Federated Self-Supervised Learning (FSSL) is a promising paradigm for extracting robust representations from decentralized unlabeled data. However, its effectiveness is often hindered by non-IID data distributions and label scarcity, which cause model divergence and limit generalization. In this paper, we propose Federated Self-Supervised and Global-Personalized Collaborative Learning (FedGP), a novel framework designed to bridge the gap between global knowledge integration and local client adaptation. The core of FedGP is the Collaborative Knowledge Anchoring (CKA) mechanism, which utilizes adaptive regularization to anchor shared global knowledge while enabling personalized refinement on local data. By dynamically balancing collaborative risks and local empirical losses via learnable coefficients, FedGP ensures stable convergence in heterogeneous environments. Extensive evaluations on multiple benchmarks, including a real-world private Flora dataset, demonstrate that FedGP consistently outperforms state-of-the-art FSSL methods. Our results confirm that FedGP achieves high-quality representation learning with significantly reduced communication overhead and annotation dependency, providing a scalable solution for privacy-preserving decentralized systems.

The deep integration of Internet of Vehicles (IoV) and edge computing technologies brings new requirements for the Vehicular Edge Computing (VEC) information system security evaluation. Facing the two core problems of resource-constrained scenarios and dynamic security evaluation, the GFCIV-CGTOPSIS model for VEC information system dynamic security evaluation is proposed. In the model, subjective and objective evaluation indexes are considered and calculated separately to improve operability. An improved grey correlation F-statistics clustering and index validity combination index screening (GFCIV) method is proposed in order to improve the operational efficiency of the traditional grey F-statistics rough set (GFRS) index screening method. The CRITIC method is used to determine the subjective and objective comprehensive index set weights, and the grey TOPSIS method is used to realize the dynamic security evaluation. Experimental results demonstrate that the GFCIV-CGTOPSIS model, compared to the pre-improved GFRS-CGTOPSIS model, achieves reduced index tree redundancy and efficient dynamic security evaluation while exhibiting less information loss and lower evaluation result deviation due to index screening.

Network traffic prediction plays a crucial role in optimizing resource allocation, mitigating congestion, and enhancing cybersecurity in large-scale Internet of Things systems. However, the multiscale temporal patterns, complexity, and nonlinearity of traffic data pose significant challenges for accurate modeling and prediction. To address these limitations, this article proposes a novel approach that combines spectrum-aided traffic decomposition with the deep learning (DL) method for network traffic prediction. Specifically, spectrum-aided traffic decomposition is performed using the seasonal-trend decomposition using loess algorithm to decompose traffic data into interpretable seasonal, trend, and residual components, where the seasonal period is dynamically optimized through frequency domain analysis. Each component is then modeled separately using a DL architecture, which is the gated recurrent units-based sequence-to-sequence model with an attention mechanism. This allows the model to effectively capture multiscale temporal dependencies, long-term relationships, and complex patterns, thereby enhancing prediction accuracy. The predictions from each decomposed component are subsequently ensembled to obtain the final prediction value. Experimental results show that our method achieves significant improvements over other methods across four real-world datasets, reducing mean square error and mean absolute error by an average of 97.6% and 84.7%, respectively.

In the era of big data and IoT, personalized federated learning (pFL) addresses privacy challenges by keeping training on-device and transmitting only parameter updates. This approach allows each client to incorporate their unique data characteristics, effectively tailoring the global model to diverse user needs. However, as pFL leverages personalized updates that capture individual client experiences, it faces the inherent risk of exposing sensitive information during the aggregation process. Global federated methods like FedAvg further exacerbate this issue when the non-IID assumption is violated across clients, as they struggle to maintain performance across varied data distributions, often leading to suboptimal results. To tackle this, we propose fine-grained Privacy-aware Parameter Coaching method for personalized Federated learning(PPCFed). We propose a dynamic privacy matrix that quantifies layer-wise privacy leakage risks and adaptively reweights inter-client knowledge transfer. This matrix acts as a fine-grained privacy-aware controller, enabling clients to selectively assimilate insights from others without additional overhead while actively suppressing high-risk information flows. Our method balances privacy and performance, showing improved results in heterogeneous FL and pFL environments.

Abstract In practical scenarios, security events triggered by abnormal network traffic often result from the collective behavior of multiple data streams, embodying group security events with collective characteristics. Existing research methods, focusing on individual data streams, lack a macroscopic analysis and struggle with challenges of analyzing massive, imbalanced data sets. To address these challenges, this paper adopts a multi-instance learning approach, mapping multiple data streams into a bag with a coarse-grained approach, where each bag corresponds to a security event label and each data stream represents an instance. We propose a multi-instance network traffic conversion method, Bag2Image, which transforms temporal multi-instance network traffic data into image representations, preserving the spatio-temporal characteristics of instances within the bag through image channels and pixels. This strategy allows the network security event prediction task to be approached as an image classification problem, leveraging advanced image classification techniques for prediction. Our cross-experiments with six advanced multi-instance learning (MIL) algorithms and six different classification models demonstrate the superior performance of our method on both the UNSW-NB15 dataset and a private dataset. Specifically, our method achieved the highest F1 scores of 77.9% and 74.4% on these datasets, respectively, representing improvements of 4.1% and 13.5% over the second-best MIL algorithm. The recall rates also saw increases of 4.1% and 13.2%, respectively.

Quantifying the deviation of testing samples from known benign traffic in the form of novelty scores is essential for identifying new malicious traffic and detecting concept drifts of benign traffic. Most existing solutions generate novelty scores based on the outputs from the supervised neural network’s last layers. However, the performance of these supervised techniques is significantly compromised in the unbalanced dynamic interconnected specialized IoT network traffic, such as electronic power grid. To solve this challenge, we investigate an improved novelty score calculation approach. It employs the targeted distance loss to ensure that different known classes form class-specific, dense clusters within the embedding space outputted by the neural network. The minimum distance between the testing sample and the centers of each known class is utilized as the novelty score. Extensive comparison experiments on public datasets and real Electronic Power Grid traffic demonstrate that our approach outperforms existing supervised techniques. As the imbalance in known class distributions increases, our proposed method consistently achieves higher AUROC and AUPR than baselines, with a reduced FPR95. This study analyzes imbalanced known class distributions’ negative impact on novelty detection and provides practical improvement.

Due to the difficulty in predefining novel attack patterns and the scarcity of sufficient malicious training samples in specialized Internet of Things (IoT) scenarios, the focus of IoT intrusion detection researchers has shifted toward anomaly-based techniques like Autoencoder. These techniques detect attacks based on the degree of deviation and rely minimally on malicious samples. However, current machine learning (ML) and deep learning (DL) implementations lack explainability. Although some methods provide post-hoc interpretations, they are limited and partial, failing to reflect the entire decision-making process. To address this challenge, we investigate an explainable anomaly-based intrusion detection system (IDS) that translates the inference process of the Autoencoder into the high-fidelity allow-list rule library, thereby balancing the detection capability and interpretability. First, we assume that benign traffic follows a complex global distribution composed of several irrelevant local distributions. The clustering algorithm is performed in an extended feature space consisting of reconstruction loss and embeddings to decompose local distributions. Then, we deploy an approach based on Gradient Ascent to explore the boundary rules of each local distribution. The allow-list rule library that reflects Autoencoder’s inference process can be constructed by merging these boundary rules. Comprehensive evaluation experiments demonstrate that the extracted allow-list rule library accurately reproduces Autoencoder’s inference process and effectively detects IoT intrusions.

The applications of unmanned aerial vehicle (UAV) swarms have effectively managed complex tasks, expanded operational areas, and executed missions either autonomously or in collaboration. Group authentication for UAV swarms is essential to ensure the authenticity, integrity, and confidentiality of sensitive data. However, the highly dynamic nature of UAV networks makes the group authentication process challenging. Traditional centralized group authentication methods are susceptible to a single point of failure. In distributed group authentication schemes, issues such as key-escrow attacks, inefficiencies in cross-domain authentication, and challenges in domain isolation within identity data management continue to persist. This paper explores a scalable and efficient authentication scheme assisted by blockchain and group decentralized identifiers (BGAS), which utilizes group decentralized identifiers (GDIDs) to extend decentralized identifiers (DIDs) for group authentication. Security and performance analyses demonstrate that our protocol offers robust security against various common types of attacks and is more efficient than current authentication protocols in terms of storage costs and group authentication time.

The development of the Internet of Vehicles (IoV) has significantly enhanced connectivity and cooperation among road entities, leading to a more efficient, economical, and safer intelligent transportation system (ITS). However, this increased connectivity also exposes vehicles to a growing risk of cybersecurity threats through intravehicle and intervehicle networks. To secure IoV networks, many studies have focused on using intrusion detection systems (IDSs) based on deep learning methods to effectively detect cyber-attacks due to their ability to learn from large-scale data. Nonetheless, most existing IDSs rely on expert knowledge to manually design features, resulting in difficulties adapting to evolving attacks and information loss. To mitigate these limitations, this article presents a tokenization representation and attention mechanism-based convolutional neural network-bidirectional long short-term memory (CNN-BiLSTM) intrusion detection method. For feature extraction, we tokenize original traffic using natural language processing technique to represent discrete hexadecimal bytes as words, thus alleviating the need for manual feature design and allowing for the direct extraction of sequence patterns. For classification, we incorporate an attention mechanism into the CNN and BiLSTM architectures to enhance the accuracy of intrusion detection by focusing on critical information and capturing sequential patterns. The effectiveness of the proposed IDS is evaluated in both intravehicle and intervehicle network scenarios. Experimental results show that our method can detect various types of attacks with 100% accuracy on the Car-Hacking data set for the intravehicle network scenario. In the intervehicle network scenario utilizing the CICIoT2023 data set, our approach also achieves a high accuracy of 98%, outperforming existing methods.

Based on the rapid development of edge computing and resource-constrained characteristics, new requirements for edge information system security evaluation are proposed. Oriented to the resource-constrained scenarios of edge information systems and the distribution characteristics of raw data for security evaluation, the adaptability of existing models is improved. The improved Principal Component Analysis (PCA-S method) based on Spearman’s Coefficient is proposed for index screening. In order to further improve the screening effect and reduce the resource consumption of the security evaluation model, the PCA-S method and the Distinction Degree Screening method are combined by taking the intersection, and the PCA-SDC combination screening method is proposed. Through the PCA-SDC index combination screening, the Coefficient of Variation weighting, and the Fuzzy Comprehensive Evaluation, the PSDC-CVF edge information system security evaluation model is finally formed. In order to assess the effect of index screening and security evaluation, three indicators, namely, the improved Average Quantity of Information Change Degree, the Average Information Contribution Change Degree, and the Fuzzy Evaluation Deviation Degree, are proposed. Through experiments, the improved PCA-S method and the combination screening PCA-SDC method are sequentially proved to be well adapted and effective in the index screening process of edge information system security evaluation. It is also verified that the PSDC-CVF model reduces the resource consumption compared with the traditional model and better balances the model energy consumption and performance.

Internet of Things (IoT) blockchains are characterized by the difference in storage resources, computing power, and high dynamic of node topologies. In view of the demand for short-time and high-concurrency execution of events in IoT blockchain transaction scenarios, we propose a lightweight segmented storage based on parallel execution for IoT blockchains. This method breaks through the technical difficulties of 2-D ledger design for event parallelism, de-blocking lightweight synchronization of cross-domain data, and global consistency maintenance of asynchronous ledger, and achieves parallel execution of events and lightweight segmented storage of resource-constrained nodes. First, the event cumulative verification and differential generation algorithm is proposed to reduce the redundancy propagation in event distribution and voting. Second, the effective concurrent consensus confirmation mechanism of events is proposed to implement the distributed batch confirmation of valid events, which guarantees the low-latency confirmation of events. Furthermore, the data lightweight segmented storage mechanism is proposed to implement the data lightweight segmented storage of the resource-constrained nodes combined with the design of the LBPS Chain ledger structure. Simulation results show that the LBPS Chain outperforms LDV, Layerchain, and GpDB by 85%, 88%, and 52% in terms of storage cost. Compared to the Hashgraph, our method effectively increases the throughput by 89% and has scalability in reducing the storage cost and increasing throughput.

Sharded blockchain offers scalability, decentralization, immutability, and linear improvement, making it a promising solution for addressing the trust problem in large-scale collaborative IoT. However, a high proportion of cross-shard transactions can severely limit the performance of decentralized blockchain. Furthermore, the dynamic assemblage characteristic of collaborative sensing in sharded blockchain is often ignored. To overcome these limitations, we propose HMMDShard, a dynamic blockchain sharding scheme based on the Hidden Markov Model. HMMDShard leverages fine-grained blockchain sharding and fully embraces the dynamic assemblage characteristic of IoT collaborative sensing. By integrating the Hidden Markov Model, we achieve adaptive dynamic incremental updating of blockchain shards, effectively reducing cross-shard transactions across all shards. We conduct a comprehensive analysis of the security issues and properties of HMMDShard, and evaluate its performance through the implementation of a system prototype. The results demonstrate that HMMDShard significantly reduces the proportion of cross-shard transactions and outperforms other baselines in terms of system throughput and transaction confirmation latency.

The frequent attacks show that no information system is absolutely safe and the security capabilities are relative. Security resilience becomes a complementary priority for improving information systems’ continuous service and security capabilities in the face of attacks, such as unknown vulnerabilities and backdoors. Endogenous security defense technology has become an important research aspect to improve the security resilience of information systems. However, there are some limitations in the research of the information system security resilience evaluation model, such as lacking indexes to characterize the system security resilience under an attack environment. In this article, a security resilience enhancement strategy based on dynamic defense is constructed to improve the security performance of the system through IP port hopping and attack surface conversion. For the adversarial behaviors of attackers and defenders, we propose a security resilience metric framework based on the evolution of attack and defense scenarios, which is evaluated using a resilient security evaluation model based on the fuzzy Choquet integral. In the model, the weights of evaluation indicators are calculated based on the decision-making trial and evaluation laboratory method. The 2-addable fuzzy measures of each indicator are calculated secondarily. Then the security performance of the system is calculated using the fuzzy Choquet integral. Absorptive capacity, adaptive capacity, and resilience factor are proposed to better supervise the metric framework’s validity. Finally, four groups of control cases were created by building the Web service system after the endogenous security transformation as the experimental simulation scenario. Experimental simulation results show the superiority of the proposed metric model.

Location-based service recommendations usually need to collect and analyze the location information of trajectories generated by users with smart phones or wearable devices. It is easy to cause the location privacy leaks. The current location privacy protection methods usually confuse the adversary by adding fake locations into real trajectories to achieve the goal of privacy protection. However, these methods fail to consider the utility of user trajectories in service recommendations. In this paper, we propose a privacy-preserving trajectory generation algorithm based on service semantic similarity. To improve the utility of privacy-preserving trajectories for service recommendations, the algorithm constructs a series of service semantic grid maps and generates fake individuals with privacy-preserving trajectories considering both utility and privacy. Simulation results show that the proposed algorithm can effectively hide the locations in real trajectories of individuals and obtain higher effectiveness for service recommendations.

With the popularity and application of the Industrial Internet of Things (1IoT), device anomaly detection is considered as one of the important challenges in IloT implementation. However, the privacy sensitivity of device data and the high heterogeneity of IloT devices make it impossible for traditional schemes to achieve efficient, accurate, and privacy-protected device anomaly detection in IloT networks. In this study, we propose an intelligent anomaly detection architecture for IloT networks based on federated optimization algorithms and deep learning (DL). In particular, an online, adaptive, and semi-supervised device anomaly detection model is designed, and a heterogeneity-aware federated learning algorithm, called Clustered-FedProx, is presented. The Clustered-FedProx algorithm considers the differences in computational power and data statistical distribution among IloT devices, whereby multiple devices can be coordinated to train a global DL model in highly heterogeneous networks. Simulation results show that the proposed scheme can achieve more stable and accurate performance than conventional schemes.

To enhance the dependability of lightweight blockchain, this paper presents a privacy protection mechanism in open networks with a secure ledger for each cell of society. We propose a swarm intelligence scheme called hierarchical RAFT (HRAFT), which accumulates valuable workloads about the credibility of all the nodes to sort for selecting candidates. Thus, the scheme achieves high-performance decentralization by progressive evolution consensus.

With the advent of the Internet of Things (IoT), crowdsensing, as a new emerging application of the IoT that employs ubiquitous mobile users with smartphones for data collection and processing, has further deepened our knowledge. However, the problems of the current crowdsensing systems regarding system security, user privacy, and user payment (UP) raise serious privacy and security concerns, which affect participants’ adoption of the system. The Blockchain technology allows for nondeterministic multiple parties to interact with each other anonymously in a network that is not fully trusted. In this article, we propose a new decentralized crowdsensing system, calledCrowdHB. Unlike other blockchain-based crowdsensing systems,CrowdHBadopts a hybrid blockchain architecture and uses smart contracts to achieve location privacy preservation and ensure data quality while improving the system performance. Furthermore, to optimize task assignments to mobile users, we propose a location privacy-preserving optimization mechanism (LPPOM) and the approach of consistency optimization (ACO) to achieve a tradeoff between user privacy and system performance. The extensive experimental results show that the proposedCrowdHBoutperforms the other crowdsensing systems in terms of task success rate and performance for a large number of mobile users and tasks.

The edge computing node plays an important role in the evolution of the artificial intelligence-empowered Internet of things (AIoTs) that converge sensing, communication, and computing to enhance wireless ubiquitous connectivity, data acquisition, and analysis capabilities. With full connectivity, the issue of data security in the new cloud-edge-terminal network hierarchy of AIoTs comes to the fore, for which blockchain technology is considered as a potential solution. Nevertheless, existing schemes cannot be applied to the resource-constrained and heterogeneous IoTs. In this paper, we consider the blockchain design for the AIoTs and propose a novel classified ledger framework based on lightweight blockchain (CLF-LB) that separates and stores data rights at the source and enables a thorough data flow protection in the open and heterogeneous network environment of AIoT. In particular, CLF-LB divides the network into five functional layers for optimal adaptation to AIoTs applications, wherein an intelligent collaboration mechanism is also proposed to enhance the across-layer operation. Unlike traditional full-function blockchain models, our framework includes novel technical modules, such as block regenesis, iterative reinforcement of proof-of-work, and efficient chain uploading via the system-on-chip system, which are carefully designed to fit the cloud-edge-terminal hierarchy in AIoTs networks. Comprehensive experimental results are provided to validate the advantages of the proposed CLF-LB, showing its potentials to address the secrecy issues of data storage and sharing in AIoTs networks.

In the process of sharing data, the costless replication of electric energy data leads to the problem of uncontrolled data and the difficulty of third-party access verification. This paper proposes a controlled sharing mechanism of data based on the consortium blockchain. The data flow range is controlled by the data isolation mechanism between channels provided by the consortium blockchain by constructing a data storage consortium chain to achieve trusted data storage, combining attribute-based encryption to complete data access control and meet the demands for granular data accessibility control and secure sharing; the data flow transfer ledger is built to record the original data life cycle management and effectively record the data transfer process of each data controller. Taking the application scenario of electric energy data sharing as an example, the scheme is designed and simulated on the Linux system and Hyperledger Fabric. Experimental results have verified that the mechanism can effectively control the scope of access to electrical energy data and realize the control of the data by the data owner.

In this article, a scheme to detect both clone and Sybil attacks by using channel-based machine learning is proposed. To identify malicious attacks, channel responses between sensor peers have been explored as a form of fingerprints with spatial and temporal uniqueness. Moreover, the machine-learning-based method is applied to provide a more accurate authentication rate. Specifically, by combining with edge devices, we apply a threshold detection method based on channel differences to provide offline training sample sets with labels for the machine learning algorithm, which avoids manually generating labels. Therefore, our proposed scheme is lightweight for resource constrained industrial wireless devices, since only an online-decision making is required. Extensive simulations and experiments were conducted in real industrial environments. Both results show that the authentication accuracy rate of our strategy with an appropriate threshold can achieve 84% without manual labeling.

In view of the difficulty in determining reasonably evaluation indicator system in the information security certification and accreditation work, an information security evaluation framework supporting the adaptation of measurement models is proposed. And the mapping-based information security evaluation indicator construction rule and a measurement model library are established. The optimal measurement model is adapted according to head-to-tail consistency and standard deviation index. Then, the evaluation model relies on the information security index feedback algorithm based on probability iteration to adjust the evaluation indicator system for more reasonableness. This paper provides a model reference for information security certification.

We study the optimized dynamic antenna parameters configuration for the 3D massive multiple-input multiple- out (MIMO) system in a heterogeneous network (HetNet) with overlaid macrocells and smallcells. In particular, we propose a deep reinforcement learning (DRL) approach to jointly adjust three key antenna parameters, namely, downtilt angle, vertical and horizontal half-power beamwidths of the macro base stations (mBSs) automatically in a dynamic environment with strong user mobility. More specifically, employing the gridded user location information (ULI), we propose a novel mix Q-learning algorithm to efficiently address the challenging joint optimization problem, which integrates a parallel hyper-parameter updating mechanism in dual sub-networks and a technique of prioritized replay buffer. The resultant neural network can efficiently learn the historical experience in an online fashion and achieve excellent sum-rate performance with affordable trials. Moreover, thanks to the proposed gridded ULI, our DRL-empowered antenna configuration framework can easily fit various HetNet deployments with variable user densities. Numerical results show that the average weighted sum-rate is increased by 4.59 bit/s/Hz, and the average performance improvement is up to 24.82% as compared to the reference scheme without gridded ULI.

In 1975, J. E. Mazo showed the potential faster‐than‐Nyquist (FTN) gain of the single‐carrier binary signal. If the inter‐symbol interference is eliminated by an optimal detector, FTN single‐carrier binary signal can transmit 24.7% more bits than the Nyquist signal without loss of bit error rate performance, which is known as the Mazo limit. In this study, the authors apply sphere decoder (SD) with box optimisation (BO) to reduce inter‐carrier interference (ICI) in the FTN non‐orthogonal frequency division multiplexing (FTN–NOFDM) system. Compared with the commonly used SD algorithm, SD with BO can achieve the same ICI cancellation performance, and the computational complexity is significantly reduced. When the bandwidth compression factor is set to 0.802, the transmission rate of FTN–NOFDM is 24.7% faster than the Nyquist rate, and quadrature phase shift keying (QPSK)‐modulated FTN–NOFDM has almost the same performance as QPSK‐modulated orthogonal frequency division multiplexing (OFDM), which agrees well with Mazo limit. In the simulation, the QPSK‐modulated FTN–NOFDM with outperforms 16‐quadrature amplitude modulation (16‐QAM)‐modulated OFDM by about 1.5 dB, and the 16‐QAM‐modulated FTN–NOFDM with and outperforms 64‐ and 256‐QAM‐modulated OFDM by about 1.5 and 2 dB, respectively.

The fusion of edge computing and artificial intelligence (AI) technology is a key enabler for the smart Internet of Things (IoT). However, these two emerging paradigms face many issues for their integration, such as data storage structure, model generation algorithms, and cloud-edge collaboration mechanisms. Moreover, edge computing is not ready for supporting AI and can be enabled to support AI via some basic network functions related to Quality of Experience (QoE), such as passive computation offloading and content caching. In this article, we present an intelligent cooperative edge (ICE) computing in IoT networks to achieve a complementary integration of AI and edge computing. The AI-related modules of edge computing are redesigned for distributing AI's core functions from the cloud to the edge. IoT-generated data are differentiated as user-private data preserved locally in IoT devices, edge-private data isolated on the edge and public data uploaded to the cloud. Therefore, a cloud-scale machine learning model can be generated, followed by privacy-preserving transfer learning running on each edge, which also has data updated more frequently that enables the model's incremental learning. The model distribution is accomplished through lightweight deployment pipelines consisting of cloud compression and edge reconstruction. Conversely, some key issues of edge computing, such as the computation offloading and content caching, achieve a better solution using the localized AI. We perform the prototype-based evaluation, which indicates that the ICE computing architecture enables a benign combination of AI and edge computing.

We consider an ultra-dense network (UDN), where serious interference may exist due to the densely deployed base stations (BSs) and user equipment (UE). Noting that the conventional interference management (IM) scheme is not readily applicable, in this paper, we propose a novel interference topology management (ITM) scheme to achieve low-complexity IM in UDNs. The proposed ITM scheme consists of two stages, namely, the optimized BS clustering stage and the decentralized UE-cluster association stage. In the first stage, a novel unsupervised learning-based BS clustering algorithm is proposed, which outperforms the conventional clustering method. Then, a decentralized UE-cluster association algorithm is proposed, which doesn't need to exchange the channel information among clusters, and significantly save system overheads as compared to the centralized solutions. The results show that our ITM scheme can improve the system throughput at lower complexity as compared to other related schemes.

Mobile edge computing (MEC) has the potential to enable computation-intensive applications in 5G networks. MEC can extend the computational capacity at the edge of wireless networks by migrating the computation-intensive tasks to the MEC server. In this paper, we consider a multi-user MEC system, where multiple user equipments (UEs) can perform computation offloading via wireless channels to an MEC server. We formulate the sum cost of delay and energy consumptions for all UEs as our optimization objective. In order to minimize the sum cost of the considered MEC system, we jointly optimize the offloading decision and computational resource allocation. However, it is challenging to obtain an optimal policy in such a dynamic system. Besides immediate reward, Reinforcement Learning (RL) also takes a long-term goal into consideration, which is very important to a time-variant dynamic systems, such as our considered multi-user wireless MEC system. To this end, we propose RL-based optimization framework to tackle the resource allocation in wireless MEC. Specifically, the Q-learning based and Deep Reinforcement Learning (DRL) based schemes are proposed, respectively. Simulation results show that the proposed scheme achieves significant reduction on the sum cost compared to other baselines.

We study the detection of misbehavior in a Gaussian relay system, where the source transmits information to the destination with the assistance of an amplify-and-forward relay node subject to unreliable channel state information (CSI). The relay node may be potentially malicious and corrupt the network by forwarding garbled information. In this situation, misleading feedback may take place, since reliable CSI is unavailable at the source and/or the destination. By classifying the action of the relay as detectable or undetectable, we propose a novel approach that is capable of coping with any malicious attack detected and continuing to work effectively in the presence of unreliable CSI. We demonstrate that the detectable class of attacks can be successfully detected with a high probability. Meanwhile, the undetectable class of attacks does not affect the performance improvements that are achievable by cooperative diversity, even though such an attack may fool the proposed detection approach. We also extend the method to deal with the case in which there is no direct link between the source and the destination. The effectiveness of the proposed approach has been validated by numerical results.

Fog computing extends the concept of cloud computing to the edge of network to relieve performance bottleneck and minimize data analytics latency at the central server of a cloud. It uses edge nodes directly to perform data input and data analysis. In public opinion analysis system, edge nodes that collect opinions from users are responsible for some data filtering jobs including sentiment analysis. Therefore, it is crucial to find suitable algorithm that is lightweight in operation and accurate in predictive performance. In this paper, we focus on Chinese sentiment analysis job in fog computing environment and propose a non‐task‐specific method called Channel Transformation Based Convolutional Neural Network (CTBCNN) for Chinese sentiment classification, which uses a new structure called channel transformation based (CTB) convolutional layer to enhance the ability of automatic feature extraction and applies global average pooling layer to prevent overfitting. Through experiments and analysis, we show that our method do achieve competitive accuracy and it is convenient to apply this method to different cases in operation.

Non-orthogonal multiple access (NOMA) has been investigated recently as a candidate radio access technology for the fifth generation (5G) networks due to its high spectrum efficiency (SE). As green radio which focuses on energy efficiency (EE) becomes an inevitable trend, energy efficient design is becoming more and more important. In this paper, we focus on energy efficient resource allocation problem in multi-user downlink NOMA system with the aim to optimize subchannel assignment and power allocation to maximize the system EE. We propose a novel low-complexity suboptimal subchannel assignment algorithm and obtain the optimal power allocation coefficients among subchannel multiplexed users. To further improve the system EE, unequal power allocation across subchannels (UPAAS) scheme including an optimal solution and a suboptimal Dinkelbach-like algorithm is studied. Simulation results show the effectiveness of our proposed resource allocation algorithms.

This paper deals with the detection of malicious attacks in a Gaussian two-hop relay network, in which the source transmits information to the destination with the assistance of an amplify and forward (AF) relay node with unreliable channel state information (CSI). We consider the case where the potentially malicious relay node attempts to corrupt the network by malicious forwarding. In addition, dishonest feedback results in reliable CSI is not available at the source and/or destination. By modeling the relay behavior with two kinds of operations, namely detectable and undetectable operations, we propose a novel detection method which can deal with any detectable malicious attacks and works effectively in the presence of unreliable CSI. For the detectable operation, it is shown that the malicious attacks can be, with high probability, successfully detected. Furthermore, our research has shown that, although for the undetectable operation each attack can fool the proposed detection method, these undetectable attacks do not affect the improvements cooperative diversity. The effectiveness of our method has been validated by numerical results obtained by means of computer simulations.

In this paper, we investigate the cache-enabled small cell networks to provide on-demand video services with differential perceptual qualities, i.e., standard definition video (SDV) and high definition video (HDV). As the extension technology of advanced video coding/H.264, scalable video coding is adopted in the considered networks and videos to be transmitted are divided into a base layer (BL) and N enhancement layers (ELs). In our proposed caching protocol, the n-th small cell base station (SBS) caches BLs and the n-th EL of the most popular videos. Depending on the distances between the typical user and SBSs in the observed cluster, the closest SBS is regarded as the serving node (SN) and the others are cooperative nodes (CNs). When SDV is required, the SN will transmit BL of the required video file to the typical user, while SN and CNs can cooperatively transmit BL and ELs to provide superior video quality if HDV is required. Based on the proposed caching and transmission protocol, we derive the expressions of the key performance indicators, i.e., local serving probability, ergodic service rate and service delay. Numerical results validate the theoretical analysis and show the superiority of our proposed scheme compared to the benchmarks.

Automatically identifying rumors from online social media especially microblogging websites is an important research issue. Most of existing work for rumor detection focuses on modeling features related to microblog contents, users and propagation patterns, but ignore the importance of the variation of these social context features during the message propagation over time. In this study, we propose a novel approach to capture the temporal characteristics of these features based on the time series of rumor's lifecycle, for which time series modeling technique is applied to incorporate various social context information. Our experiments using the events in two microblog datasets confirm that the method outperforms state-of-the-art rumor detection approaches by large margins. Moreover, our model demonstrates strong performance on detecting rumors at early stage after their initial broadcast.

In this paper, bootstrap percolation is introduced to control the information propagation for efficient cooperative positioning in wireless networks. Particularly, we obtain a novel linear least square (LLS) estimator for the localization of agent nodes. Exploiting the idea of bootstrap percolation, agent nodes sequentially get activated and estimate their positions with an adaptive location updating rule. The rule is designed to first localize the more reliable agent nodes with at least three connections to the active nodes, and then gradually relax such connection constraints in each iteration so as to localize the agent nodes with fewer connections. Due to the activation characteristic, error propogation can be mitigated and energy is well managed. In addition, taking the uncertainty of the positional information into account, positioning errors can be further reduced. Simulations show that the proposed schemes improve the localization accuracy and use fewer links than traditional methods.

We investigate the secrecy beamforming in two-way relay channels (TWRC) with physical layer network coding (PNC). The multi-antenna relay broadcasts the superimposed signal of two user messages with secrecy beamforming after receiving the signals transmitted by the two legitimate users. We first propose a lower bound of the secrecy sum rate to quantify the secrecy performance of the TWRC with PNC. Because the maximization of the lower bound is non-convex under total power constraint, we propose a joint beamforming and power allocation scheme, in which the problem is successively approximated by several convex semidefinite programs. In order to reduce the complexity, we further propose an suboptimal scheme with closed-form solution. Numerical results indicate that the proposed schemes with PNC achieve much better secrecy sum-rate performance than the traditional AF schemes.

In this paper, the exact closed-form outage probability expression is derived for cognitive relay network with incremental decode-and-forward (IDF) protocol in independent non-identically distributed (i.n.i.d.) Nakagami-m fading channels. The outage performance comparisons are made between IDF and DF protocols. Besides, the impact of channel fading parameters is investigated for both secondary transmission links and interference links. The results show that a significant gain can be made by using IDF protocol, especially when the secondary direct link is in good channel condition. Moreover, the outage performance is dominated by the channel quality of the secondary transmission links and is also impacted by the channel quality of the interference links.

The key of time of arrival (TOA) estimation in ultra wideband (UWB) ranging is to detect the first path (FP). In this paper, we propose an optimized FP detector with the adaptive threshold in the absence of prior channel state information (CSI). In particular, the error information (EI) set is introduced to guide the threshold adjustment and determine the TOA estimate with a novel iterative algorithm. The EI set captures the characteristics of major errors regarding the inappropriate threshold. After the iterative process, the proposed scheme is shown to achieve the asymptotic optimal threshold without large number of repeated pulses. Therefore, the proposed scheme efficiently improves the TOA estimation accuracy as compare to the traditional schemes. Simulation results validate the effectiveness and superiority of the proposed scheme.

In this paper, we consider a K-cell multi-user interference network, where S single-antenna users are selected within each cell to carry out the uplink transmission with their M-antenna home base station, where 2 ≤ S ≤ M <; KS. For the considered scenario, a novel intra-cell performance aware opportunistic interference alignment (OIA) scheme is proposed to mitigate the inter-cell interference while reducing the intra-cell power loss caused by zero-forcing receiving. Unlike the traditional OIA schemes, the proposed scheme reuses the reference signal subspace (RSS) to balance not only the inter-cell interference but also the desired signal power and the intra-cell power leakage. It is shown that the refined selection further improves the achievable sum rate as compared to the existing schemes, and this observation is theoretically analyzed. Finally, numerical results validate that our scheme outperforms the existing uplink OIA schemes.

In this paper, semidefinite relaxation (SDR) technology is exploited for the multiple-symbol detection (MSD) over the multiple-input multiple-output (MIMO) ultra-wideband (UWB) systems. The existing scheme generalized likelihood ratio test (GLRT) MSD jointly detect multiple symbols, however, it entails a complexity of O(2M), where M is the observation window size. To this end, SDR is employed to reformulate the GLRT detection into a semidefinite programming (SDP) model, and two detectors, randomization-SDR (RSDR) and eigenvector-SDR (ESDR) are proposed on the order of O(M3.5) and O(M3), respectively. Complexity analysis validates that the SDR-MSD strategy is desirable owing to its reduced complexity, compared with the exponential-complexity sphere decoding (SD) MSD. Furthermore, Monte-Carlo simulations demonstrate that the proposed SDR detectors provide the bit error rate (BER) performance almost the same with that of the SD method, and the RSDR outperforms the ESDR at the price of slightly higher complexity.

Large-scale multi-cell multiuser multiple-input multiple-output (LS-MIMO) systems can greatly increase the spectral efficiency. But the performance of these systems is deteriorated by pilot contamination. In this paper, first, a pilot design criterion is proposed by exploiting the orthogonality of channel vectors of LS-MIMO systems. Second, following this criterion, Chu sequences based pilots are designed. Because of the proposed pilots, the channel estimate of most terminals of a cell is only interfered by the partial cells rather than all the other cells, where the latter is caused by traditional pilots. As a result, pilot contamination is mitigated. Numerical results verify the effectiveness of the proposed pilots.

In this paper, we study the amplify-and-forward (AF) relaying based signaling scheme for the Y channel consisting of three single-antenna users and a two-antenna relay. In such a particular scenario, traditional signal space alignment for network coding (SSA-NC) scheme is not feasible. Moreover, the time division based multi-user multiple-input multiple-output (MU-MIMO) scheme has to rely on time division mode, i.e., more than two time slots are required to complete the whole communication process, which results in throughput loss. We develop an asymmetric signal space alignment (ASSA) scheme to enable all the users to finish information exchange with each other via the relay within two time slots. Brief degrees of freedom (DOF) analysis and numerical simulations have been provided to demonstrate that the proposed signaling technique significantly outperforms the conventional time division based MU-MIMO scheme.

In this paper, we propose two kinds of interference alignment (IA) schemes with limited feedback for the two-cell interfering multi-user multiple-input multiple-output multiple access channel (MIMO-MAC). Since IA with limited feedback results in residual interference for the quantization error, more effective schemes are introduced to reduce the residual interference in this paper compared with the ever work. The first kind of schemes means that the precoding matrices at the transmitters are the quantization value after obtaining the IA close-form solution of the precoding and decoding matrices at the receivers. This kind of schemes has been generalized to K users in this paper, and decoding matrices design is considered to reduce the quantization error to improve the performance. For the second kind of schemes, the beamforming vectors are chosen in the codebooks directly which guarantee the inter-cell interference (ICI) are most aligned. Monte-Carlo simulations illustrate that the proposed schemes outperform the existing schemes.

In this paper, an improved method for reconstruction of doubly selective wireless channels in piloted-aided OFDM systems based an existing estimation method is proposed. In this re-expansion channel estimation process, the first few Fourier coefficients of each channel tap are estimated from the pilot information and the received signal firstly. Then the channel taps are estimated in the framework of Basis Expansion Model (BEM) from their respective Fourier coefficients. In the process of recovering BEM coefficients, instead of using the inverse method which is a Least Square (LS) problem, this paper proposes an improved method of recovering BEM coefficients from the estimated Fourier coefficients based on the Minimum Mean Square Error (MMSE) criterion. The proposed method is validated by simulating a system conforming to the IEEE 802.16e standard. Numerical results illustrate the performance gains achieved by the improved method.

Video multicast over wireless networks suffers from both heterogeneous packet loss resulting from different channel conditions and user capacity heterogeneity in screen resolution and mobile device battery life. To solve resource scheduling problem in video multicasting in heterogeneous network, an Asymmetric Nash Bargaining Game model in layered hybrid FEC/ARQ for scalable video multicast is proposed in this paper. The scheme is applied in multicast server for each time slot, and the server will play the bargaining game for all users according to their real-time channel conditions and device capacities. By solving the bargaining problem, the server achieves to provide fair and efficient multicast utility for each user. Moreover, a formula of bargaining power is proposed in the asymmetric game model to adjust resource allocation according to system bias and user priority.

Large-scale multi-cell multiuser multiple-input multiple-output (LS-MIMO) systems have received much attention recently. But the performance of these systems is deteriorated by imperfect channel state information (CSI). Hence, in this paper, a subspace-based semi-blind channel estimator is proposed. Based on the approximate orthogonality of the channel vectors of LS-MIMO systems, singular value decomposition (SVD) is employed on the received signals to determine the channel matrix up to an ambiguity matrix. Then matrix inversion is avoided in resolving the ambiguity matrix, which is essential to traditional subspace-based estimators and loses the partial CSI. The properties of the estimators are analyzed, and the analysis shows that the estimation accuracy of the proposed estimator is improved. Simulations comparing the proposed approach with others illustrate improvement of the performance of the proposed approach.

In this paper, we propose a pilot-aided channel estimation scheme for Orthogonal Frequency-Division Multiplexing (OFDM) systems where channels are assumed to be both time-varying and sparse. Basis Expansion Models (BEM) are often used to model and reconstruct time-varying channel taps. In this paper, the framework of BEM is applied to OFDM systems with time-varying sparse channels. A new method to detect the positions of significant taps is proposed based on the use of Constant Amplitude Zero Auto Correlation (CAZAC) sequence. Based on the results of detection, BEM based estimation is implemented to estimate the detected taps. The numerical simulations illustrate that the proposed two-step estimation scheme for significant channel taps outperforms direct estimation methods for all the channel taps and also this method can reduce the required pilots and thus reduce the computational load and improve the spectral efficiency.

The effective cross-tier interference (CTI) mitigation is a key technique for the macrocell and femtocell two-tier heterogeneous networks. In this paper, a cognitive radio relay selection (CRRS) scheme is proposed to improve the received power of the macrocell users served by the macrocell base station (MBS). The scheme is intended to ensure the macrocell users can endure much more interference from the femtocell users in order to improve the quality of service (QoS) of the femto-networks. In addition, the fairness of the femto-networks is analyzed, with the objective of accounting for the maximal density of femtocell base stations that located in a certain area. Simulation results demonstrate that, compared with the existing works, the proposed scheme can not only increase the number of the femtocell users whose Signal-to-Interference-plus-Noise-Ratio (SINR) requirements are guaranteed, but also improve the average SINR of the femtocell users.

In this paper, we analyze the non-cooperative power control algorithm based on game theory in the two-tier femtocell networks, and find that the outcome of the game in a Nash equilibrium (NE) is inefficient. That is to say, there are still many users that can not achieve the target Signal-to-Interference-Plus-Noise Ratio (SINR) at the NE point, especially femtocell user equipments (FUE). Therefore, we propose a novel power control scheme in heterogeneous Femto-Macro cell networks, which can guarantee the target SINR of the macrocell user (MUEs), and make as many as possible FUEs to achieve their target SINRs. The proposed power control algorithm introduces the user selection and channel re-allocation in the conventional non-cooperative power control game. In addition, to further optimize the proposed scheme, we propose a novel FUE-SINR based MUE link quality protection algorithm. The propose scheme is able to improve the efficiency of Nash equilibrium, i.e., ensure more users to attain the target SINRs. Numerical simulations verify the conclusions.

The combination of physical-layer network coding (PNC) and multiple-input multiple-output (MIMO) is expected to improve the throughput of two-way relay network. In this paper, we propose a zero-forcing based MIMO two-way relay scheme in conjunction with a simple Max-Min relay antenna selection. This scheme solves the unpractical constraint encountered by many existing MIMO two-way relay schemes for application, which requires the relay to equip fewer antennas than the end node. Our scheme, on the other hand, benefits from the dedicated relay that has more antennas than the end node. A notable diversity advantage is obtained from judicious relay antenna selection. The reliability of the simple ZF based MIMO two-way relay is therefore improved. Of particular note, this paper extends our previous study to 1) support the more general application with non-binary PNC and 2) give a complete analysis on the attained end-to-end diversity with explicit theoretical result under i.i.d. Rayleigh fading channel.

In this paper, we study the cognitive interference management (CIM) scheme in the coexistence networks of macro-cells and femtocells. A single-channel detection based spectrum allocation algorithm is proposed to enhance the performance of femtocells considering the co-tier interference and co-tier interference. The proposed scheme converts the complex interference environment into several specific channel categories and successfully recognizes these channel patterns, while the group-channel resource allocation algorithm does not distinguish the interference channels in detail and operates on the continuous group of channels. Therefore, femtocell base stations (FBSs) can recognize the spectral environment more completely and allocate more available channels. As a result, our scheme is able to significantly improve the femtocell spectral efficiency and the signal-to-interference-and-noise ratio (SINR) performance of femtocell users (FUEs), meanwhile, it avoids the strong interference on the existing macrocell. Numerical simulations verify the conclusions.

In this paper, we propose a new interference alignment scheme with limited feedback for the two-cell interfering multi-user multiple-input multiple-output multiple access channel (MIMO-MAC), which provides better performance compared with other schemes when the number of feedback bits is same. Then, we analyse the rate loss for the quantization error, and show that the rate loss is only impacted by the residual inter-cell interference. By characterizing the rate loss as a function of the number of feedback bits, a bits allocation algorithm is introduced to further improve the system throughput. Monte-Carlo simulations illustrate that our proposed scheme outperforms the existing schemes.

In this paper, we consider a multi-way relaying channel where 2K users are divided into two groups averagely and each of them exchanges messages with every user of the other group via an intermediate relay. We term it multi-user multi-way relaying X network. We design the beamforming vectors at the users and the relay to achieve an interference alignment (IA) solution. Meanwhile, we investigate the feasibility conditions on the required amount of antennas for each node. Brief theoretical analysis and numerical simulations have been provided to demonstrate that the degrees of freedom (DOF) of 2K2is obtained.

In this paper, we consider the resource allocation problem in the uplink transmission of an orthogonal frequency-division multiple access (OFDMA) based cognitive radio (CR) network. The resource allocation aims to maximize the uplink throughput of secondary users (SUs) in CR network under the constraints of the primary user (PU) interference and the transmit power limits of SUs. In general, the optimal joint power and subchannel allocation is known as NP-hard. To ease the computation complexity while maintain good performance, we propose a novel particle swarm optimization (PSO) based joint uplink power and subchannel allocation algorithm to solve this resource allocation problem. Due to the combinatorial nature of the resource allocation problem, our algorithm in which power continuously changes while the subchannel allocation strategy alters in the iteration process can obtain better performance than the existing decomposition based algorithm that subchannel assignment and power allocation are implemented separately. Simulation results show the effectiveness of the proposed algorithm.

In this paper, a practical dynamic resource allocation scheme for real-time variable bit rate (VBR) video transmission is investigated. This scheme uses time-domain adaptive linear prediction, instead of the conventional size prediction of I-, P- and B- frames, to forecast future bandwidth requirement in time. Media delivery index (MDI) is used here as QoS measurement and complementary reference for adjustment. Besides, considering that the practical multi-user application may arouse adjustment collision, an adjustment priorities classified strategy based on the user-level classifications is put forward. Last, a test-bed based on the proposed scheme is constructed and the experimental results show that the bandwidth effective utilization has increased by 20%-60% compared to a fixed service rate with QoS guaranteed and no collisions.