98.6% in detecting attacks.

This article introduces a novel lightweight framework using ambient backscattering communications to counter eavesdroppers. In particular, our framework divides an original message into two parts. The first part, i.e., the active-transmit message, is transmitted by the transmitter using conventional RF signals. Simultaneously, the second part, i.e., the backscatter message, is transmitted by an ambient backscatter tag that backscatters upon the active signals emitted by the transmitter. Notably, the backscatter tag does not generate its own signal, making it difficult for an eavesdropper to detect the backscattered signals unless they have prior knowledge of the system. Here, we assume that without decoding/knowing the backscatter message, the eavesdropper is unable to decode the original message. Even in scenarios where the eavesdropper can capture both messages, reconstructing the original message is a complex task without understanding the intricacies of the message-splitting mechanism. A challenge in our proposed framework is to effectively decode the backscattered signals at the receiver, often accomplished using the maximum likelihood (MLK) approach. However, such a method may require a complex mathematical model together with perfect channel state information (CSI). To address this issue, we develop a novel deep meta-learning-based signal detector that can not only effectively decode the weak backscattered signals without requiring perfect CSI but also quickly adapt to a new wireless environment with very little knowledge. Simulation results show that our proposed learning approach, without requiring perfect CSI and complex mathematical model, can achieve a bit error ratio close to that of the MLK-based approach. They also clearly show the efficiency of the proposed approach in dealing with eavesdropping attacks and the lack of training data for deep learning models in practical scenarios.

This paper investigates the antenna selection problem in massive multiple-input multiple-out (MIMO) systems under incomplete channel state information (CSI), with a particular interest on risk-aware planning subjected to practical constraints such as transmit power budgets and quality of services (QoS). Due to a very large number of antennas, obtaining complete channel measurements becomes a cost-prohibitive, energy-inefficient and spectral-inefficient task. To reduce pilot overhead, incomplete CSI and antenna selection (AS) are expected in practical massive MIMO systems. However, most existing AS algorithms heavily rely on the complete CSI, which imposes a high probability of violating the practical constraints in the scenarios of our interests. Motivated by this, we propose a joint channel prediction and antenna selection framework (JCPAS) which efficiently performs AS and is robust against the incomplete CSI and practical constraints. The proposed framework comprises i) a channel tracker which estimates the channel dynamics based on historical incomplete observations, and ii) a risk-aware Monte Carlo tree search (RA-MCTS) algorithm which utilizes the estimated channel dynamics to select antennas in a risk-aware manner. Simulation results show that the proposed RA-MCTS not only achieves much lower energy consumption compared to the existing typical algorithms, but also significantly reduces the probability of violating the practical constraints.

This article proposes a novel framework that utilizes an over-the-air Brain-Computer Interface (BCI) to learn Metaverse users’ expectations. By interpreting users’ brain activities, our framework can optimize physical resources and enhance Quality-of-Experience (QoE) for users. To achieve this, we leverage a Wireless Edge Server (WES) to process electroencephalography (EEG) signals via uplink wireless channels, thus eliminating the computational burden for Metaverse users’ devices. As a result, the WES can learn human behaviors, adapt system configurations, and allocate radio resources to tailor personalized user settings. Despite the potential of BCI, the inherent noisy wireless channels and uncertainty of the EEG signals make the related resource allocation and learning problems especially challenging. We formulate the joint learning and resource allocation problem as a mixed integer programming problem. Our solution involves two algorithms: a hybrid learning algorithm and a meta-learning algorithm. The hybrid learning algorithm can effectively find the solution for the formulated problem. Specifically, the meta-learning algorithm can further exploit the neurodiversity of the EEG signals across multiple users, leading to higher classification accuracy. Extensive simulation results with real-world BCI datasets show the effectiveness of our framework with low latency and high EEG signal classification accuracy.

In this paper, we propose a framework that uses Brain-Computer Interface (BCI) technology to create human-like avatars for user-driven Metaverse applications. This framework is designed to work efficiently with fast wireless connectivity and high computing demand, making it ideal for future infrastructures, e.g., 5G and beyond. The Metaverse system uses brain signals sent through wireless channels to create intelligent digital avatars that can provide helpful recommendations and assist in user-driven applications. To eliminate the computational burden on the user equipments, the computational tasks and resource allocation decisions are shifted to the centralized base station. As a result, our framework involves solving a mixed decision-making and classification problem. The goal is for the base station to efficiently allocate its computing and radio resources to users, as well as classify their brain signals. To this end, we develop a hybrid training algorithm that uses the latest advancements in deep reinforcement learning to solve the problem. Our algorithm involves three deep neural networks working together to handle both decision-making and classification tasks. Simulation results indicate that our framework can effectively manage system resources while accurately classifying users' brain signals.

In this paper, we aim to enable an intelligent traffic (TS) steering application in the open radio access network (O-RAN) by jointly optimizing the flow-split distribution, congestion control and scheduling (i.e. so-called JFCS). To do so, we develop a multi-layer optimization framework based on network utility maximization and stochastic optimization methods. The proposed algorithm provides fast convergence, long-term utility-optimality and significantly low latency compared to state-of-the-art RAN approaches. In particular, our main contributions are as follows: i) we propose the novel JFCS framework to efficiently and adaptively route traffic to indented users in appropriate radio units, and ii) we develop low-complexity algorithms to effectively solve the JFCS problem in different time scales, enabling a closed-loop control of the TS in the O-RAN context. The insights presented in this work will pave the way for 0- RAN that are completely automated, offering improved control and flexibility.

Although Rate Splitting Multiple Access (RSMA) is a promising scheme to effectively manage interference and enhance data rate and spectral utilization, its applications for Virtual Reality (VR) streaming have not been well studied. In addition to the strict latency requirement as in conventional High-Definition streaming, VR streaming further requires more computing resources at the transmitter to promptly react to the dynamic of users' Field-of-View interests. Unfortunately, current conventional RSMA approaches could not effectively handle these problems since they are not intentionally developed to deal with the special features of VR streaming. To address these challenges, we first propose a novel hierarchical multicast technique to effectively integrate the RSMA and VR streaming by exploiting the Field-of-Views of VR users. Then, the VR streaming problem established based on RSMA is formulated as a joint computation and communication optimization problem which can not only guarantee VR streaming latency requirement but also effectively manage interferences among users. Finally, due to the dynamic and uncertainty of wireless channels and users' demands, we develop a deep reinforcement learning approach to find the optimal policy for the system. This learning solution allows us to find the optimal parameters for the system via trail-and-error learning process, and thus it is effective in dealing with the uncertainty and unknown information from surrounding environment. Simulation results demonstrate that our proposed solution can satisfy the VR requirement of millisecond latency that is much lower than those of the baselines.

Federated Learning (FL) with quantization and deliberately added noise over wireless networks is a promising approach to preserve the user differential privacy while reducing the wireless resources. Specifically, an FL learning process can be fused with quantized Binomial mechanism-based updates contributed by multiple users to reduce the communication overhead/cost as well as to protect the privacy of participating users. However, the optimization of wireless transmission and quantization parameters (e.g., transmit power, bandwidth, and quantization bits) as well as the added noise while guaranteeing the privacy requirement and the performance of the learned FL model remains an open and challenging problem. In this paper, we aim to jointly optimize the level of quantization, parameters of the Binomial mechanism, and devices' transmit powers to minimize the training time under the constraints of the wireless networks. The resulting optimization turns out to be a Mixed Integer Non-linear Programming (MINLP) problem, which is known to be NP-hard. To tackle it, we transform this MINLP problem into a new problem whose solutions are proved to be the optimal solutions of the original one. We then propose an approximate algorithm that can solve the transformed problem with an arbitrary relative error guarantee. Intensive simulations show that for the same wireless resources the proposed approach achieves the highest accuracy, close to that of the conventional FL with no quantization and no noise added. This suggests the faster convergence/training time of the proposed wireless FL framework while optimally preserving users' privacy.

This work proposes a novel framework to dynamically and effectively manage and allocate different types of resources for Metaverse applications, which are forecasted to demand massive resources of various types that have never been seen before. Specifically, by studying functions of Metaverse applications, we first propose an effective solution to divide applications into groups, namely MetaInstances, where common functions can be shared among applications to enhance resource usage efficiency. Then, to capture the real-time, dynamic, and uncertain characteristics of request arrival and application departure processes, we develop a semi-Markov decision process-based framework and propose an intelligent algorithm that can gradually learn the optimal admission policy to maximize the revenue and resource usage efficiency for the Metaverse service provider and at the same time enhance the Quality-of-Service for Metaverse users. Extensive simulation results show that our proposed approach can achieve up to 120% greater revenue for the Metaverse service providers and up to 178.9% higher acceptance probability for Metaverse application requests than those of other baselines.

Unlike conventional anti-eavesdropping methods that always require additional energy or computing resources (e.g., in friendly jamming and cryptography-based solutions), this work proposes a novel anti-eavesdropping solution that comes with mostly no extra power nor computing resource requirement. This is achieved by leveraging the ambient backscatter technology in which secret information can be transmitted by backscattering it over ambient radio signals. Specifically, the original message at the transmitter is first encoded into two parts: (i) active transmit message and (ii) backscatter message. The active transmit message is then transmitted by using the conventional wireless transmission method while the backscatter message is transmitted by backscattering it on the active transmit signals via an ambient backscatter tag. As the backscatter tag does not generate any active RF signals, it is intractable for the eavesdropper to detect the backscatter message. Therefore, secret information, e.g., a secret key for decryption, can be carried by the backscattered message, making the adversary unable to decode the original message. Simulation results demonstrate that our proposed solution can significantly enhance security protection for communication systems.

Unmanned-aerial-vehicle (UAV)-assisted data collection has been emerging as a prominent application due to its flexibility, mobility, and low operational cost. However, under the dynamic and uncertainty of Internet of Things data collection and energy replenishment processes, optimizing the performance for UAV collectors is a very challenging task. Thus, this article introduces a novel framework that jointly optimizes the flying speed and energy replenishment for each UAV to significantly improve the overall system performance (e.g., data collection and energy usage efficiency). Specifically, we first develop a Markov decision process to help the UAV automatically and dynamically make optimal decisions under the dynamics and uncertainties of the environment. Although traditional reinforcement learning algorithms, such as$Q$-learning and deep$Q$-learning, can help the UAV to obtain the optimal policy, they often take a long time to converge and require high computational complexity. Therefore, it is impractical to deploy these conventional methods on UAVs with limited computing capacity and energy resource. To that end, we develop advanced transfer learning techniques that allow UAVs to “share” and “transfer” learning knowledge, thereby reducing the learning time as well as significantly improving learning quality. Extensive simulations demonstrate that our proposed solution can improve the average data collection performance of the system up to 200% and reduce the convergence time up to 50% compared with those of conventional methods.

Integrated communications and sensing (ICS) has recently emerged as an enabling technology for ubiquitous sensing and IoT applications. For ICS application to autonomous vehicles (AVs), optimizing the waveform structure is one of the most challenging tasks due to strong influences between sensing and data communication functions. Specifically, the preamble of a data communication frame is typically leveraged for the sensing function. As such, the higher number of preambles in a coherent processing interval (CPI) is, the greater sensing task’s performance is. In contrast, communication efficiency is inversely proportional to the number of preambles. Moreover, surrounding radio environments are usually dynamic with high uncertainties due to their high mobility, making the ICS’s waveform optimization problem even more challenging. To that end, this article develops a novel ICS framework established on the Markov decision process and recent advanced techniques in deep reinforcement learning. By doing so, without requiring complete knowledge of the surrounding environment in advance, the ICS-AV can adaptively optimize its waveform structure (i.e., number of frames in the CPI) to maximize sensing and data communication performance under the surrounding environment’s dynamic and uncertainty. Extensive simulations show that our proposed approach can improve the joint communication and sensing performance up to 46.26% compared with other baseline methods.

Federated learning (FL) has recently become an effective approach for cyberattack detection systems, especially in Internet of Things (IoT) networks. By distributing the learning process across IoT gateways, FL can improve learning efficiency, reduce communication overheads, and enhance privacy for cyberattack detection systems. However, one of the biggest challenges for deploying FL in IoT networks is the unavailability of labeled data and dissimilarity of data features for training. In this article, we propose a novel collaborative learning framework that leverages Transfer Learning (TL) to overcome these challenges. Particularly, we develop a novel collaborative learning approach that enables a target network with unlabeled data to effectively and quickly learn “knowledge” from a source network that possesses abundant labeled data. It is important that the state-of-the-art studies require the participated data sets of networks to have the same features, thus limiting the efficiency, flexibility, as well as scalability of intrusion detection systems. However, our proposed framework can address these problems by exchanging the learning knowledge among various deep learning (DL) models, even when their data sets have different features. Extensive experiments on recent real-world cybersecurity data sets show that the proposed framework can improve more than 40% as compared to the state-of-the-art DL-based approaches.

Rate Splitting Multiple Access (RSMA) has emerged as an effective interference management scheme for applications that require high data rates. Although RSMA has shown advantages in rate enhancement and spectral efficiency, it has yet not to be ready for latency-sensitive applications such as virtual reality streaming, which is an essential building block of future 6G networks. Unlike conventional High-Definition streaming applications, streaming virtual reality applications requires not only stringent latency requirements but also the computation capability of the transmitter to quickly respond to dynamic users’ demands. Thus, conventional RSMA approaches usually fail to address the challenges caused by computational demands at the transmitter, let alone the dynamic nature of the virtual reality streaming applications. To overcome the aforementioned challenges, we first formulate the virtual reality streaming problem assisted by RSMA as a joint communication and computation optimization problem. A novel multicast approach is then proposed to cluster users into different groups based on a Field-of-View metric and transmit multicast streams in a hierarchical manner. After that, we propose a deep reinforcement learning approach to obtain the solution for the optimization problem. Extensive simulations show that our framework can achieve the millisecond-latency requirement, which is much lower than other baseline schemes.

Rate Splitting Multiple Access (RSMA) has recently emerged as a promising technique to enhance the transmission rate for multiple access networks. Unlike conventional multiple access schemes, RSMA requires splitting and transmitting messages at different rates. The joint optimization of the power allocation and rate control at the transmitter is challenging given the uncertainty and dynamics of the environment. Furthermore, securing transmissions in RSMA networks is a crucial problem because the messages transmitted can be easily exposed to adversaries. This work first proposes a stochastic optimization framework that allows the transmitter to adaptively adjust its power and transmission rates allocated to users, and thereby maximizing the sum-rate and fairness of the system under the presence of an adversary. We then develop a highly effective learning algorithm that can help the transmitter to find the optimal policy without requiring complete information about the environment in advance. Extensive simulations show that our proposed scheme can achieve non-saturated transmission rates at high SNR values with infinite blocklength. More significantly, our proposed scheme can achieve positive covert transmission rates in the finite blocklength regime, compared with zero-valued covert rates of a conventional multiple access scheme.

Intrusion detection (ID) on the cloud environment has received paramount interest over the last few years. Among the latest approaches, machine learning-based ID methods allow us to discover unknown attacks. However, due to the lack of malicious samples and the rapid evolution of diverse attacks, constructing a cloud ID system (IDS) that is robust to a wide range of unknown attacks remains challenging. In this article, we propose a novel solution to enable robust cloud IDSs using deep neural networks. Specifically, we develop two deep generative models to synthesize malicious samples on the cloud systems. The first model, conditional denoising adversarial autoencoder (CDAAE), is used to generate specific types of malicious samples. The second model (CDAEE-KNN) is a hybrid of CDAAE and the K -nearest neighbor algorithm to generate malicious borderline samples that further improve the accuracy of a cloud IDS. The synthesized samples are merged with the original samples to form the augmented datasets. Three machine learning algorithms are trained on the augmented datasets and their effectiveness is analyzed. The experiments conducted on four popular IDS datasets show that our proposed techniques significantly improve the accuracy of the cloud IDSs compared with the baseline technique and the state-of-the-art approaches. Moreover, our models also enhance the accuracy of machine learning algorithms in detecting some currently challenging distributed denial of service (DDoS) attacks, including low-rate DDoS attacks and application layer DDoS attacks.

Most conventional Federated Learning (FL) models are using a star network topology where all users aggregate their local models at a single server (e.g., a cloud server). That causes significant overhead in terms of both communications and computing at the server, delaying the training process, especially for large scale FL systems with straggling nodes. This article proposes a novel edge network architecture that enables decentralizing the model aggregation process at the server, thereby significantly reducing the training delay for the whole FL network. Specifically, we design a highly-effective in-network computation framework (INC) consisting of a user scheduling mechanism, an in-network aggregation process (INA) which is designed for both primal- and primal-dual methods in distributed machine learning problems, and a network routing algorithm with theoretical performance bounds. The in-network aggregation process, which is implemented at edge nodes and cloud node, can adapt two typical methods to allow edge networks to effectively solve the distributed machine learning problems. Under the proposed INA, we then formulate a joint routing and resource optimization problem, aiming to minimize the aggregation latency. The problem turns out to be NP-hard, and thus we propose a polynomial time routing algorithm which can achieve near optimal performance with a theoretical bound. Simulation results showed that the proposed algorithm can achieve more than 99$\%$of the optimal solution and reduce the FL training latency, up to 5.6 times w.r.t other baselines. The proposed INC framework can not only help reduce the FL training latency but also significantly decrease cloud's traffic and computing overhead. By embedding the computing/aggregation tasks at the edge nodes and leveraging the multi-layer edge-network architecture, the INC framework can liberate FL from the star topology to enable large-scale FL.

Federated learning (FL) is a new artificial intelligence concept that enables Internet-of-Things (IoT) devices to learn a collaborative model without sending the raw data to centralized nodes for processing. Despite numerous advantages, low computing resources at IoT devices and high communication costs for exchanging model parameters make applications of FL in massive IoT networks very limited. In this work, we develop a novel compression scheme for FL, calledhigh-compression federated learning (HCFL), for very large scale IoT networks. HCFL can reduce the data load for FL processes without changing their structure and hyperparameters. In this way, we not only can significantly reduce communication costs, but also make intensive learning processes more adaptable on low-computing resource IoT devices. Furthermore, we investigate a relationship between the number of IoT devices and the convergence level of the FL model and thereby better assess the quality of the FL process. We demonstrate our HCFL scheme in both simulations and mathematical analyses. Our proposed theoretical research can be used as a minimum level of satisfaction, proving that the FL process can achieve good performance when a determined configuration is met. Therefore, we show that HCFL is applicable in any FL-integrated networks with numerous IoT devices.

Federated learning (FL) can empower Internet-of-Vehicles (IoV) networks by leveraging smart vehicles (SVs) to participate in the learning process with minimum data exchanges and privacy disclosure. The collected data and learned knowledge can help the vehicular service provider (VSP) improve the global model accuracy, e.g., for road safety as well as better profits for both VSP and participating SVs. Nonetheless, there exist major challenges when implementing the FL in IoV networks, such as dynamic activities and diverse quality-of-information (QoI) from a large number of SVs, VSP's limited payment budget, and profit competition among SVs. In this paper, we propose a novel dynamic FL-based economic framework for an IoV network to address these challenges. Specifically, the VSP first implements an SV selection method to determine a set of the best SVs for the FL process according to the significance of their current locations and information history at each learning round. Then, each selected SV can collect on-road information and propose a payment contract to the VSP based on its collected QoI. For that, we develop a multi-principal one-agent contract-based policy to maximize the profits of the VSP and learning SVs under the VSP's limited payment budget and asymmetric information between the VSP and SVs. Through experimental results using real-world on-road datasets, we show that our framework can converge 57% faster (even with only 10% of active SVs in the network) and obtain much higher social welfare of the network (up to 27.2 times) compared with those of other baseline FL methods.

This work proposes a novel framework to address straggling and privacy issues for federated learning (FL)-based mobile application services, considering limited computing/communications resources at mobile users (MUs)/mobile application provider (MAP), privacy cost, the rationality and incentive competition among MUs in contributing data to the MAP. Particularly, the MAP first determines a set of the best MUs for the FL process based on MUs' provided information/features. Then, each selected MU can encrypt part of local data and upload the encrypted data to the MAP for an encrypted training process, in addition to the local training process. For that, the selected MU can propose a contract to the MAP according to its expected local and encrypted data. To find optimal contracts that can maximize utilities while maintaining high learning quality of the system, we develop a multi-principal one-agent contract-based problem considering the MUs' privacy cost, the MAP's limited computing resources, and asymmetric information between the MAP and MUs. Experiments with a real-world dataset show that our framework can speed up training time up to 49% and improve prediction accuracy up to 4.6 times while enhancing network's social welfare up to 114% under the privacy cost consideration compared with those of baseline methods.

Deep neural networks have shown vulnerability to well-designed inputs called adversarial examples. Researchers in industry and academia have proposed many adversarial example defense techniques. However, they offer partial but not full robustness. Thus, complementing them with another layer of protection is a must, especially for mission-critical applications. This article proposes a novel online selection and relabeling algorithm (OSRA) that opportunistically utilizes a limited number of crowdsourced workers to maximize the machine learning (ML) system's robustness. The OSRA strives to use crowdsourced workers effectively by selecting the most suspicious inputs and moving them to the crowdsourced workers to be validated and corrected. As a result, the impact of adversarial examples gets reduced, and accordingly, the ML system becomes more robust. We also proposed a heuristic threshold selection method that contributes to enhancing the prediction system's reliability. We empirically validated our proposed algorithm and found that it can efficiently and optimally utilize the allocated budget for crowdsourcing. It is also effectively integrated with a state-of-the-art black box defense technique, resulting in a more robust system. Simulation results show that the OSRA can outperform a random selection algorithm by 60% and achieve comparable performance to an optimal offline selection benchmark. They also show that OSRA's performance has a positive correlation with system robustness.

In this article, we propose FedChain, a novel framework for federated-blockchain systems, to enable effective transferring of tokens between different blockchain networks. Particularly, we first introduce a federated-blockchain system together with a cross-chain transfer protocol to facilitate the secure and decentralized transfer of tokens between chains. We then develop a novel PoS-based consensus mechanism for FedChain, which can satisfy strict security requirements, prevent various blockchain-specific attacks, and achieve a more desirable performance compared to those of other existing consensus mechanisms. Moreover, a Stackelberg game model is developed to examine and address the problem of centralization in the FedChain system. Furthermore, the game model can enhance the security and performance of FedChain. By analyzing interactions between the stakeholders and chain operators, we can prove the uniqueness of the Stackelberg equilibrium and find the exact formula for this equilibrium. These results are especially important for the stakeholders to determine their best investment strategies and for the chain operators to design the optimal policy to maximize their benefits and security protection for FedChain. Simulations results then clearly show that the FedChain framework can help stakeholders to maximize their profits and the chain operators to design appropriate parameters to enhance FedChain's security and performance.

In this paper, we focus on a wireless-powered sensor network coordinated by a multi-antenna access point (AP). Each node can generate sensing information and report the latest information to the AP using the energy harvested from the AP’s signal beamforming. We aim to minimize the average age-of-information (AoI) by adapting the nodes’ scheduling and the transmission control strategies jointly. To reduce the transmission delay, an intelligent reflecting surface (IRS) is used to enhance the channel conditions by controlling the AP’s beamforming strategy and the IRS’s phase shifting matrix. Considering dynamic data arrivals at different sensing nodes, we propose a hierarchical deep reinforcement learning (DRL) framework for AoI minimization in two steps. The users’ transmission scheduling is firstly determined by the outer-loop DRL approach, e.g. the DQN or PPO algorithm, and then the inner-loop optimization is used to adapt either the uplink information transmission or downlink energy transfer to all nodes. A simple and efficient approximation is also proposed to reduce the inner-loop rum time overhead. Numerical results verify that the hierarchical learning framework outperforms typical baselines in terms of the average AoI and proportional fairness among different nodes.

In this paper, we propose a robust secure transmission scheme for an active reconfigurable intelligent surface (RIS) enabled symbiotic radio (SR) system in the presence of multiple eavesdroppers (Eves). In the considered system, the active RIS is adopted to enable the secure transmission of primary signals from the primary transmitter to multiple primary users in a multicasting manner, and simultaneously achieve its own information delivery to the secondary user by riding over the primary signals. Taking into account the imperfect channel state information (CSI) related with Eves, we formulate the system power consumption minimization problem by optimizing the transmit beamforming and reflection beamforming for the bounded and statistical CSI error models, taking the worst-case SNR constraints and the SNR outage probability constraints at the Eves into considerations, respectively. Specifically, the S-Procedure and the Bernstein-Type Inequality are implemented to approximately transform the worst-case SNR and the SNR outage probability constraints into tractable forms, respectively. After that, the formulated problems can be solved by the proposed alternating optimization (AO) algorithm with the semi-definite relaxation and sequential rank-one constraint relaxation techniques. Numerical results show that the proposed active RIS scheme can reduce up to 27.0% system power consumption compared to the passive RIS.

Coded distributed computing (CDC) has recently emerged to be a promising solution to address the straggling effects in conventional distributed computing systems. By assigning redundant workloads to the computing nodes, CDC can significantly enhance the performance of the whole system. However, since the core idea of CDC is to introduce redundancies to compensate for uncertainties, it may lead to a large amount of wasted energy at the edge nodes. It can be observed that the more redundant workload added, the less impact the straggling effects have on the system. However, at the same time, the more energy is needed to perform redundant tasks. In this work, we develop a novel framework, namely CERA, to elastically allocate computing resources for CDC processes. Particularly, CERA consists of two stages. In the first stage, we model a joint coding and node selection optimization problem to minimize the expected processing time for a CDC task. Since the problem is NP-hard, we propose a linearization approach and a hybrid algorithm to quickly obtain the optimal solutions. In the second stage, we develop a smart online approach based on Lyapunov optimization to dynamically turn off straggling nodes based on their actual performance. As a result, wasteful energy consumption can be significantly reduced with minimal impact on the total processing time. Simulations using real-world datasets have shown that our proposed approach can reduce the system’s total processing time by more than 200% compared to that of the state-of-the-art approach, even when the nodes’ actual performance is not known in advance. Moreover, the results have shown that CERA’s online optimization stage can reduce the energy consumption by up to 37.14% without affecting the total processing time.

Intrusion detection systems (IDSs) provide an ef-fective solution for protecting loT systems. However, due to the massive number of loT devices (in billions) and their heterogeneity, IDSs face challenges posed by the complexity of loT data such as correlation-based features, high dimensions, and imbalance. To address these problems, this paper proposes a novel neural network architecture, called Balanced Twin Auto-Encoder (BTAE) which consists of three components, i.e., an encoder, a hermaphrodite, and a decoder. The encoder of BTAE first aims to transfer the input data into the latent space before data samples (pre-images) are translated into this space by different translation vectors. In addition, the data of the skewed labels are also generated in the latent space to address the problem of imbalanced data in which the number of attack samples is often significantly lower than those of the benign samples. Second, the hermaphrodite component serves as a bridge to move the data from the encoder to the decoder. Third, the decoder tries to copy the distribution of the samples in the latent space. BTAE is trained by a supervised learning technique, and its data representation extracted from the decoder can well distinguish the attack from the normal data. The experiments on five loT botnet datasets show that BTAE outperforms three existing groups of methods, e.g., the typical supervised learning, the well-known sampling, and the state-of-the-art representation learning. In addition, the false alarm rate (FAR) of BTAE applied for loT intrusion detection is less than equal to 1.2%.

In this paper, we develop a novel privacy-aware framework to address straggling problem in a federated learning (FL)-based mobile edge network through maximizing profit for the mobile service provider (MSP). In particular, unlike the conventional FL process when participating mobile users (MUs) have to train their all data locally, we propose a highly-effective solution that allows MUs to encrypt parts of local data and upload/cache the encrypted data to nearby mobile edge nodes (MENs) and/or a cloud server (CS) to perform additional training processes. In this way, we can not only mitigate the straggling problem caused by limited computing/communications resources at MUs but also enhance the usage efficiency of learning data from all MUs in the FL process. To optimize portions of encrypted data cached and trained at MENs/CS given constraints from MUs and the MSP while considering data privacy and training costs, we first formulate the profit maximization problem for the MSP as an optimal in-network encrypted data caching and learning optimization. We then prove that the objective function is concave, and thus an interior-point method algorithm can be effectively adopted to quickly find the optimal solution. The numerical results demonstrate that our proposed framework can enhance the profit of the MSP up to 5.39 times compared with other FL methods.

By executing offloaded tasks from mobile users, edge computing augments mobile devices with computing/communications resources from edge nodes (ENs), enabling new services/applications (e.g., real-time gaming, virtual/augmented reality). However, despite being more resourceful than mobile devices, allocating ENs’ computing/communications resources to given favorable sets of users may block other devices from their service. This is often the case for most existing task offloading and resource allocation approaches that only aim to maximize the network social welfare (e.g., minimizing the total energy consumption) but not consider the computing/battery status of each mobile device. This work develops a proportional fair task offloading and resource allocation framework for a multi-layer cooperative edge computing network to serve all user equipment (UEs) while considering both their service requirements and individual energy/battery levels. The resulting optimization involves both binary (offloading decisions) and real variables (resource allocations), making it NP-hard. To tackle it, we leverage the fact that the relaxed problem is convex and propose a distributed algorithm, namely the dynamic branchand-bound Benders decomposition (DBBD). DBBD decomposes the original problem into a master problem (MP) for the offloading decision and subproblems (SPs) for resource allocation. The SPs can either find their closed-form solutions or be solved in parallel at ENs, thus help reduce the complexity. The numerical results show that the DBBD returns the optimal solution of the problem maximizing the fairness between UEs. The DBBD has higher fairness indexes, i.e., Jain’s index and min-max ratio, in comparing with the existing ones that minimize the total consumed energy.

In this paper, we consider an IRS-assisted and wireless-powered mobile edge computing (MEC) system that allows both edge users and the IRS to harvest energy from the hybrid access point (HAP), co-located with the MEC server. Each edge user uses the harvested energy to offload its data to the MEC server. The IRS not only assists downlink energy transfer to the edge users, but also improves the users' uplink offloading rates. To minimize the overall energy consumption, we jointly optimize the users' offloading decisions, the HAP's active beamforming, as well as the IRS's energy harvesting and passive beamforming strategies. The energy minimization problem is intractable due to complicated couplings in both the objective function and constraints. We decompose this problem into the downlink energy transfer and the uplink data offloading phases. The uplink phase can be efficiently optimized by the conventional semi-definite relaxation (SDR) method, while the downlink phase depends on the alternating optimization between the users' offloading decisions and the joint active and passive beamforming strategies. Numerical results demonstrate that the proposed offloading scheme can significantly reduce the HAP's energy consumption compared with typical benchmarks.

This article proposes a cooperative friendly jamming framework for swarm unmanned aerial vehicle (UAV)-assisted amplify-and-forward (AF) relaying networks with wireless energy harvesting. We consider a swarm of hovering UAVs that relays information from a terrestrial source to a distant mobile user and simultaneously generates jamming signals to obfuscate an eavesdropper. Due to the limited energy of the UAVs, we develop a collaborative time-switching relaying protocol that allows the UAVs to collaborate to harvest wireless energy, relay information, and jam the eavesdropper. To evaluate the secrecy rate, we derive the expressions of the secrecy outage probability (SOP) in the integral form for two popular detection techniques used by the eavesdropper, i.e., selection combining and maximum-ratio combining in high signal-to-noise ratio regime. Monte Carlo simulations validate the derived SOP and show that the proposed framework outperforms the conventional AF relaying system, in terms of SOP. The insights from SOP and analysis in this work sheds light on optimizing the energy harvesting time, the number of UAVs in the swarm as well as their placements, to achieve the required secrecy protection level.

Metaverse has recently attracted paramount attention due to its potential for future Internet. However, to fully realize such potential, Metaverse applications have to overcome various challenges such as massive resource demands, interoperability among applications, and security and privacy concerns. In this paper, we propose MetaChain, a novel blockchain-based framework to address emerging challenges for the development of Metaverse applications. In particular, by utilizing the smart contract mechanism, MetaChain can effectively manage and automate complex interactions among the Metaverse Service Provider (MSP) and the Metaverse users (MUs). In addition, to allow the MSP to efficiently allocate its resources for Metaverse applications and MUs’ demands, we design a novel sharding scheme to improve the underlying blockchain’s scalability. Moreover, to leverage MUs’ resources as well as to attract more MUs to support Metaverse operations, we develop an incentive mechanism using the Stackelberg game theory that rewards MUs’ contributions to the Metaverse. Through numerical experiments, we clearly show the impacts of the MUs’ behaviors and how the incentive mechanism can attract more MUs and resources to the Metaverse.

In this paper, we propose a reconfigurable intelligent surface (RIS) assisted secure transmission scheme for a symbiotic radio multicast system, where the RIS not only assists the confidential information multicasting from a primary transmitter (PT) to multiple primary users (PUs) to against the information interception by eavesdroppers, but also delivers its own signal to a secondary user (SU) by passive reflections. We formulate a signal-to-noise ratio (SNR) maximization problem for the SU by jointly optimizing the active beamforming at the PT, amplitude reflection coefficients and phase shifts of the RIS. To address the non-convexity of the formulated problem, we propose to decompose the original problem into two sub-problems and solve them independently in an iteratively alternating manner. For the first sub-problem, we adopt the successive convex approximation (SCA) and semidefinite relaxation (SDR) techniques to design the active beamforming by proving the tightness of SDR. For the second sub-problem, the sequential rank-one constraint relaxation (SROCR) technique is adopted to handle the rank-one constraint for reflection coefficients optimization. Numerical results show that compared to the benchmark schemes, the proposed scheme can achieve up to 68.3% performance gain in terms of SNR.

Intrusion detection systems (IDSs) play a pivotal role in defending IoT systems. However, developing a robust and efficient IDS is challenging due to the rapid and continuing evolving of various forms of cyber-attacks as well as a massive number of low-end IoT devices. In this paper, we introduce a novel deep learning architecture based on auto-encoders that allows to develop a robust intrusion detection system. Specifically, we propose a novel neural network architecture called Twin Variational Auto-Encoder (TVAE) for representation learning. TVAE includes a variational Auto-Encoder (VAE) and an Auto-Encoder (AE) that share a common stage where the decoder of the VAE is used as the encoder of the AE. The TVAE is trained in an unsupervised manner to effectively transform the original representation of data at the input of the VAE into a new representation at the output of the AE. In the new representation space, the difference between normal and attack data is more distinguishable. A variant of TVAE, namely Twin Sparse Variational Auto-Encoder (TSVAE) is also introduced by imposing a sparsity constraint on the representation units. The effectiveness of TVAE and TSVAE is evaluated using popular IDS and IoT botnet datasets. The simulation results show that the accuracy of TVAE and TSVAE can achieve the best results on six datasets, which is higher than those of state-of-the-art AE and VAE variants. We also investigate various characteristics of TVAE in the latent space as well as in the data extraction process. Besides applications on the IoT IDS, TVAE can also be applicable to all conventional network IDSs.

For wireless networks under jamming attacks, suppressing the jammer is essential to guarantee a rehable communication link. However, it can be problematic to nullify the jamming signal when the correlations between transmitted jamming signals are deliberately varied over tone. Specifically recent studies reveal that the time-varying correlations create a "virtual change" m the jamming channel and thus their nullspace, even when the physical channels remain unchanged Unlike existing studies that only consider unchanged correlations or merely propose a heuristic solution to the "virtual change" problem by continuously monitoring the residual jamming signal then updating the beam-forming matrix, we develop a deep dueling Q-learning technique to minimize the magnitude of the "virtual change" by choosing a suitable allocated time for different phases of each communication frame. Extensive simulations show that the proposed techniques can suppress the jamming signal, even when the correlations vary over time, and the correlations’ trajectory is unrevealed. Moreover, our techniques do not require monitoring the residual jamming signals then updating the beam-forming matrix. Therefore, our technique can improve the system’s spectral efficiency and reduce the outage probability.

This paper aims to develop a highly-effective framework to significantly enhance the efficiency in using coded computing techniques for distributed computing tasks over heterogeneous wireless edge networks. In particular, we first formulate a joint coding and node selection optimization problem to minimize the expected total processing time for computing tasks, taking into account the heterogeneity in the nodes’ computing resources and communication links. The problem is shown to be NP-hard. To circumvent it, we leverage the unique characteristic of the problem to develop a linearization approach and a hybrid algorithm based on binary search and branch-and-bound (BB) algorithms. This hybrid algorithm can not only guarantee to find the optimal solution, but also significantly reduce the computational complexity of the BB algorithm. Simulations based on real-world datasets show that the proposed approach can reduce the total processing time up to 2.4 times compared with that of state-of-the-art approach, even without perfect knowledge regarding the node’s performance and their straggling parameters.

In this paper, we focus on a multi-user wireless network coordinated by a multi-antenna access point (AP). Each user can generate the sensing information randomly and report it to the AP. The freshness of information is measured by the age of information (AoI). We formulate the AoI minimization problem by jointly optimizing the users’ scheduling and transmission control strategies. Moreover, we employ the intelligent reflecting surface (IRS) to enhance the channel conditions and thus reduce the transmission delay by controlling the AP’s beamforming vector and the IRS’s phase shifting matrices. The resulting AoI minimization becomes a mixed-integer program and difficult to solve due to uncertain information of the sensing data arrivals at individual users. By exploiting the problem structure, we devised a hierarchical deep reinforcement learning (DRL) framework to search for optimal solution in two iterative steps. Specifically, the users’ scheduling strategy is firstly determined by the outer-loop DRL approach, and then the inner-loop optimization adapts either the uplink information transmission or downlink energy transfer to all users. Our numerical results verify that the proposed algorithm can outperform typical baselines in terms of the average AoI performance.

This article proposes a cooperative friendly jamming framework for swarm unmanned aerial vehicle (UAV)-assisted amplify-and-forward (AF) relaying networks with wireless energy harvesting (EH). In particular, we consider a swarm of hovering UAVs that relays information from a terrestrial base station to a distant mobile user and simultaneously generates friendly jamming signals to interfere/obfuscate an eavesdropper. Due to the limited energy of the UAVs, we develop a collaborative time-switching relaying protocol that allows the UAVs to collaborate in harvesting wireless energy, relay information, and jam the eavesdropper. To evaluate the performance, we derive the secrecy outage probability (SOP) for two popular detection techniques at the eavesdropper, i.e., selection combining and maximum-ratio combining. Monte Carlo simulations are then used to validate the theoretical SOP derivation. Using the derived SOP, one can obtain engineering insights to optimize the EH time and the number of UAVs in the swarm to achieve a given secrecy protection level. Furthermore, simulations show the effectiveness of the proposed framework in terms of SOP compared to the conventional AF relaying system. The analytical SOP derived in this work can also be helpful in future UAV secure-communication optimizations (e.g., trajectory and locations of UAVs). As an example, we present a case study to find the optimal corridor to locate the swarm so as to minimize the system SOP. Our proposed framework helps secure communications for various applications that require large coverage, e.g., industrial IoT, smart city, intelligent transportation systems, and critical IoT infrastructures, such as energy and water.

Unlike theoretical analysis of distributed learning (DL) in the literature, DL over wireless edge networks faces the inherent dynamics/uncertainty of wireless connections and edge nodes, making DL less efficient or even inapplicable under the highly dynamic wireless edge networks. This article addresses these problems by leveraging recent advances in coded computing and the deep dueling neural network architecture. By introducing coded structures/redundancy, a distributed learning task can be completed without waiting for straggling nodes. Unlike conventional coded computing that only optimizes the code structure, coded distributed learning over the wireless edge also requires to optimize the selection/scheduling of wireless edge nodes with heterogeneous connections, computing capability, and straggling effects. However, even neglecting the aforementioned dynamics/uncertainty, the resulting joint optimization of coding and scheduling to minimize the distributed learning time turns out to be NP-hard. To tackle this and to account for the dynamics and uncertainty of wireless connections and edge nodes, we reformulate the problem as a Markov Decision Process and design a novel deep reinforcement learning algorithm that employs the deep dueling neural network architecture to find the jointly optimal coding scheme and the best set of edge nodes for different learning tasks without explicit information about the wireless environment and edge nodes’ straggling parameters. Simulations show that the proposed framework reduces the average learning delay in wireless edge computing up to 66% compared with other DL approaches. The jointly optimal framework in this article is also applicable to any distributed learning scheme with heterogeneous and uncertain computing nodes.

Harvesting wireless power to energize miniature devices has been envisioned as a promising solution to sustain future-generation energy-sensitive networks, e.g., Internet-of-Things systems. However, due to the limited computing and communication capabilities, wirelessly powered networks (WPNs) may be incapable of employing complex security practices, e.g., encryption, which may incur considerable computation and communication overheads. This challenge makes securing energy harvesting communications an arduous task and, thus, limits the use of WPNs in many high-security applications. In this context, security at the physical layer (PHY) that exploits the intrinsic properties of the wireless medium to achieve secure communication has emerged as an alternative paradigm. This article first introduces the fundamental principles of primary PHY attacks, covering jamming, eavesdropping, and detection of covert, and then presents an overview of the prevalent countermeasures to secure both active and passive communications in WPNs. Furthermore, a number of open research issues are identified to inspire possible future research.

With outstanding features, machine learning (ML) has become the backbone of numerous applications in wireless networks. However, the conventional ML approaches face many challenges in practical implementation, such as the lack of labeled data, the constantly changing wireless environments, the long training process, and the limited capacity of wireless devices. These challenges, if not addressed, can impede the effectiveness and applicability of ML in wireless networks. To address these problems, transfer learning (TL) has recently emerged to be a promising solution. The core idea of TL is to leverage and synthesize distilled knowledge from similar tasks and valuable experiences accumulated from the past to facilitate the learning of new problems. By doing so, TL techniques can reduce the dependence on labeled data, improve the learning speed, and enhance the ML methods’ robustness to different wireless environments. This article aims to provide a comprehensive survey on the applications of TL in wireless networks. Particularly, we first provide an overview of TL, including formal definitions, classification, and various types of TL techniques. We then discuss diverse TL approaches proposed to address emerging issues in wireless networks. The issues include spectrum management, signal recognition, security, caching, localization, and human activity recognition, which are all important to next-generation networks, such as 5G and beyond. Finally, we highlight important challenges, open issues, and future research directions of TL in future wireless networks.

Autonomous Vehicles (AVs) are required to operate safely and efficiently in dynamic environments. For this, the AVs equipped with Joint Radar-Communications (JRC) functions can enhance the driving safety by utilizing both radar detection and data communication functions. However, optimizing the performance of the AV system with two different functions under uncertainty and dynamic of surrounding environments is very challenging. In this work, we first propose an intelligent optimization framework based on the Markov Decision Process (MDP) to help the AV make optimal decisions in selecting JRC operation functions under the dynamic and uncertainty of the surrounding environment. We then develop an effective learning algorithm leveraging recent advances of deep reinforcement learning techniques to find the optimal policy for the AV without requiring any prior information about surrounding environment. Furthermore, to make our proposed framework more scalable, we develop a Transfer Learning (TL) mechanism that enables the AV to leverage valuable experiences for accelerating the training process when it moves to a new environment. Extensive simulations show that the proposed transferable deep reinforcement learning framework reduces the obstacle miss detection probability by the AV up to 67% compared to other conventional deep reinforcement learning approaches. With the deep reinforcement learning and transfer learning approaches, our proposed solution can find its applications in a wide range of autonomous driving scenarios from driver assistance to full automation transportation.

Internet of Things (IoT) has emerged as a cutting-edge technology that is changing human life. The rapid and widespread applications of IoT, however, make cyberspace more vulnerable, especially to IoT-based attacks in which IoT devices are used to launch attack on cyber-physical systems. Given a massive number of IoT devices (in order of billions), detecting and preventing these IoT-based attacks are critical. However, this task is very challenging due to the limited energy and computing capabilities of IoT devices and the continuous and fast evolution of attackers. Among IoT-based attacks, unknown ones are far more devastating as these attacks could surpass most of the current security systems and it takes time to detect them and "cure" the systems. To effectively detect new/unknown attacks, in this article, we propose a novel representation learning method to better predictively "describe" unknown attacks, facilitating supervised learning-based anomaly detection methods. Specifically, we develop three regularized versions of autoencoders (AEs) to learn a latent representation from the input data. The bottleneck layers of these regularized AEs trained in a supervised manner using normal data and known IoT attacks will then be used as the new input features for classification algorithms. We carry out extensive experiments on nine recent IoT datasets to evaluate the performance of the proposed models. The experimental results demonstrate that the new latent representation can significantly enhance the performance of supervised learning methods in detecting unknown IoT attacks. We also conduct experiments to investigate the characteristics of the proposed models and the influence of hyperparameters on their performance. The running time of these models is about 1.3 ms that is pragmatic for most applications.

Mobile service providers (MSPs) are particularly vulnerable to roaming frauds, especially ones that exploit the long delay in the data exchange process of the contemporary roaming management systems, causing multi-billion dollars loss each year. In this paper, we introduce BlockRoam, a novel blockchain-based roaming management system that provides an efficient data exchange platform among MSPs and mobile subscribers. Utilizing the Proof-of-Stake (PoS) consensus mechanism and smart contracts, BlockRoam can significantly shorten the information exchanging delay, thereby addressing the roaming fraud problems. Through intensive analysis, we show that the security and performance of such PoS-based blockchain network can be further enhanced by incentivizing more users (e.g., subscribers) to participate in the network. Moreover, users in such networks often join stake pools (e.g., formed by MSPs) to increase their profits. Therefore, we develop an economic model based on Stackelberg game to jointly maximize the profits of the network users and the stake pool, thereby encouraging user participation. We also propose an effective method to guarantee the uniqueness of this game's equilibrium. The performance evaluations show that the proposed economic model helps the MSPs to earn additional profits, attracts more investment to the blockchain network, and enhances the network's security and performance.

In this paper, we propose a novel economic-efficiency framework for an electric vehicle (EV) network to maximize the profits (i.e., the amount of money that can be earned) for charging stations (CSs). To that end, we first introduce an energy demand prediction method for CSs leveraging federated learning approaches, in which each CS can train its own energy transactions locally and exchange its learned model with other CSs to improve the learning quality while protecting the CS's information privacy. Based on the predicted energy demands, each CS can reserve energy from the smart grid provider (SGP) in advance to optimize its profit. Nonetheless, due to the competition among the CSs as well as unknown information from the SGP, i.e., the willingness to transfer energy, we develop a multi-principal one-agent (MPOA) contract-based method to address these issues. In particular, we formulate the CSs’ profit maximization as a non-collaborative energy contract problem under the SGP's unknown information and common constraints as well as other CSs’ contracts. To solve this problem, we transform it into an equivalent low-complexity optimization problem and develop an iterative algorithm to find the optimal contracts for the CSs. Through simulation results using a real CS dataset, we demonstrate that our proposed framework can enhance energy demand prediction accuracy up to 24.63 percent compared with other machine learning algorithms. Furthermore, our proposed framework can outperform other economic models by 48 and 36 percent in terms of the CSs’ utilities and social welfare (i.e., the total profits of all participating entities) of the network, respectively.

In this paper, we employ multiple wireless-powered relays to assist information transmission from a multi-antenna access point to a single-antenna receiver. The wireless relays can operate in either the passive mode via backscatter communications or the active mode via RF communications, depending on their channel conditions and energy states. We aim to maximize the overall throughput by jointly optimizing the transmit beamforming and the relays’ radio modes and operating parameters. Due to the non-convex and combinatorial problem structure, we develop a novel optimization-driven hierarchical deep deterministic policy gradient (H-DDPG) approach to adapt the beamforming and relay strategies. The optimization-driven H-DDPG algorithm firstly decomposes the binary relay mode selection into the outer-loop deep$Q$-network (DQN) algorithm and then optimizes the continuous beamforming and relaying strategies by using the inner-loop DDPG algorithm. Secondly, to improve the learning efficiency, we integrate the model-based optimization into the inner-loop DDPG framework by providing a better-informed target estimation for DNN training. Simulation results reveal that these two special designs ensure a more stable learning performance and achieve a higher reward, up to 20%, compared to the conventional model-free DDPG approach.

This paper develops a novel framework to defeat a super-reactive jammer, one of the most difficult jamming attacks to deal with in practice. Specifically, the jammer has an unlimited power budget and is equipped with the self-interference suppression capability to simultaneously attack and listen to the transmitter’s activities. Consequently, dealing with super-reactive jammers is very challenging. Thus, we introduce a smart deception mechanism to attract the jammer to continuously attack the channel and then leverage jamming signals to transmit data based on the ambient backscatter communication technology. To detect the backscattered signals, the maximum likelihood detector can be adopted. However, this method is notorious for its high computational complexity and requires the model of the current propagation environment as well as channel state information. Hence, we propose a deep learning-based detector that can dynamically adapt to any channels and noise distributions. With a Long Short-Term Memory network, our detector can learn the received signals’ dependencies to achieve a performance close to that of the optimal maximum likelihood detector. Through simulation and theoretical results, we demonstrate that with our approaches, the more power the jammer uses to attack the channel, the better bit error rate performance the transmitter can achieve.

Major bottlenecks of large-scale Federated Learning (FL) networks are the high costs for communication and computation. This is due to the fact that most of current FL frameworks only consider a star network topology where all local trained models are aggregated at a single server (e.g., a cloud server). This causes significant overhead at the server when the number of users are huge and local models' sizes are large. This paper proposes a novel edge network architecture which decentralizes the model aggregation process at the server, thereby significantly reducing the aggregation latency of the whole network. In this architecture, we propose a highly-effective in-network computation protocol consisting of two components. First, an in-network aggregation process is designed so that the majority of aggregation computations can be offloaded from cloud server to edge nodes. Second, a joint routing and resource allocation optimization problem is formulated to minimize the aggregation latency for the whole system at every learning round. The problem turns out to be NP-hard, and thus we propose a polynomial time routing algorithm which can achieve near optimal performance with a theoretical bound. Numerical results show that our proposed framework can dramatically reduce the network latency, up to 4.6 times. Furthermore, this framework can significantly decrease cloud's traffic and computing overhead by a factor of$K$/ M, where$K$is the number of users and$M$is the number of edge nodes, in comparison with conventional baselines.

This paper proposes a novel framework that can effectively address key challenges for the development of distributed learning over wireless edge networks. In particular, we first introduce a highly effective distributed learning model leveraging the most recent advanced coded distributed computing algorithm together with collaborative computing resources from wireless edge nodes to securely and effectively execute learning tasks. To minimize the average delay of learning tasks, the coding and scheduling policies must be jointly optimized. However, determining the optimal coding scheme together with the optimal edge nodes for different learning tasks is NP-hard due to the dynamics and uncertainty of the wireless environment and straggling problems at the computing nodes. Thus, we develop a highly effective approach utilizing advances of both reinforcement learning algorithms and the dueling network architecture to quickly find the optimal coding scheme together with the best edge nodes for different learning tasks without requiring completed information about the surrounding environment and straggling parameters in advance. Through extensive simulation results, we show that our proposed framework can reduce the average delay for the whole system up to 66% compared with other conventional learning and optimization approaches.

The Internet-of-Vehicles (IoV) can make driving safer and bring more services to smart vehicle (SV) users. Specif-ically, with IoV, the road service provider (RSP) can collaborate with SVs to provide high-accurate on-road information-based services by implementing federated learning (FL). Nonetheless, SVs' activities are very diverse in IoV networks, e.g., some SVs move frequently while other SVs are occasionally disconnected from the network. Consequently, obtaining information from all SVs for the learning process is costly and impractical. Furthermore, the quality-of-information (QoI) obtained by SVs also dramatically varies. That makes the learning process from all SVs simultaneously even worse when some SVs have low QoI. In this paper, we propose a novel selective FL approach for an IoV network to address these issues. Particularly, we first develop an SV selection method to determine a set of active SVs based on their location significance. In this case, we adopt a K-means algorithm to classify significant and insignificant areas where the SVs are located according to the areas' average annual daily flow of vehicles. From the set of SVs in the significant areas, we select the best SVs for the FL execution based on the SVs' QoI at each learning round. Through simulation results using a real-world on-road dataset, we observe that our proposed approach can converge to the FL results even with only 10% of active SVs in the network. Moreover, our results reveal that the RSP can optimize on-road services with faster convergence up to 63% compared with other baseline FL methods.

Federated learning (FL) has emerged as a highly-effective distributed learning framework for various AI-based mobile applications. However, in conventional FL, participating mobile users (MUs) may have limited computing resources to train their local data, which leads to learning quality degradation for the whole FL process. To address this problem, coded FL (codFL) has been recently introduced, allowing MUs to upload part of their coded data to a mobile application provider (MAP) before the learning process. As a result, codFL can not only deal with the MUs' limited computing resources, but also provide more benefits for the MUs to participate in the learning process. Nonetheless, in practice, the MAP and MUs often belong to different parties who unilaterally aim to maximize their individual utility functions. Thus, in this paper, we propose an effective mechanism for the codFL process to incentivize all the participating MUs while improving the learning quality of the MAP. Specifically, we first design a codFL contract optimization problem leveraging a multi-principal one-agent (MPOA) approach in contract theory, under limited computing resources at the MAP and MUs as well as information asymmetry between them. To find the optimal contracts for MUs, we develop an iterative contract algorithm which can produce maximum utilities for all MUs while satisfying all the constraints of the MAP. Numerical results show that our framework can enhance the utilities of MUs up to 113% and system performance in terms of social welfare up to 42% compared with the baseline method.

Conventional anti-jamming solutions like frequency hopping and rate adaptation that are more suitable for proactive jammers are not effective in dealing with reactive jammers. These advanced jammers with recent advances in signal detection can discern the activities of legitimate radios then attack them as soon as the transmission is detected. To combat this type of jammer, we develop an intelligent deception strategy in which the transmitter generates "fake" transmissions to attract the jammer. After that, the transmitter can either harvest energy from the jamming signals or backscatter the jamming signals to transmit data. As such, we can leverage jamming signals to improve the average throughput and reduce the packet loss. To effectively learn from and adapt to the dynamic and uncertainty of jamming attacks, we develop a Markov decision process (MDP) that can dynamically construct two decision epochs in each time slot to capture the special properties of our proposed deception mechanism. The Q-learning algorithm then can be adopted to find the optimal deception strategy for the transmitter. Nevertheless, due to very-slow convergence rates, conventional Q-learning algorithms may not be effective in dealing with smart jamming attacks. We thus develop an advanced deep reinforcement learning model based on deep dueling architecture to quickly obtain the optimal defense policy. Simulation results show that the proposed framework can improve the system throughput up to 173% and reduce the packet loss by 42% compared with other anti-jamming strategies that are not equipped with the proposed deception mechanism.

Symbiotic radio (SR) has recently emerged as a promising technology to boost spectrum efficiency of wireless communications by allowing reflective communications underlying the active RF communications. In this paper, we leverage SR to boost physical layer security by using an array of passive reflecting elements constituting the intelligent reflecting surface (IRS), which is reconfigurable to induce diverse RF radiation patterns. In particular, by switching the IRS’s phase shifting matrices, we can proactively create dynamic channel conditions, which can be exploited by the transceivers to extract common channel features and thus used to generate secret keys for encrypted data transmissions. As such, we firstly present the design principles for IRS-assisted key generation and verify a performance improvement in terms of the secret key generation rate (KGR). Our analysis reveals that the IRS’s random phase shifting may result in a non-uniform channel distribution that limits the KGR. Therefore, to maximize the KGR, we propose both a heuristic scheme and deep reinforcement learning (DRL) to control the switching of the IRS’s phase shifting matrices. Simulation results show that the DRL approach for IRS-assisted key generation can significantly improve the KGR.

Unmanned Aerial Vehicles (UAVs) have been emerging as an effective solution for IoT data collection networks thanks to their outstanding flexibility, mobility, and low operation costs. However, due to the limited energy and uncertainty from the data collection process, speed control is one of the most important factors while optimizing the energy usage efficiency and performance for UAV collectors. This work aims to develop a novel autonomous speed control approach to address this issue. To that end, we first formulate the dynamic speed control task of a UAV as a Markov decision process taking into account its energy status and location. In this way, the Q-learning algorithm can be adopted to obtain the optimal speed control policy for the UAV. To further improve the system performance, we develop a highly-effective deep dueling double Q-learning algorithm utilizing outstanding features of the deep neural networks as well as advanced dueling architecture to quickly stabilize the learning process and obtain the optimal policy. Through simulations, we show that our proposed solution can achieve up to 40% greater performance, i.e., an average throughput of the system, compared with other conventional methods. Importantly, the simulation results also reveal significant impacts of UAV's energy and charging time on the system performance.

In this paper, we consider IRS-assisted transmissions from a multi-antenna access point (AP) to a receiver with uncertain channel information. By adjusting the magnitude of reflecting coefficients, the IRS can sustain its operations by harvesting energy from the AP's signal beamforming. Considering channel estimation errors, we model both the AP-IRS channel and the AP-IRS-receiver as a cascaded channel by norm-based uncertainty sets. This allows us to formulate a robust optimization problem to minimize the AP's transmit power, subject to the receiver's worst-case data rate requirement and the IRS's worst-case power budget constraint. Instead of using the alternating optimization (AO) method, we firstly propose a heuristic scheme to decompose the IRS's phase shift optimization and the AP's active beamforming. Based on semidefinite relaxations of the worst-case constraints, we further devise an iterative algorithm to optimize the AP's transmit beamforming and the magnitude of the IRS's reflecting coefficients efficiently by solving a set of semidefinite programs. Simulation results reveal that the AP requires a higher transmit power to deal with the channel uncertainty. Moreover, the negative effect of channel uncertainty can be alleviated by using a larger-size IRS.

In this paper, we propose a novel blockchain-based platform for the coalition loyalty program management. The platform allows the customers to freely exchange loyalty points from different existing blockchain-based loyalty programs by utilizing the sidechain technology. Moreover, by adopting the Proof-of-Stake consensus mechanism, we can further increase customer engagement by allowing the customers to participate in the consensus process to earn additional tokens. However, this might lead to situations where the customers centralize all tokens to a single chain/loyalty program if the chain offers more rewards for consensus participation. Through security and performance analyses, we show that such centralization of stakes poses a threat to the security and performance of the platform. Therefore, we develop a non-cooperative game model to analyze the rational behavior of the users. We reveal that the consensus participation rewards govern the user behavior and the decentralization of the system. Numerical experiments confirm our analytical results and show that the ratios between the consensus rewards have a significant impact on the system's security and performance.

In intelligent transportation systems (ITS), vehicles are expected to feature with advanced applications and services which demand ultra-high data rates and low-latency communications. For that, the millimeter wave (mmWave) communication has been emerging as a very promising solution. However, incorporating the mmWave into ITS is particularly challenging due to the high mobility of vehicles and the inherent sensitivity of mmWave beams to dynamic blockages. This article addresses these problems by developing an optimal beam association framework for mmWave vehicular networks under high mobility. Specifically, we use the semi-Markov decision process to capture the dynamics and uncertainty of the environment. The Q-learning algorithm is then often used to find the optimal policy. However, Q-learning is notorious for its slow-convergence. Instead of adopting deep reinforcement learning structures (like most works in the literature), we leverage the fact that there are usually multiple vehicles on the road to speed up the learning process. To that end, we develop a lightweight yet very effective parallel Q-learning algorithm to quickly obtain the optimal policy by simultaneously learning from various vehicles. Extensive simulations demonstrate that our proposed solution can increase the data rate by 47% and reduce the disconnection probability by 29% compared to other solutions.

This paper proposes a hybrid-relaying scheme empowered by a self-sustainable intelligent reflecting surface (IRS) in a wireless powered communication network (WPCN), to simultaneously improve the performance of downlink energy transfer (ET) from a hybrid access point (HAP) to multiple users and uplink information transmission (IT) from users to the HAP. We propose time-switching (TS) and power-splitting (PS) schemes for the IRS, where the IRS can harvest energy from the HAP's signals by switching between energy harvesting and signal reflection in the TS scheme or adjusting its reflection amplitude in the PS scheme. For both the TS and PS schemes, we formulate the sum-rate maximization problems by jointly optimizing the IRS's phase shifts for both ET and IT and network resource allocation. To address each problem's non-convexity, we propose a two-step algorithm to obtain the near-optimal solution with high accuracy. To show the structure of resource allocation, we also investigate the optimal solutions for the schemes with random phase shifts. Through numerical results, we show that our proposed schemes can achieve significant system sum-rate gain compared to the baseline scheme without IRS.

We develop a joint offloading and resource allocation framework for a multi-layer cooperative fog computing network, aiming to minimize the total energy consumption of multiple mobile devices subject to their service delay requirements. The resulting optimization involves both binary (offloading decisions) and real variables (resource allocations), making it an NP-hard and computationally intractable problem. To tackle it, we first propose an improved branch-and-bound algorithm (IBBA) that is implemented in a centralized manner. However, due to the large size of the cooperative fog computing network, the computational complexity of the proposed IBBA is relatively high. To speed up the optimal solution searching as well as to enable its distributed implementation, we then leverage the unique structure of the underlying problem and the parallel processing at fog nodes. To that end, we propose a distributed framework, namely feasibility finding Benders decomposition (FFBD), that decomposes the original problem into a master problem for the offloading decision and subproblems for resource allocation. The master problem (MP) is then equipped with powerful cutting-planes to exploit the fact of resource limitation at fog nodes. The subproblems (SP) for resource allocation can find their closed-form solutions using our fast solution detection method. These (simpler) subproblems can then be solved in parallel at fog nodes. The numerical results show that the FFBD always returns the optimal solution of the problem with significantly less computation time (e.g., compared with the centralized IBBA approach). The FFBD with the fast solution detection method, namely FFBD-F, can reduce up to 60% and 90% of computation time, respectively, compared with those of the conventional FFBD, namely FFBD-S, and IBBA.

Mobile edge caching/computing (MEC) has been emerging as a promising paradigm to provide ultra-high rate, ultra-reliable, and/or low-latency communications in future wireless networks. In this paper, we introduce a novel MEC network architecture that leverages the optimal joint caching-delivering with horizontal cooperation among mobile edge nodes (MENs). To that end, we first formulate the content-access delay minimization problem by jointly optimizing the content caching and delivering decisions under various network constraints (e.g., network topology, storage capacity and users' demands at each MEN). However, the strongly mutual dependency between the decisions makes the problem a nested dual optimization that is proved to be NP-hard. To deal with it, we propose a novel transformation method to transform the nested dual problem to an equivalent mixed-integer nonlinear programming (MINLP) optimization problem. Then, we design a centralized solution using an improved branch-and-bound algorithm with the interior-point method to find the joint caching and delivering policy which is within 1 percent of the optimal solution. Since the centralized solution requires the full network topology and information from all MENs, to make our solution scalable, we develop a distributed algorithm which allows each MEN to make its own decisions based on its local observations. Extensive simulations demonstrate that the proposed solutions can reduce the total average delay for the whole network up to 40 percent compared with other current caching policies. Furthermore, the proposed solutions also increase the cache hit ratio for the network up to 4 times, thereby dramatically reducing the traffic load on the backhaul network.

In this paper, we introduce DeepFake, a novel deep reinforcement learning-based deception strategy to deal with reactive jamming attacks. In particular, for a smart and reactive jamming attack, the jammer is able to sense the channel and attack the channel if it detects communications from the legitimate transmitter. To deal with such attacks, we propose an intelligent deception strategy which allows the legitimate transmitter to transmit “fake” signals to attract the jammer. Then, if the jammer attacks the channel, the transmitter can leverage the strong jamming signals to transmit data by using ambient backscatter communication technology or harvest energy from the strong jamming signals for future use. By doing so, we can not only undermine the attack ability of the jammer, but also utilize jamming signals to improve the system performance. To effectively learn from and adapt to the dynamic and uncertainty of jamming attacks, we develop a novel deep reinforcement learning algorithm using the deep dueling neural network architecture to obtain the optimal policy with thousand times faster than those of the conventional reinforcement algorithms. Extensive simulation results reveal that our proposed DeepFake framework is superior to other anti-jamming strategies in terms of throughput, packet loss, and learning rate.

This article studies the strategic interactions between an IoT service provider (IoTSP) which consists of heterogeneous IoT devices and its energy service provider (ESP). To that end, we propose an economic framework using the Stackelberg game to maximize the network throughput and energy efficiency of both the IoTSP and ESP. To obtain the Stackelberg equilibrium (SE), we apply a backward induction technique which first derives a closed-form solution for the ESP (follower). Then, to tackle the non-convex optimization problem for the IoTSP (leader), we leverage theblock coordinate descentandconvex-concave proceduretechniques to design two partitioning schemes (i.e., partial adjustment (PA) and joint adjustment (JA)) to find the optimal energy price and service time that constitute local SEs. Numerical results reveal that by jointly optimizing the energy trading and time allocation for IoT devices, one can achieve significant improvements in terms of the IoTSP’s profit compared with those of conventional transmission methods (up to 38.7 folds). Different tradeoffs between the ESP’s and IoTSP’s profits and complexities of the PA/JA schemes can also be numerically tuned. Simulations also show that the obtained local SEs approach the optimal social welfare when the benefit per transmitted bit exceeds a given threshold.

We investigate the performance of multi-user multiple-antenna downlink systems in which a base station (BS) serves multiple users via a shared wireless medium. In order to fully exploit the spatial diversity while minimizing the passive energy consumed by radio frequency (RF) components, the BS is equipped with$M$RF chains and$N$antennas, where$M < N$. Upon receiving pilot sequences to obtain the channel state information (CSI), the BS determines the best subset of$M$antennas for serving the users. We propose a joint antenna selection and precoding design (JASPD) algorithm to maximize the system sum rate subject to a transmit power constraint and quality of service (QoS) requirements. The JASPD algorithm overcomes the non-convexity of the formulated problem via a doubly iterative algorithm, in which an inner loop successively optimizes the precoding vectors, followed by an outer loop that tests all valid antenna subsets. Although approaching (near) global optimality, the JASPD suffers from a combinatorial complexity, which may limit its application in real-time network operations. To overcome this limitation, we propose a learning-based antenna selection and precoding design algorithm (L-ASPA), which employs a deep neural network (DNN) to establish underlaying relations between key system parameters and the selected antennas. The proposed L-ASPD algorithm is robust against the number of users and their locations, the transmit power of the BS, as well as the small-scale channel fading. With a well-trained learning model, it is shown that the L-ASPD algorithm significantly outperforms baseline schemes based on the block diagonalization and a learning-assisted solution for broadcasting systems and achieves a better effective sum rate than that of the JASPA under limited processing time. In addition, we observed that the proposed L-ASPD algorithm can reduce the computation complexity by 95% while retaining more than 95% of the optimal performance.

This paper develops a beam association framework for mm Wave vehicular networks to improve the system performance in terms of handover, disconnection time, and data rate under the high mobility of vehicles. In particular, we recruit the semi Markov decision process to capture the uncertainty and dynamic of the environment such as locations of beams, received signal strength indicator profiles, velocities, and blockages. Instead of adopting complex deep learning structures such as deep dueling and double deep Q-learning, we develop a lightweight yet very effective parallel Q-learning algorithm to quickly derive the optimal beam association policy by simultaneously learning from various vehicles on the road. Through extensive simulation results, we demonstrate that the proposed framework can reduce the average disconnection time by 33% and increase the data rate by 60% compared to other solutions. We also observed that the proposed parallel Q-learning algorithm converges much faster to the optimal solution than state-of-the-art deep-learning based algorithms.

Intelligent reflecting surface (IRS) is a promising technology to assist downlink information transmissions from a multi-antenna access point (AP) to a receiver. In this paper, we minimize the AP's transmit power by a joint optimization of the AP's active beamforming and the IRS's passive beamforming. Due to uncertain channel conditions, we formulate a robust power minimization problem subject to the receiver's signal-to-noise ratio (SNR) requirement and the IRS's power budget constraint. We propose a deep reinforcement learning (DRL) approach that can adapt the beamforming strategies from past experiences. To improve the learning performance, we derive a convex approximation as a lower bound on the robust problem, which is integrated with the DRL framework and thus promoting a novel optimization-driven deep deterministic policy gradient (DDPG) approach. In particular, when the DDPG algorithm generates a part of the action (e.g., passive beamforming), we can use the model-based convex approximation to optimize the other part of the action (e.g., active beamforming) efficiently. Our simulation results demonstrate that the optimization-driven DDPG algorithm can improve both the learning rate and reward significantly compared to the conventional DDPG algorithm.

In this paper, an economic model is proposed to jointly optimize profits for participants in a heterogeneous IoT wireless-powered backscatter communication network. In the network under considerations, a power beacon and IoT devices (with various communication types and energy constraints) are assumed to belong to different service providers, i.e., energy service provider (ESP) and IoT service provider (ISP), respectively. To jointly maximize the utility for both service providers in terms of energy efficiency and network throughput, a Stackelberg game model is proposed to study the strategic interaction between the ISP and ESP. In particular, the ISP first evaluates its benefits from providing IoT services to its customers and then sends its requested price together with the service time to the ESP. Based on the request from the ISP, the ESP offers an optimized transmission power that maximizes its utility while meeting energy demands of the ISP. To study the Stackelberg equilibrium, we first obtain a closed-form solution for the ESP and propose a low-complexity iterative method based on block coordinate descent (BCD) to address the non-convex optimization problem for the ISP. Through simulation results, we show that our approach can significantly improve the profits for both providers compared with those of conventional transmission methods, e.g., bistatic backscatter and harvest-then-transmit communication methods.

The rapid adoption of electric or hybrid vehicles (EVs) has called for wide deployment of charging stations. These stations can be launched/owned by different owners, referred to as charging station providers (CSPs), which make energy contracts with a smart grid provider (SGP). However, there exists a shortage of mutual economic strategy between the SGP and CSPs in an energy request/transfer competition due to the selfish nature among them. In this paper, we propose an economic model leveraging a multi-principal single-agent (referred to as common agency) contract policy, aiming at maximizing the utilities of multiple CSPs while optimizing the utility of the SGP in an EV network. In particular, we first develop the common agency-based contract problem as a non-cooperative energy contract optimization problem, in which each CSP can maximize its utility given the common constraints from the SGP and the contracts of other CSPs. To deal with this problem, we develop an iterative energy contract algorithm to find an equilibrium contract solution where the contracts from the CSPs can produce maximum utilities of the CSPs and satisfy the constraints of the SGP. Through numerical results, we show that our proposed model can improve the social welfare of the EV network up to 54% and the utilities of CSPs up to 60% compared with the baseline method in which each CSP obtains the amount of energy that is proportional to its energy request.

Among all wireless jammers, dealing with reactive ones is most challenging. This kind of jammer attacks the channel whenever it detects transmission from legitimate radios. With recent advances in self-interference suppression or in-band full-duplex radios, a reactive jammer can jam and simultaneously sense/discern/detect the legitimate transmission. Such a jammer is referred to as a smart reactive jammer. However, all existing solutions, e.g., frequency hopping and rate adaptation, cannot effectively deal with this type of jammer. This is because a smart reactive jammer with sufficient power budget can theoretically jam most, if not all, frequency channels at sufficiently high power. This work proposes to augment the transmitter with an ambient backscatter tag. Specifically, when the jammer attacks the channel, the transmitter deceives it by continuing to transmit data to attract the jammer while the tag backscatters data based on both the jamming signals and active signals from the jammer and transmitter, respectively. However, backscattering signals from multiple radio sources results in a high bit error rate (BER). Thus, we propose to use multiple antennas at the receiver. The theoretical analysis and simulation results show that by using multiple antennas at the receiver, the BER and hence the throughput of the system can be significantly improved. More importantly, we demonstrate that with our proposed solutions, the average throughput increases and the BER decreases when the jammer attacks with higher power levels. We believe that this is the first anti-jamming solution that can cope effectively with a high- or even unlimited-power jammers.

Although the development of IoT Industry 4.0 has brought breakthrough achievements in many sectors, e.g., manufacturing, healthcare, and agriculture, it also raises many security issues to human beings due to a huge of emerging cybersecurity threats recently. In this paper, we propose a novel collaborative learning-based intrusion detection system which can be efficiently implemented in IoT Industry 4.0. In the system under consideration, we develop smart “filters” which can be deployed at the IoT gateways to promptly detect and prevent cyberattacks. In particular, each filter uses the collected data in its network to train its cyberattack detection model based on the deep learning algorithm. After that, the trained model will be shared with other IoT gateways to improve the accuracy in detecting intrusions in the whole system. In this way, not only the detection accuracy is improved, but our proposed system also can significantly reduce the information disclosure as well as network traffic in exchanging data among the IoT gateways. Through thorough simulations on real datasets, we show that the performance obtained by our proposed method can outperform those of the conventional machine learning methods.

Roaming fraud is one of the most significant financial losses for mobile service providers. The inefficiency of current exchanging data management methods among mobile service providers is the main obstacle for roaming fraud prevention. In this paper, we introduce a novel blockchain-based data exchange management system to address roaming fraud problems in mobile networks. This system provides a secure and automatic data exchange service among mobile service providers and mobile subscribers. In addition, we introduce an emerging Proof-of-Stake (PoS) consensus mechanism for the proposed blockchain-based roaming fraud prevention system, which can significantly reduce the delay in exchanging information as well as implementation costs for mobile service providers. To further enhance benefits and security efficiency for the proposed blockchain system, we develop an economic model based on Stackelberg game. This game model is very effective in maximizing profits for both the stakeholders and stake pool and useful in designing a robust blockchain-based mobile roaming management system. Through performance analysis and numerical results, we show that our proposed framework not only provides an effective solution to prevent mobile roaming fraud but also opens many business opportunities for future mobile networks.

In this paper, we employ multiple wireless-powered user devices as wireless relays to assist information transmission from a multi-antenna access point to a single-antenna receiver. To improve energy efficiency, we design a hybrid relaying communication strategy in which wireless relays are allowed to operate in either the passive mode via backscatter communications or the active mode via RF communications, depending on their channel conditions and energy states. We aim to maximize the overall SNR by jointly optimizing the access point's beamforming strategy as well as individual relays' radio modes and operating parameters. Due to the non-convex and combinatorial structure of the SNR maximization problem, we develop a deep reinforcement learning approach that adapts the beamforming and relaying strategies dynamically. In particular, we propose a novel optimization-driven hierarchical deep deterministic policy gradient (H-DDPG) approach that integrates the model-based optimization into the framework of conventional DDPG approach. It decomposes the discrete relay mode selection into the outer-loop by using deep Q-network (DQN) algorithm and then optimizes the continuous beamforming and relays' operating parameters by using the inner-loop DDPG algorithm. Simulation results reveal that the H-DDPG is robust to the hyper parameters and can speed up the learning process compared to the conventional DDPG approach.

The Internet of Things (IoT) for smart ocean is a promising paradigm that will support emerging applications in the areas of maritime transport, emergency search and rescue, security and border surveillance, environmental protection, etc. There has been a surging amount of data acquired from different maritime terminals, such as vessels, buoys, and offshore platforms. As a result, the demand for high-speed, ultrareliable, and low-latency maritime communications and data processing is proliferating. In this context, transmission and processing of maritime data have become a research hotspot. IoT technologies are expected to dramatically enhance the capacity, safety, and efficiency of connected vessels and other maritime terminals. Meanwhile, the unique characteristics of smart ocean applications create heterogeneous challenges in achieving viable, reliable, and secure communications and data processing. Addressing the challenges calls for novel approaches and consideration for the deployment of next-generation maritime communication networks. Therefore, it is essential to pursue research on new theories, architecture, and technologies to fully exploit the capability that is delivered by IoT for smart ocean to form efficient and intelligent maritime communication systems. This special issue aims to create a platform for researchers from both academia and industry to disseminate state-of-the-art results and to advance the applications of IoT for the smart ocean.

In this article, we employ multiple energy harvesting relays to assist information transmission from a multiantenna hybrid access point (HAP) to a receiver. All the relays are wirelessly powered by the HAP in the power-splitting (PS) protocol. We introduce the novel concept of hybrid relay communications, which allows each relay to switch between two radio modes, i.e., the active RF communications and the passive backscatter communications, according to its channel and energy conditions. We aim to jointly optimize the HAP's beamforming, individual relays' radio modes, PS ratios, and the relays' collaborative beamforming strategies to enhance the throughput performance at the receiver. The resulting formulation becomes a combinatorial and nonconvex problem. We first propose a convex approximation to the original problem, which serves as a lower bound of the relay performance. Then, we design an iterative algorithm that decomposes the binary relay mode optimization from the other operating parameters. In the inner loop of the algorithm, we exploit the structural properties to optimize the relay performance with the fixed relay mode by using alternating optimization. In the outer loop, different performance metrics are derived to guide the search for a set of passive relays to further improve the relay performance. The simulation results verify that the hybrid relaying communications can achieve 20% performance improvement compared to the conventional relay communications with all active relays.

Internet of things (IoT) produces massive data from devices embedded with sensors. The IoT data allows creating profitable services using machine learning. However, previous research does not address the problem of optimal pricing and bundling of machine learning-based IoT services. In this paper, we define the data value and service quality from a machine learning perspective. We present an IoT market model which consists of data vendors selling data to service providers, and service providers offering IoT services to customers. Then, we introduce optimal pricing schemes for the standalone and bundled selling of IoT services. In standalone service sales, the service provider optimizes the size of bought data and service subscription fee to maximize its profit. For service bundles, the subscription fee and data sizes of the grouped IoT services are optimized to maximize the total profit of cooperative service providers. We show that bundling IoT services maximizes the profit of service providers compared to the standalone selling. For profit sharing of bundled services, we apply the concepts of core and Shapley solutions from cooperative game theory as efficient and fair allocations of payoffs among the cooperative service providers in the bundling coalition.

In this paper, we consider a smart jammer that only attacks the channel if it detects activities of legitimate devices on that channel. To cope with smart jamming attacks, we propose an intelligent deception strategy in which the legitimate device will send fake transmissions to lure the jammer. Then, if the jammer launches attacks to the channel, the legitimate device can either backscatter the jamming signals to transmit data or harvest energy from the jamming signals for future active transmission. In this way, we can not only undermine the attack ability of the jammer, but also leverage jamming attacks as means to enhance system performance. In addition, to find an optimal defense strategy for the legitimate device under uncertainty of wireless environment as well as incomplete information from the jammer, we develop Q-learning and deep Q-learning algorithms based on the Markov decision process. Through simulation results, we demonstrate that our proposed solution is able to not only deal with smart jamming attacks, but also successfully leverage jamming attacks to improve the system performance.

We propose a jointly optimal caching and routing framework (JOCAR) for a cooperative mobile edge caching network. This novel network architecture enables mobile edge servers/nodes (MENs) to collaborate in not only caching but also routing contents to users, in order to simultaneously minimize the total content-access delay for all mobile users and reduce the traffic on the backhaul network. To that end, we first formulate an access- delay minimization problem by jointly optimizing the content caching and routing decisions while accounting for various network configurations. Solving this problem requires us to deal with a nested dual optimization due to the strong mutual dependence between content caching and routing decisions. To tackle it, we first transform the nested dual problem to an equivalent mixed-integer nonlinear programming (MINLP) problem. Then, we design a branch-and-bound based algorithm with the interior-point method to find the near-optimal policy for the MINLP problem. Extensive simulations show that JOCAR can reduce the total average delay and increase the cache hit rate for the whole network by more than 40% and by four times, respectively, compared with other conventional policies.

In this paper, we propose novel approaches using state-of-the-art machine learning techniques, aiming at predicting energy demand for electric vehicle (EV) networks. These methods can learn and find the correlation of complex hidden features to improve the prediction accuracy. First, we propose an energy demand learning (EDL)-based prediction solution in which a charging station provider (CSP) gathers information from all charging stations (CSs) and then performs the EDL algorithm to predict the energy demand for the considered area. However, this approach requires frequent data sharing between the CSs and the CSP, thereby driving communication overhead and privacy issues for the EVs and CSs. To address this problem, we propose a federated energy demand learning (FEDL) approach which allows the CSs sharing their information without revealing real datasets. Specifically, the CSs only need to send their trained models to the CSP for processing. In this case, we can significantly reduce the communication overhead and effectively protect data privacy for the EV users. To further improve the effectiveness of the FEDL, we then introduce a novel clustering- based EDL approach for EV networks by grouping the CSs into clusters before applying the EDL algorithms. Through experimental results, we show that our proposed approaches can improve the accuracy of energy demand prediction up to 24.63% and decrease communication overhead by 83.4% compared with other baseline machine learning algorithms.

We investigate a joint offloading and resource allocation under a multi-layer cooperative fog and cloud computing architecture, aiming to minimize the total energy consumption of mobile devices while meeting users' QoS requirements, e.g., delay, security, and application compatibility. Due to the mutual coupling amongst offloading decision and resource allocation variables, the resulting optimization is a mixed integer non- linear programming problem that is NP-hard. Such problem often requires exponential time to find the optimal solution. In this work, we propose a distributed approach, namely feasibility-finding Benders decomposition (FFBD), that decomposes the original problem into a master problem for the offloading decision and subproblems for resource allocation. These (simpler) subproblems can be solved in parallel at fog nodes, thereby reducing both the complexity and the computational time. The numerical results show that the FFBD always returns the optimal solution of the problem with significantly less computation time (e.g., in comparing with the branch-and-bound method).

In this work, we consider multiple energy harvesting relays to assist information transmission from a hybrid access point (HAP) to a distant receiver. The multi-antenna HAP also beamforms RF power to the relays by using a power-splitting protocol. We aim to maximize the throughput by jointly optimizing the HAP's beamforming strategy as well as individual relays' energy harvesting and collaborative beamforming strategies. With dense user devices, the throughput maximization takes account of the direct links from the HAP to the receiver as they are short and contribute considerably to the overall throughput. Moreover, we introduce the concept of hybrid relaying communications which allows the energy harvesting relays to switch between two radio modes. In particular, the relays can operate either in RF communications or backscatter communications, depending on their channel conditions and energy status. This results in a non-convex and combinatorial throughput maximization problem. With the fixed relay mode, we can find a feasible lower performance bound via convex approximation, which further motivates our algorithm design to update the relay mode in an iterative manner. Simulation results verify that the proposed hybrid relaying strategy can achieve significant performance improvement compared to the conventional relaying strategy with all relays operating in the RF communications mode.

Practical and efficient network slicing often faces real-time dynamics of network resources and uncertain customer demands. This work provides an optimal and fast resource slicing solution under such dynamics by leveraging the latest advances in deep learning. Specifically, we first introduce a novel system model which allows the network provider to effectively allocate its combinatorial resources, i.e., spectrum, computing, and storage, to various classes of users. To allocate resources to users while taking into account the dynamic demands of users and resources constraints of the network provider, we employ a semi-Markov decision process framework. To obtain the optimal resource allocation policy for the network provider without requiring environment parameters, e.g., uncertain service time and resource demands, a Q-learning algorithm is adopted. Although this algorithm can maximize the revenue of the network provider, its convergence to the optimal policy is particularly slow, especially for problems with large state/action spaces. To overcome this challenge, we propose a novel approach using an advanced deep Q-learning technique, called deep dueling that can achieve the optimal policy at few thousand times faster than that of the conventional Q-learning algorithm. Simulation results show that our proposed framework can improve the long-term average return of the network provider up to 40% compared with other current approaches.

In this paper, we propose a novel approach to jointly address energy management and network throughput maximization problems for heterogeneous IoT low-power wireless communication networks. In particular, we consider a low-power communication network in which the IoT devices can harvest energy from a dedicated RF energy source to support their transmissions or backscatter the signals of the RF energy source to transmit information to the gateway. Different IoT devices may have dissimilar hardware configurations, and thus they may have various communications types and energy requirements. In addition, the RF energy source may have a limited energy supply source which needs to be minimized. Thus, to maximize the network throughput, we need to jointly optimize energy usage and operation time for the IoT devices under different energy demands and communication constraints. However, this optimization problem is non-convex due to the strong relation between energy supplied by the RF energy source and the IoT communication time, and thus obtaining the optimal solution is intractable. To address this problem, we study the relation between energy supply and communication time, and then transform the non-convex optimization problem to an equivalent convex-optimization problem which can achieve the optimal solution. Through simulation results, we show that our solution can achieve greater network throughputs (up to five times) than those of other conventional methods, e.g., TDMA. In addition, the simulation results also reveal some important information in controlling energy supply and managing low-power IoT devices in heterogeneous wireless communication networks.

We develop a novel deep learning model, Multidistributed Variational AutoEncoder (MVAE), for the network intrusion detection. To make the traffic more distinguishable, MVAE introduces the label information of data samples into the Kullback-Leibler (KL) term of the loss function of Variational AutoEncoder (VAE). This label information allows MVAEs to force/partition network data samples into different classes with different regions in the latent feature space. As a result, the network traffic samples are more distinguishable in the new representation space (i.e., the latent feature space of MVAE), thereby improving the accuracy in detecting intrusions. To evaluate the efficiency of the proposed solution, we carry out intensive experiments on two popular network intrusion datasets, i.e., NSL-KDD and UNSWNB15 under four conventional classifiers including Gaussian Naive Bayes (GNB), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF). The experimental results demonstrate that our proposed approach can significantly improve the accuracy of intrusion detection algorithms up to 24.6% compared to the original one (using area under the curve metric).

Effective network slicing requires an infrastructure/network provider to deal with the uncertain demands and real-time dynamics of the network resource requests. Another challenge is the combinatorial optimization of numerous resources, e.g., radio, computing, and storage. This paper develops an optimal and fast real-time resource slicing framework that maximizes the long-term return of the network provider while taking into account the uncertainty of resource demands from tenants. Specifically, we first propose a novel system model that enables the network provider to effectively slice various types of resources to different classes of users under separate virtual slices. We then capture the real-time arrival of slice requests by a semi-Markov decision process. To obtain the optimal resource allocation policy under the dynamics of slicing requests, e.g., uncertain service time and resource demands, a Q-learning algorithm is often adopted in the literature. However, such an algorithm is notorious for its slow convergence, especially for problems with large state/action spaces. This makes Q-learning practically inapplicable to our case, in which multiple resources are simultaneously optimized. To tackle it, we propose a novel network slicing approach with an advanced deep learning architecture, called deep dueling, that attains the optimal average reward much faster than the conventional Q-learning algorithm. This property is especially desirable to cope with the real-time resource requests and the dynamic demands of the users. Extensive simulations show that the proposed framework yields up to 40% higher long-term average return while being few thousand times faster, compared with the state-of-the-art network slicing approaches.

With conventional anti-jamming solutions like frequency hopping or spread spectrum, legitimate transceivers often tend to “escape” or “hide” themselves from jammers. These reactive anti-jamming approaches are constrained by the lack of timely knowledge of jamming attacks (especially from smart jammers). Bringing together the latest advances in neural network architectures and ambient backscattering communications, this work allows wireless nodes to effectively “face” the jammer (instead of escaping) by first learning its jamming strategy, then adapting the rate or transmitting information right on the jamming signals (i.e., backscattering modulated information on the jamming signals). Specifically, to deal with unknown jamming attacks (e.g., jamming strategies, jamming power levels, and jamming capability), existing work often relies on reinforcement learning algorithms, e.g., Q -learning. However, the Q -learning algorithm is notorious for its slow convergence to the optimal policy, especially when the system state and action spaces are large. This makes the Q -learning algorithm pragmatically inapplicable. To overcome this problem, we design a novel deep reinforcement learning algorithm using the recent dueling neural network architecture. Our proposed algorithm allows the transmitter to effectively learn about the jammer and attain the optimal countermeasures (e.g., adapt the transmission rate or backscatter or harvest energy or stay idle) thousand times faster than that of the conventional Q -learning algorithm. Through extensive simulation results, we show that our design (using ambient backscattering and the deep dueling neural network architecture) can improve the average throughput (under smart and reactive jamming attacks) by up to 426% and reduce the packet loss by 24%. By augmenting the ambient backscattering capability on devices and using our algorithm, it is interesting to observe that the (successful) transmission rate increases with the jamming power. Our proposed solution can find its applications in both civil (e.g., ultra-reliable and low-latency communications or URLLC) and military scenarios (to combat both inadvertent and deliberate jamming).

Ambient backscatter has been introduced with a wide range of applications for low power wireless communications. In this paper, we propose an optimal and low-complexity dynamic spectrum access framework for the RF-powered ambient backscatter system. In this system, the secondary transmitter not only harvests energy from ambient signals but also reflects these signals to transmit its modulated data to the receiver. Under the dynamics of the ambient signals, we first adopt the Markov decision process (MDP) framework to obtain the optimal policy for the secondary transmitter, aiming to maximize the system throughput. However, the MDP-based optimization requires complete knowledge of environment parameters, e.g., the probability of a channel to be idle and the probability of a successful packet transmission, that may not be practical to obtain. To cope with such incomplete knowledge of the environment, we develop a low-complexity online reinforcement learning algorithm that allows the secondary transmitter to “learn” from its decisions and then attain the optimal policy. Simulation results show that the proposed learning algorithm not only efficiently deals with the dynamics of the environment but also improves the average throughput up to 50% and reduces the blocking probability and delay up to 80% compared with conventional methods.

For an RF-powered cognitive radio network with ambient backscattering capability, while the primary channel is busy, the RF-powered secondary user (RSU) can either backscatter the primary signal to transmit its own data or harvest energy from the primary signal (and store in its battery). The harvested energy then can be used to transmit data when the primary channel becomes idle. To maximize the throughput for the secondary system, it is critical for the RSU to decide when to backscatter and when to harvest energy. This optimal decision has to account for the dynamics of the primary channel, energy storage capability, and data to be sent. To tackle that problem, we propose a Markov decision process (MDP)-based framework to optimize RSU's decisions based on its current states, e.g., energy, data as well as the primary channel state. As the state information may not be readily available at the RSU, we then design a low-complexity online reinforcement learning algorithm that guides the RSU to find the optimal solution without requiring prior-and complete-information from the environment. The extensive simulation results then clearly show that the proposed solution achieves higher throughputs, i.e., up to 50%, than that of conventional methods.

We propose a novel edge computing network architecture that enables edge nodes to cooperate in sharing computing and radio resources to minimize the total energy consumption of mobile users while meeting their delay requirements. To find the optimal task offloading decisions for mobile users, we first formulate the joint task offloading and resource allocation optimization problem as a mixed integer non-linear programming (MINLP). The optimization involves both binary (offloading decisions) and real variables (resource allocations), making it an NP-hard and computational intractable problem. To circumvent, we relax the binary decision variables to transform the MINLP to a relaxed optimization problem with real variables. After proving that the relaxed problem is a convex one, we propose two solutions namely ROP and IBBA. ROP is adopted from the interior point method and IBBA is developed from the branch and bound algorithm. Through the numerical results, we show that our proposed approaches allow minimizing the total energy consumption and meet all delay requirements for mobile users.

To effectively utilize/harvest short-lived whitespace that accounts for more than 30% of the cellular bands, it is critical to build a real-time radio environment map. Note that existing radio spectrum maps/databases (e.g., Google Spectrum Database) are updated on a daily or weekly basis. In this paper, we introduce a novel real-time crowdsourcing incentive solution that rewards mobile users who contribute their qualified spectrum sensing data to a radio environment map. First, we develop a feature-based model based on advanced machine learning techniques in order to estimate model parameters of the radio environment map. Based on the prediction model, we then propose a smart dynamic pricing strategy including prepaid and postpaid pricing schemes. The prepaid scheme is to guarantee the minimum payment for participants, and the postpaid scheme is to reward the participants according to their contributions. Importantly, in our model, the postpaid scheme will be adjusted iteratively in a real-time manner based on the contributions of participants to the spectrum map. After that we carry out real experiments through a mobile application and a cloud spectrum database. The experiment results show that our proposed solution can achieve not only better users' utilities, but also a lower overall system cost compared with those of some existing works.

5G mobile networks encompass the capabilities of hosting a variety of services such as mobile social networks, multimedia delivery, healthcare, transportation, and public safety. Therefore, the major challenge in designing the 5G networks is how to support different types of users and applications with different quality-of-service requirements under a single physical network infrastructure. Recently, network slicing has been introduced as a promising solution to address this challenge. Network slicing allows programmable network instances which match the service requirements by using network virtualization technologies. However, how to efficiently allocate resources across network slices has not been well studied in the literature. Therefore, in this paper, we first introduce a model for orchestrating network slices based on the service requirements and available resources. Then, we propose a Markov decision process framework to formulate and determine the optimal policy that manages cross-slice admission control and resource allocation for the 5G networks. Through simulation results, we show that the proposed solution is efficient not only in providing slice-as-a-service based on service requirements, but also in maximizing the provider's revenue.

The development of mobile cloud computing has brought many benefits to mobile users as well as cloud service providers. However, mobile cloud computing is facing some challenges, especially security-related problems due to the growing number of cyberattacks which can cause serious losses. In this paper, we propose a dynamic framework together with advanced risk management strategies to minimize losses caused by cyberattacks to a cloud service provider. In particular, this framework allows the cloud service provider to select appropriate security solutions, e.g., security software/hardware implementation and insurance policies, to deal with different types of attacks. Furthermore, the stochastic programming approach is adopted to minimize the expected total loss for the cloud service provider under its financial capability and uncertainty of attacks and their potential losses. Through numerical evaluation, we show that our approach is an effective tool in not only dealing with cyberattacks under uncertainty, but also minimizing the total loss for the cloud service provider given its available budget.

With the rapid growth of mobile applications and cloud computing, mobile cloud computing has attracted great interest from both academia and industry. However, mobile cloud applications are facing security issues such as data integrity, users' confidentiality, and service availability. A preventive approach to such problems is to detect and isolate cyber threats before they can cause serious impacts to the mobile cloud computing system. In this paper, we propose a novel framework that leverages a deep learning approach to detect cyberattacks in mobile cloud environment. Through experimental results, we show that our proposed framework not only recognizes diverse cyberattacks, but also achieves a high accuracy (up to 97.11%) in detecting the attacks. Furthermore, we present the comparisons with current machine learning-based approaches to demonstrate the effectiveness of our proposed solution.

In this paper, we study the transmission strategy adaptation problem in an RF-powered cognitive radio network, in which hybrid secondary users are able to switch between the harvest-then-transmit mode and the ambient backscatter mode for their communication with the secondary gateway. In the network, a monetary incentive is introduced for managing the interference caused by the secondary transmission with imperfect channel sensing. The sensing-pricing-transmitting process of the secondary gateway and the transmitters is modeled as a single-leader-multi-follower Stackelberg game. Furthermore, the follower sub-game among the secondary transmitters is modeled as a generalized Nash equilibrium problem with shared constraints. Based on our theoretical discoveries regarding the properties of equilibria in the follower sub-game and the Stackelberg game, we propose a distributed, iterative strategy searching scheme that guarantees the convergence to the Stackelberg equilibrium. The numerical simulations show that the proposed hybrid transmission scheme always outperforms the schemes with fixed transmission modes. Furthermore, the simulations reveal that the adopted hybrid scheme is able to achieve a higher throughput than the sum of the throughput obtained from the schemes with fixed transmission modes.

In this work, we investigate a system where users are served by an access point that is equipped with energy harvesting and caching mechanism. Focusing on the design of an efficient content delivery scheduling, we propose a joint scheduling and caching scheme. The scheduling problem is formulated as a Markov decision process and solved by an on-line learning algorithm. To deal with large state space, we apply the linear approximation method to the state-action value functions, which significantly reduces the memory space for storing the function values. In addition, the preference learning is incorporated to speed up the convergence when dealing with the requests from users that have obvious content preferences. Simulation results confirm that the proposed scheme outperforms the baseline scheme in terms of convergence and system throughput, especially when the personal preference is concentrated to one or two contents.

In this paper, we propose a novel network model for RF-powered cognitive radio networks and ambient backscatter communications. In the network under consideration, each secondary transmitter is able to backscatter primary signals to the gateway for data transfer or to harvest energy from the primary signals and then use that energy to transmit data to the gateway. To maximize overall network throughput of the network, we formulate an optimization problem with the aim of finding not only an optimal tradeoff between data backscattering time and energy harvesting time, but also time sharing among multiple secondary transmitters. Through the numerical results, we demonstrate that the solution of the optimization problem always achieves the best performance compared with two other baseline schemes.

In this paper, we study an overlay RF-powered cognitive radio network with ambient backscatter communications. In the network, when the channel is occupied, the secondary transmitter (ST) can perform either energy harvesting or data transmission using ambient backscattering technique to a gateway. We consider the case that the gateway charges the ST a certain price if the ST transmits information. This leads to questions of how to determine the best price for the gateway and how to find the optimal backscatter time. To address this problem, we propose a Stackelberg game in which the gateway is the leader adapting the price to maximize its profit in the first stage. Meanwhile, the ST chooses its backscatter time to maximize its utility in the second stage. To analyze the game, we apply the backward induction technique. We show that the game always has a unique subgame perfect Nash equilibrium. Additionally, our results provide insights on the impact of the competition on the players' profit and utility.

This paper introduces a cyber insurance policy management for the mobile networks in which if a mobile user agrees to purchase an insurance policy from an insurer, the loss of the mobile user, i.e., the insured, will be covered by the insurance policy when the risks happen. To protect mobile users from cyber attacks, the insurer can deploy security protection solutions, e.g., anti-virus software or personal firewall, to the insureds, thereby reducing the risks for mobile users. However, when the solutions are deployed, they will incur a certain cost to the insurer. Therefore, we propose a stochastic optimization based on the reserve state of the insurer and the number of active mobile users to determine whether the protection solutions should be deployed or not to maximize the revenue for the insurer. The performance evaluation reveals that the optimal policy can achieve significantly higher revenue than those of baseline schemes for the insurer. Alternatively, the coalitional game is studied to share the reward among the insurers, and we show that the insurers can gain higher individual rewards through the cooperation.

This paper introduces a new solution to improve the performance for secondary systems in radio frequency (RF) powered cognitive radio networks (CRNs). In a conventional RF-powered CRN, the secondary system works based on the harvest-then-transmit protocol. That is, the secondary transmitter (ST) harvests energy from primary signals and then uses the harvested energy to transmit data to its secondary receiver (SR). However, with this protocol, the performance of the secondary system is much dependent on the amount of harvested energy as well as the primary channel activity, e.g., idle and busy periods. Recently, ambient backscatter communication has been introduced, which enables the ST to transmit data to the SR by backscattering ambient signals. Therefore, it is potential to be adopted in the RF-powered CRN. We investigate the performance of RF-powered CRNs with ambient backscatter communication over two scenarios, i.e., overlay and underlay CRNs. For each scenario, we formulate and solve the optimization problem to maximize the overall transmission rate of the secondary system. Numerical results show that by incorporating such two techniques, the performance of the secondary system can be improved significantly compared with the case when the ST performs either harvest-then-transmit or ambient backscatter technique.

This paper studies the data scheduling and admission control problem for a backscatter sensor network (BSN). In the network, instead of initiating their own transmissions, the sensors can send their data to the gateway just by switching their antenna impedance and reflecting the received RF signals. As such, we can reduce remarkably the complexity, the power consumption, and the implementation cost of sensor nodes. Different sensors may have different functions, and data collected from each sensor may also have a different status, e.g., urgent or normal, and thus we need to take these factors into account. Therefore, in this paper, we first introduce a system model together with a mechanism in order to address the data collection and scheduling problem in the BSN. We then propose an optimization solution using the Markov decision process framework and a reinforcement learning algorithm based on the linear function approximation method, with the aim of finding the optimal data collection policy for the gateway. Through simulation results, we not only show the efficiency of the proposed solution compared with other baseline policies, but also present the analysis for data admission control policy under different classes of sensors as well as different types of data.

In this paper, we introduce a new model for RF-powered cognitive radio networks with the aim to improve the performance for secondary systems. In our proposed model, when the primary channel is busy, the secondary transmitter is able either to backscatter the primary signals to transmit data to the secondary receiver or to harvest RF energy from the channel. The harvested energy then will be used to transmit data to the receiver when the channel becomes idle. We first analyze the tradeoff between backscatter communication and harvest-then-transmit protocol in the network. To maximize the overall transmission rate of the secondary network, we formulate an optimization problem to find time ratio between taking backscatter and harvest-thentransmit modes. Through numerical results, we show that under the proposed model can achieve the overall transmission rate higher than using either the backscatter communication or the harvest-then-transmit protocol.

Google's Project Loon [1] was launched in 2013 with the aim of providing Internet access to rural and remote areas. In the Loon network, balloons travel around the Earth and bring access points to the users who cannot connect directly to the global wired Internet. The signals from the users will be transmitted through the balloon network to the base stations connected to the Internet service provider (ISP) on Earth. The process of transmitting and receiving data consume a certain amount of energy from the balloon, while the energy on balloons cannot be supplied by stable power source or by replacing batteries frequently. Instead, the balloons can harvest energy from natural energy sources, e.g., solar energy, or from radio frequency energy by equipping with appropriate circuits. However, such kinds of energy sources are often dynamic and thus how to use this energy efficiently is the main goal of this paper. In this paper, we study the optimal energy allocation problem for the balloons such that network performance is optimized and the revenue for service providers is maximized. We first formulate the stochastic optimization problem as a Markov decision process and then apply a learning algorithm based on simulation-based method to obtain optimal policies for the balloons. Numerical results obtained by extensive simulations clearly show the efficiency and convergence of the proposed learning algorithm.

We study the performance optimization problem for a cognitive radio network with radio frequency (RF) energy harvesting capability for secondary users. In such networks, the secondary users are able to not only transmit packets on a channel licensed to a primary user when the channel is idle, but also harvest RF energy from the primary users' transmissions when the channel is busy. Specifically, we propose a system model where the secondary users are able to cooperate to maximize the overall network throughput through sensing a set of common channels. We first consider the case where the secondary users cooperate in a TDMA fashion and propose a novel solution based on a learning algorithm to find optimal channel access policies for the secondary users. Then, we examine the case where the secondary users cooperate in a decentralized manner and we formulate the cooperative decentralized optimization problem as a decentralized partially observable Markov decision process (DEC-POMDP). To solve the cooperative decentralized stochastic optimization problem, we apply a decentralized learning algorithm based on the policy gradient and the Lagrange multiplier method to obtain optimal channel access policies. Extensive performance evaluation is conducted and it shows the efficiency and the convergence of the learning algorithms.

This paper considers an admission control problem for a mobile cloud computing hotspot with a cloudlet. We first formulate the admission control problem as a Markov decision process (MDP). The objective is to maximize the average reward in terms of revenue for cloudlet service providers. However, the MDP could suffer from the complexity problem (i.e., curse of dimensionality). Therefore, we apply the simulation-based algorithm to obtain the optimal policy for the MDP. The algorithm can estimate the performance measure to update the policy gradient in an online fashion. The performance evaluation of the proposed algorithm uses parameters setting profiled from real mobile applications. The extensive simulation results clearly show the convergence and the efficiency of the proposed algorithm.

In this paper, we consider a decentralized wireless communication system with wireless energy transfer capability. We aim to minimize the total number of packets waiting at wireless nodes for the whole system. We first formulated the optimization problem as a decentralized partially observable Markov decision process (DEC-POMDP). To solve an optimization problem with constraints, we applied the Lagrangian multiplier and the policy gradient method. In addition, to reduce the complexity of DEC-POMDP, we proposed a decentralized online learning algorithm with minimum communication among the wireless nodes. Under appropriate conditions, we showed that the proposed algorithm converges to a local optimal solution. The simulation results clearly showed the convergence as well as the efficiency of the proposed algorithm.

This paper proposes a model and a mechanism for cooperative bidding of wireless data transmission and energy transfer in a decentralized wireless communication system. We aim to minimize the total number of packets waiting at wireless nodes and the total bid prices from the nodes. We first formulated the optimization problem as a decentralized partially observable Markov decision process (DEC-POMDP) with constraints for wireless nodes and then proposed a decentralized learning algorithm to obtain an optimal policy for the DEC-POMDP. Through simulations, we showed the convergence as well as the significantly better performance of the proposed algorithm compared with other algorithms.

Radio frequency (RF) energy harvesting is a promising technique to sustain operations of wireless networks. In a cognitive radio network, a secondary user can be equipped with RF energy harvesting capability. In this paper, we consider such a network where the secondary user can perform channel access to transmit a packet or to harvest RF energy when the selected channel is idle or occupied by the primary user, respectively. We present an optimization formulation to obtain the channel access policy for the secondary user to maximize its throughput. Both the case that the secondary user knows the current state of the channels and the case that the secondary knows the idle channel probabilities of channels in advance are considered. However, the optimization requires model parameters (e.g., the probability of successful packet transmission, the probability of successful RF energy harvesting, and the probability of channel to be idle) to obtain the policy. To obviate such a requirement, we apply an online learning algorithm that can observe the environment and adapt the channel access action accordingly without any a prior knowledge about the model parameters. We evaluate both the efficiency and convergence of the learning algorithm. The numerical results show that the policy obtained from the learning algorithm can achieve the performance in terms of throughput close to that obtained from the optimization.

Abstract—We consider the joint resource allocation and admis-sion control problem for Orthogonal Frequency-Division Multi-ple Access (OFDMA)-based femtocell networks. We assume that Macrocell User Equipments (MUEs) can establish connections with Femtocell Base Stations (FBSs) to mitigate the excessive cross-tier interference and achieve better throughput. A cross-layer design model is considered where multiband opportunistic scheduling at the Medium Access Control (MAC) layer and admission control at the network layer working at different time-scales are assumed. We assume that both MUEs and Femtocell User Equipments (FUEs) have minimum average rate constraints, which depend on their geographical locations and their application requirements. In addition, blocking probability constraints are imposed on each FUE so that the connections from MUEs only result in controllable performance degradation for FUEs. We present an optimal design for the admission control problem by using the theory of Semi-Markov Decision Process (SMDP). Moreover, we devise a novel distributed femtocell power adaptation algorithm, which converges to the Nash equilibrium of a corresponding power adaptation game. This power adaptation algorithm reduces energy consumption for femtocells while still maintaining individual cell throughput by adapting the FBS power to the traffic load in the network. Finally, numerical results are presented to demonstrate the desirable operation of the optimal admission control solution, the significant performance gain of the proposed hybrid access strategy with respect to the closed access counterpart, and the great power saving gain achieved by the proposed power adaptation algorithm. Index Terms—Femtocell network, admission control, Markov decision process, blocking probability, channel assignment.

ABSTRACT Together with an explosive growth of the mobile applications and emerging of cloud computing concept, mobile cloud computing (MCC) has been introduced to be a potential technology for mobile services. MCC integrates the cloud computing into the mobile environment and overcomes obstacles related to the performance (e.g., battery life, storage, and bandwidth), environment (e.g., heterogeneity, scalability, and availability), and security (e.g., reliability and privacy) discussed in mobile computing. This paper gives a survey of MCC, which helps general readers have an overview of the MCC including the definition, architecture, and applications. The issues, existing solutions, and approaches are presented. In addition, the future research directions of MCC are discussed. Copyright © 2011 John Wiley & Sons, Ltd.

In this paper, we consider the admission control problem for hybrid access in OFDMA-based femtocell networks. We assume that Macrocell User Equipments (MUEs) can establish connections with Femtocell Base Stations (FBSs) to improve their QoSs. Both MUEs and Femtocell User Equipments (FUEs) have minimum rate requirements, which depend on their geographical locations and maybe their running applications. In addition, blocking probability constraints are imposed on each FUE so that connections from MUEs only result in controllable performance degradation for FUEs. We show how to formulate the admission control problem as a Semi-Markov Decision Process (SMDP) and present a Linear Programming (LP) based solution approach. Moreover, we develop a novel femtocell power adaptation algorithm, which can be implemented in a distributed manner jointly with the proposed admission control scheme. This power adaptation algorithm enables to achieve better cell throughput and more energy-efficient operation of the femtocell network considering the heterogeneity of traffic load in the network. Finally, numerical results are presented to illustrate the desirable performance of the optimal admission control solution and the significant throughput and power saving gains of the proposed cross-layer solution.

We consider an admission control problem and adaptive resource allocation for running mobile applications on a cloudlet. We formulate an optimization problem for dynamic resource sharing of mobile users in mobile cloud computing (MCC) hotspot with a cloudlet as a semi-Markov decision process (SMDP). SMDP is transformed into a linear programming (LP) model and it is solved to obtain an optimal solution. In the optimization model, the quality of service (QoS) for different classes of mobile user is taken into account under resource constraints (i.e., bandwidth and server). The numerical results are presented to illustrate that the proposed admission control scheme can achieve a desirable performance and improve throughput of an MCC hotspot significantly.