Artificial intelligence (AI) is pivotal in advancing fifth-generation (5G)-Advanced and sixth-generation systems, capturing substantial research interest. Both the 3rd Generation Partnership Project (3GPP) and leading corporations champion AI's standardization in wireless communication. This piece delves into AI's role in channel state information (CSI) prediction, a sub-use case acknowledged in 5G-Advanced by the 3GPP. We offer an exhaustive survey of AI-driven CSI prediction, highlighting crucial elements like accuracy, generalization, and complexity. Further, we touch on the practical side of model management, encompassing training, monitoring, and data gathering. Moreover, we explore prospects for CSI prediction in future wireless communication systems, entailing integrated design with feedback, multitasking synergy, and predictions in rapid scenarios. This article seeks to be a touchstone for subsequent research in this burgeoning domain.

This paper proposes a hybrid wireless communication framework that integrates an active intelligent reflecting surface (IRS) with a decode-and-forward (DF) relay to enhance spectral efficiency in extended-range scenarios. By combining the amplification capability of the active IRS and the signal regeneration function of the DF relay, the proposed system effectively mitigates path loss and fading. We derive closed-form upper bounds on the achievable rate and develop an optimal power allocation strategy under a total power constraint. Numerical results demonstrate that the hybrid scheme significantly outperforms conventional passive IRS-assisted or active IRS-only configurations, particularly under conditions of limited reflecting elements or moderate signal-to-noise ratios.

Many applications for locating a radio signal source employ Global Navigation Satellite System (GNSS) to obtain a sensor's position. By using GNSS, a sensor can also synchronize with other sensors. For a sensor that is equipped with a GNSS receiver, it can be independent and is readily to be loaded on a flexible platform, such as an unmanned aerial vehicle (UAV). In this paper, we consider using such sensors and time-of-arrival (TOA) techniques to locate a radio signal source, and analyze the performance limit of source localization. Besides the performance analysis, this paper provides the geometric interpretation of the performance limit, which can illustrate how a sensor contributes to the source localization accuracy. The performance analysis and the geometric interpretation together give important insights into how to make better use of GNSS receiver for passive localization. Another contribution is we propose a modified closed-form solution for this localization problem. Compared with previous literature, this solution takes both sensor position and synchronization uncertainty into account, and it does not need proper initial guess of source position and is computationally efficient. Our simulation results validate the efficiency of this solution.

Satellite communication plays an important role in 6G systems. However, satellite communication systems are more susceptible to intentional or unintentional interference signals than other communication systems because of their working mechanism of transparent forwarding. For the purpose of eliminating the influence of interference, this paper develops an angle reciprocal interference suppression scheme based on the reconstruction of interference-plus-noise covariance matrix (ARIS-RIN). Firstly, we utilize the reciprocity between the known beam central angle and the unknown signal arrival angle to estimate the angle of arrival (AOA) of desired signal due to the multi-beam coverage. Then, according to the priori known spatial spectrum distribution, the interference-plus-noise covariance matrix (INCM) is reconstructed by integrating within the range except the direction of desired signal. In order to correct the estimation bias of the first two steps, the worst-case performance optimization technology is adopted in the process of solving the beamforming vector. Numerical simulation results show that the developed scheme: 1) has a higher output signal-to-interference-plus-noise ratio (SINR) under arbitrary signal-to-noise ratio (SNR); 2) still has good performance under small snapshots; 3) is robuster and easier to be realized when comparing with minimum variance distortionless response (MVDR) and the traditional diagonal loading algorithms.

The linear transmission impairments, such as the timing offset (TO), frequency offset (FO), and chromatic dispersion (CD), are major factors of signal degradations in coherent optical fiber communication systems. The estimation and compensation of such impairments play significant roles in the receiver side digital signal processing (DSP) unit. In this paper, we propose to combat the linear impairments systematically (including TO, FO and CD) with a joint time-frequency signal processing by taking the advantage of fractional Fourier transform (FrFT). In view of geometrical analysis, TO/FO induces a shift in time/frequency coordinate and the CD leads to the rotation in the fractional domain. Both mathematical derivations and geometrical interpretations have been established to unveil the relationships between impairments and linear frequency modulated (LFM) training symbols (TSs). By considering a typical coherent optical orthogonal frequency-division multiplexing (CO-OFDM) transmission system, three kinds of linear impairments have been jointly estimated by simple geometric calculations using appropriately designed TS based on FrFTs. Simulation and experimental results confirmed the feasibility of time-frequency techniques with better accuracy, less complexity, and improved spectral efficiency.

Multiple-input multiple-output (MIMO) technology has been promoted to achieve high-speed data transmission. Compared to the orthogonal frequency division multiplexing (OFDM) based MIMO systems, the single-carrier scheme (SC) has attracted lots of attention due to its low peak-to-average ratio (PAPR) feature and is quite suitable for long-distance transmission. However, one of the significant challenges of SC-MIMO systems is the vast complexity of channel equalization and signal detection, especially for the inter-symbol interference (ISI) caused by the multipath channel. The single carrier frequency domain equalization-based minimum mean square error (SCFDE-MMSE) algorithm is proved to achieve satisfying performance. However, it involves large numbers of DFTs and large-scale matrix inversions, which is unacceptable for practical systems. In this paper, a low-complexity SCFDE-MMSE (LC-SCFDE-MMSE) algorithm is proposed. Firstly, the characteristics of cyclic matrix and DFT cyclic shift are combined to reduce the number of DFTs. Secondly, SCFDE is transformed into a symbol-wise manner to avoid large-scale matrix inversions. Finally, a frequency-domain interpolation method is proposed to reduce the number of small-scale matrix inversions further. According to the evaluation results, the proposed LC-SCFDE-MMSE reduces the complexity of the traditional SCFDE-MMSE algorithm by more than one order of magnitude with less than 0.1 dB performance loss.

With the rapid development and commercialization of wireless communications, the execution of OTA testing requires a tremendous amount of test time. Therefore, test time reduction is of great significance. The objective of this article is to determine optimal measurement grids for SISO OTA testing of 5G Sub-6 GHz user equipments (UEs) in anechoic chamber with satisfactory accuracy and efficiency. The effect of different grid configurations on OTA performance is analyzed quantitatively using reference radiation patterns at different bands. These patterns are utilized to mimic the worst-case radiation patterns of 5G Sub-6 GHz UEs. Subsequently, the associated measurement uncertainty (MU) terms are quantitatively analyzed and determined based on statistical analysis. According to the comparison of calculated MUs, reduction of grid points from currently-required 62 (30/30, $\Delta\theta/\Delta\phi$) to 26 (45/45) could achieve roughly 60% test time reduction for Sub-6 GHz, while still maintaining an uncertainty level of $\leq $ 0.25 dB. These values can be further reduced to 14 (60/60) with 80% reduction for Sub-3 GHz. More importantly, the recommended grid configurations in this research are applicable to both TIS and TRP testing.

The uplink massive multiple-input multi-ple-output (MIMO) status update system is very concerned about information freshness performance, especially for some central control Internet of Things (IoT) applications. In this context, age of information (AoI), as the metric of information freshness, gets more and more recognition, and simultaneously, the status packet blocklength plays an important role in improving the information freshness. In this work, we firstly consider a case with perfect channel state information (CSI) at the base station (BS), and derive the closed-form expression of the average AoI by using the Shannon theory. Guided by this, we obtain the tradeoff relationship among the status packet blocklength, transmission time and transmission failure probability. Accordingly, we optimize the status packet blocklength to minimize the average AoI. Then, we consider a more practical case with finite blocklength and imperfect CSI at the BS. In this case, we exploit pilot sequence to assist channel estimation, and derive an approximated closed-form expression of the average AoI according to short packet communication theory. It is found that increasing pilot blocklength can improve the accuracy of channel estimation but reduce the frequency of status updates. Hence, we jointly optimize the pilot blocklength and status packet blocklength to improve the AoI performance. Extensive simulation results validate that the proposed methods can achieve almost the same performance as the exhaustive search methods.

Advanced persistent threat (APT) can use malware, vulnerabilities, and obfuscation countermeasures to launch cyber attacks against specific targets, spy and steal core information, and penetrate and damage critical infrastructure and target systems. Also, the APT attack has caused a catastrophic impact on global network security. Traditional APT attack detection is achieved by constructing rules or manual reverse analysis using expert experience, with poor intelligence and robustness. However, current research lacks a comprehensive effort to sort out the intelligent methods of APT attack detection. To this end, we summarize and review the research on intelligent detection methods for APT attacks. Firstly, we propose two APT attack intelligent detection frameworks for endpoint samples and malware, and for malware-generated audit logs. Secondly, this paper divides APT attack detection into four critical tasks: malicious attack detection, malicious family detection, malicious behavior identification, and malicious code location. In addition, we further analyze and summarize the strategies and characteristics of existing intelligent methods for each task. Finally, we look forward to the forefront of research and potential directions of APT attack detection, which can promote the development of intelligent defense against APT attacks.

The integration of blockchain and edge-to-end collaborative computing offers a solution to address the trust issues arising from untrusted IIoT devices. However, ensuring efficiency and energy-saving in applying blockchain to edge-to-end collaborative computing remains a significant challenge. To tackle this, this paper proposes an innovative task-oriented blockchain architecture. The architecture comprises trusted Edge Computing (EC) servers and untrusted Industrial Internet of Things (IIoT) devices. We organize untrusted IIoT devices into several clusters, each executing a task in the form of smart contracts, and package the work logs of a task into a block. Executing a task with smart contracts within a cluster ensures the reliability of the task result. Reducing the scope of nodes involved in block consensus increases the overall throughput of the blockchain. Packaging task logs into blocks, storing and propagating blocks through corresponding Edge Computing (EC) servers reduces network load and avoids computing power competition. The paper also presents the proposed architecture's theoretical TPS (Transactions Per Second) and failure probability calculations. Experimental results demonstrate that this architecture ensures computational security, improves TPS, and reduces resource consumption.

One of the evolving hand biometric features considered so far is finger knuckle printing, because of its ability towards unique identification of individuals. Despite many attempts have been made in this area of research, the accuracy of the recognition model remains a major issue. To overcome this problem, a novel biometric-based method, named finger-knuckle-print (FKP), has been developed for individual verification. The proposed system carries key steps such as preprocessing, segmentation, feature extraction and classification. Initially input FKP image is fed into the preprocessing stage where colour images are converted to gray scale image for augmenting the system performance. Afterwards, segmentation process is carried out with the help of CROI (Circular Region of Interest) and Morphological operation. Then, feature extraction stage is carried out using Gabor-Derivative line approach for extracting intrinsic features. Finally, DCNN (Deep Convolutional Neural Network) is trained for the processed knuckle images to recognize imposter and genuine individuals. Extensive experiments on standard FKP database demonstrates that the proposed method attains considerable improvement compared with state-of-the-art methods. The overall accuracy attained for the proposed methodology is 95.6% which is achieved better than the existing techniques.

Network virtualization is the development trend and inevitable requirement of hybrid wireless sensor networks (HWSNs). Low mapping efficiency and service interruption caused by mobility seriously affect the reliability of sensing tasks and ultimately affect the long-term revenue of the infrastructure providers. In response to these problems, this paper proposes an efficient virtual network embedding algorithm with a reliable service guarantee. Based on the topological attributes of nodes, a method for evaluating the physical network resource importance degree is proposed, and the nodes with rich resources are selected to improve embedding efficiency. Then, a method for evaluating the physical network reliability degree is proposed to predict the probability of mobile sensors providing uninterrupted services. The simulation results show that the proposed algorithm improves the acceptance rate of virtual sensor networks (VSN) embedding requests and the long-term revenue of the infrastructure providers.% so as to reduce the out-of-band (OOB) radiation as much as possible. Parameters of the proposed scheme are solved under joint con-straints of constant power and unity cumulative distribution. A new receiving method is also proposed to improve the bit error rate (BER) performance of OFDM systems. Simulation results indicate the proposed scheme can achieve better OOB radiation and BER performance at same PAPR levels, compared with existing similar companding algorithms.

To improve the traffic scheduling capability in operator data center networks, an analysis prediction and online scheduling mechanism (APOS) is designed, considering both the network structure and the network traffic in the operator data center. Fibonacci tree optimization algorithm (FTO) is embedded into the analysis prediction and the online scheduling stages, the FTO traffic scheduling strategy is proposed. By taking the global optimal and the multi-modal optimization advantage of FTO, the traffic scheduling optimal solution and many suboptimal solutions can be obtained. The experiment results show that the FTO traffic scheduling strategy can schedule traffic in data center networks reasonably, and improve the load balancing in the operator data center network effectively.

Wireless Power Transmission (WPT) has been widely used to replenish energy for various rechargeable devices. The ElectroMagnetic Radiation (EMR) of WPT has attracted great attention of safety concerns. It is possible for the malicious attacker to launch the EMR attack by capturing multiple wireless chargers. Little work has studied the EMR attack itself . In this paper, we propose a realistic EMR hazard model, which outputs the diminishing marginal hazard with EMR, with adjustable parameters to the target entities. We formulate three EMR attack models, termed Cumulative EMR Attack (CEA), Overall EMR Attack (OEA) and Unsafety EMR Attack (UEA), and propose the performance guaranteed algorithm of EMR attack for each model. We conduct extensive simulations and field experiments on a testbed. The results show that the proposed algorithms can output the near-optimal solution with much less running time than the optimal algorithms. The results of field experiments in a small testbed show that the utilities of CEAA and OEAA are increased by 70.5% and 12.9% than the comparison algorithms, respectively. Moreover, the number of captured chargers of UEAA is 5.9% less than the comparison algorithms. Our simulations also show the designed algorithms can perform better in a large-scale charging network.

This paper explores the development of interpretable data elements from raw data using Kolmogorov-Arnold Networks (KAN). With the exponential growth of data in contemporary society, there is an urgent need for effective data processing methods to unlock the full potential of this resource. The study focuses on the application of KAN in the transportation sector to transform raw traffic data into meaningful data elements. The core of the research is the KAN-T-GCN model, which synergizes Kolmogorov-Arnold Networks with Temporal Graph Convolutional Networks (T-GCN). This innovative model demonstrates superior performance in predicting traffic speeds, outperforming existing methods in terms of accuracy, reliability, and interpretability. The model was evaluated using real-world datasets from Shenzhen, Los Angeles, and the San Francisco Bay Area, showing significant improvements in different metrics. The paper highlights the potential of KAN-T-GCN to revolutionize data-driven decision-making in traffic management and other sectors, underscoring its ability to handle dynamic updates and maintain data integrity.

Recent advances in integrating Digital Twins (DTs) with Heterogeneous Vehicular Networks (HetVNets) enhance decision-making and improve network performance. Additionally, developments in Mobile Edge Computing (MEC) support the computational demands of DTs. However, the decentralized nature of MEC systems introduces security challenges and traditional HetVNets fail to efficiently integrate diverse computing and network resources, limiting their ability to handle services for vehicles. This paper presents a novel service request offloading framework for DT-HetVNets to address these issues. In this framework, we design utility functions for vehicles and infrastructures to maximize satisfaction of their requirements through data synchronization and decision-making between DTs and entities. Furthermore, we propose a new honestly based distributed PoA(HDPoA) via scalable work. The interactions between infrastructures and vehicles are modeled as a multi-leader multi-follower (MLMF) game, and we develop a dynamic iterative algorithm to achieve the Nash equilibrium (NE) of the proposed game-theoretic model. Experimental results validate the effectiveness and accuracy of our scheme.

In the heterogeneous power internet of things(IoT) environment, data signals are acquired to support different business systems to realize advanced intelligent applications, with massive, multi-source, heterogeneous and other characteristics. Reliable perception of information and efficient transmission of energy in multi-source heterogeneous environments are crucial issues. Compressive sensing (CS), as an effective method of signal compression and transmission, can accurately recover the original signal only by very few sampling. In this paper, we study a new method of multi-source heterogeneous data signal reconstruction of power IoT based on compressive sensing technology. Based on the traditional compressive sensing technology to directly recover multi-source heterogeneous signals, we fully use the interference subspace information to design the measurement matrix, which directly and effectively eliminates the interference while making the measurement. The measure matrix is optimized by minimizing the average cross-coherence of the matrix, and the reconstruction performance of the new method is further improved. Finally, the effectiveness of the new method with different parameter settings under different multi-source heterogeneous data signal cases is verified by using orthogonal matching pursuit (OMP) and sparsity adaptive matching pursuit (SAMP) for considering the actual environment with prior information utilization of signal sparsity and no prior information utilization of signal sparsity.

Federated learning (FL) is an intricate and privacy-preserving technique that enables distributed mobile devices to collaboratively train a machine learning model. However, in real-world FL scenarios, the training performance is affected by a combination of factors such as the mobility of user devices, limited communication and computational resources, thus making the user scheduling problem crucial. To tackle this problem, we jointly consider the user mobility, communication and computational capacities, and develop a stochastic optimization problem to minimize the convergence time. Specifically, we first establish a convergence bound on the training performance based on the heterogeneity of users' data, and then leverage this bound to derive the participation rate for each user. After deriving the user-specific participation rate, we aim to minimize the training latency by optimizing user scheduling under the constraints of the energy consumption and participation rate. Afterward, we transform this optimization problem to the contextual multi-armed bandit framework based on the Lyapunov method and solve it with the submodular reward enhanced linear upper confidence bound (SR-linUCB) algorithm. Experimental results demonstrate the superiority of our proposed algorithm on the training performance and time consumption compared with state-of-the-art algorithms for both independent and identically distributed (IID) and non-IID settings.

Non-panoramic virtual reality (VR) provides users with immersive experiences involving strong interactivity, thus attracting growing research and development attention. However, the demand for high bandwidth and low latency in VR services presents greater challenges to existing networks. Inspired by mobile edge computing (MEC), VR users can offload rendering tasks to other devices. The main challenge of task offloading is to minimize latency and energy consumption. Yet, in non-panoramic VR scenarios, it is essential to consider the Quality of Perceptual Experience (QOPE) for users. Simultaneously, one must also take into account the diverse requirements of users in real-world scenarios. Therefore, this paper proposes a QOPE model to measure the visual quality of non-panoramic VR users and models the non-panoramic VR task offloading problem based on MEC as a constrained multi-objective optimization problem (CMOP) that minimizes latency and energy consumption while providing a satisfied QOPE. And we propose an evolutionary algorithm (EA), G-NSGA-II, to solve the CMOP. Simulation results show that the algorithm can effectively find various trade-off solutions among the objectives, satisfying the requirements of different users.

The phasor data concentrator placement (PDCP) in wide area measurement systems (WAMS) is an optimization problem in the communication network planning for power grid. Instead of using the traditional integer linear programming (ILP) based modeling and solution schemes that ignore the graph-related features of WAMS, in this work, the PDCP problem is solved through a heuristic graph-based two-phase procedure (TPP): topology partitioning, and phasor data concentrator (PDC) provisioning. Based on the existing minimum $k$-section algorithms in graph theory, the $k$-base topology partitioning algorithm is proposed. To improve the performance, the ``center-node-last" pre-partitioning algorithm is proposed to give an initial partition before the $k$-base partitioning algorithm is applied. Then, the PDC provisioning algorithm is proposed to locate PDCs into the decomposed sub-graphs. The proposed TPP was evaluated on five different IEEE benchmark test power systems and the achieved overall communication performance compared to the ILP based schemes show the validity and efficiency of the proposed method.

With the increasingly complex and changeable electromagnetic environment, wireless communication systems are facing jamming and abnormal signal injection, which significantly affects the normal operation of a communication system. In particular, the abnormal signals may emulate the normal signals, which makes it very challenging for abnormal signal recognition. In this paper, we propose a new abnormal signal recognition scheme, which combines time-frequency analysis with deep learning to effectively identify synthetic abnormal communication signals. Firstly, we emulate synthetic abnormal communication signals including seven jamming patterns. Then, we model an abnormal communication signals recognition system based on the communication protocol between the transmitter and the receiver. To improve the performance, we convert the original signal into the time-frequency spectrogram to develop an image classification algorithm. Simulation results demonstrate that the proposed method can effectively recognize the abnormal signals under various parameter configurations, even under low signal-to-noise ratio (SNR) and low jamming-to-signal ratio (JSR) conditions.