The problem of combination MIMO and NOMA technologies was considered in this paper. Combining these technologies is the key challenge for future 6G networks. The main goal of this combination is to have the advantages of each of the technologies separately when they are in use together. It is analyzed that the proposed approaches which are highlighted in the various papers do not allow to realize full potential of these both technologies by using them together. The main goal of this article is to propose a new combination of MIMO and NOMA technologies using a multi-volume codebook, which can be formed from extended codebooks. Multi-volume codebooks allow individual codewords to be used to transmit the signal of each user and each transmitting antenna. It should also be noted that codebooks generation is also a separate independent task. The proposed method makes it possible to obtain additional energy gain by separating of uncorrelated fading in resources and increasing the number of correlations between the signals of different users. The results of the proposed method are a reduction in the required signal-to-noise ratio, increase in capacity and improvement in the efficiency of orthogonal resource use.

Channel state information (CSI) is essential to unlock the potential of reconfigurable intelligent surfaces (RISs) in wireless communication systems. Since massive RIS elements are typically implemented without baseband signal processing capabilities, limited CSI feedback is necessary when designing the reflection/refraction coefficients of the RIS. In this article, the unique RIS-assisted channel features, such as the RIS position-dependent channel fluctuation, the ultra-high dimensional sub-channel matrix, and the structured sparsity, are distilled from recent advances in limited feedback and used as guidelines for designing feedback schemes. We begin by illustrating the use cases and the corresponding challenges associated with RIS feedback. We then discuss how to leverage techniques such as channel customization, structured-sparsity, autoencoders, and others to reduce feedback overhead and complexity when devising feedback schemes. Finally, we identify potential research directions by considering the unresolved challenges, the new RIS architecture, and the integration with multi-modal information and artificial intelligence.

This work explores physical-layer security for a cooperative cognitive radio wireless sensor network (CCR-WSN) with hardware impairments (HIs) and channel estimation errors (CEEs), where the cognitive transmissions are supported by multiple cognitive relays amidst potential passive eavesdropping. For enhancing security-reliability tradeoff (SRT), an opportunistic relay with random jammer (ORRJ) scheme and an opportunistic relay with oriented jammer (OROJ) scheme are proposed. We derive closed-form expressions for outage probabilities (OPs) and intercept probabilities (IPs) to evaluate reliability and security for both the proposed schemes. Numerical simulations demonstrate that the OROJ and ORRJ schemes outperform the random relay with random jammer (RRRJ) scheme in terms of SRT performance. The HIs at legitimate users and CEEs in main channels deteriorate SRT performance, while HIs affecting eavesdropper and CEEs in wiretap channels enhance SRT performance. The SRT performance of our proposed schemes benefits from multiple relays. Notably, the jammer-aided schemes outperform the pure opportunistic relay (POR) scheme from SRT perspective in low and medium signal-to-noise ratio (SNR) scenarios. However, the POR scheme excel our proposed schemes in a high SNR condition.

Low Earth Orbit (LEO) satellite constel-lations, with their advantages of higher power trans-mission and geometric diversity, have become a de-velopment trend of satellite navigation systems. How-ever, as the number of LEO satellites increases, is-sues such as decreased anti-interference performanceand compatibility have arisen. This paper proposes anovel modulation method. By replacing the periodicbinary cosine subcarrier in traditional Binary OffsetCarrier (BOC) signals with a periodic binary Hyper-bolic Frequency Modulation (HFM) subcarrier signal,this method achieves time-frequency domain charac-teristics that differ from traditional BOC signals. Thenew spectral features bring about performance im-provements: under a frontend bandwidth of 30 MHz,the anti-interference ability of matched interferencehas improved by 26.22% compared to BOCc modu-lation; for the spectrally crowded L1C/B1C/E1OS andL1C/A signals, the compatibility is superior to BOCsby 2 dB; it also surpasses traditional BOC signals interms of code tracking performance, multipath resis-tance, and narrowband interferenceresistance.Thenovelmodulationmethod proposed in this paper provides a potential modulation approach for future large-scale LEO satellite navigation-augmented systems.

SIMON is a family of lightweight block ciphers designed by the U.S. National Security Agency (NSA) in 2013. Differential-linear cryptanalysis is an important cryptanalysis method in cryptography, which has received widespread attention since its introduction by Langford and Hellman in 1994. In this paper, we investigate differential-linear cryptanalysis on SIMON. First, we explore the search method for differential-linear distinguishers from an algebraic perspective. By combining the algebraic transitional forms (ATF) technique proposed at CRYPTO 2021 and the automatic search method based on SAT, we obtain differential-linear distinguishers for all versions of SIMON. Second, we introduce dynamic key guessing techniques into differential-linear cryptanalysis, and propose a new framework for key recovery attacks. Finally, we app- ly this new framework to SIMON. As a result, for 20-round SIMON32/64, the data and time complexities are $2^{17.45}$ and $2^{60.63}$, respectively. For the 21-r- ound SIMON48/72, the data and time complexities are $2^{34.95}$ and $2^{67.71}$, respectively. For the 22-round SIMON48/96, 26-round SIMON64/96, 27-round SIMON64/128, 32-round SIMON96/144, 38-round SIMON128/192, and 38-round SIMON128/256, the data complexities are $2^{34.95}$, $2^{43.96}$, $2^{43.96}$, $2^{74.95}$, $2^{106.96}$, and $2^{106.96}$, respectively. The corresponding time complexities are $2^{82.18}$, $2^{88.40}$, $2^{106.89}$, $2^{143.64}$, $2^{178.40}$, and $2^{181.22}$, respectively. To the best of our knowledge, this is the first time that a key-recovery attack has been conducted on all versions of SIMON in differential-linear cryptanalysis. Furthermore, our attack rounds are currently the longest among existing differential-linear c- ryptanalysis results for SIMON.

Ambient Internet of Things (IoT) is a 3rd Generation Partnership Project (3GPP) technology that enables low-power, low-complexity, low-cost, and maintenance-free devices. These devices can harvest and store limited ambient energy, enabling them to amplify backscattered signals to extend communication range. Due to the constrained energy storage and increased device population, Ambient IoT seeks a lightweight but highly time and energy-efficient anti-collision algorithm. Existing anti-collision algorithms present opportunities to further reduce device access attempts and relax channel stability requirements, which have been underexplored in prior studies. This paper proposes a time and energy-efficient Retraverse Tree (RT) with novel traversal methods and introduces the Full Retraverse Tree (FRT) to skip redundant queries in RT. An efficient integrated ALOHA tree (IAT) method is also proposed for grouping tags and relaxing the channel stability requirement. Simulation results and theoretical analysis verify that the proposed IAT-FRT algorithm can extend a device's life span significantly, with about 0.94 system efficiency and averaging 1.4 access attempts per tag.

Cognitive radio (CR) embodies an intelligent approach to wireless communication technique, enabling it to sense the spectrum and assign the free bands to secondary user when those bands are not occupied by primary user (PU) without interfering with primary user through spectrum sensing (SS) technique. Designing efficient and reliable SS method is always the main part to realize CR. In the paper, we propose a novel SS algorithm that does not need to consider noise power and some prior information about the signal. In this algorithm, considering higher-order cumulants are more sensitive to Gaussian noise we choose higher-order cumulants matrix as the statistical criterion for SS at low signal-to-noise ratio (SNR). Specifically, without the need to pre-know the noise power and some prior information related to the PU, we can determine whether the PU exists by comparing the differences between the eigenvalues of higher-order cumulants matrix of the noise and primary user. Simulation results indicate that compared to energy detection (ED), M-estimators of covariance matrix(MECM), mean-to-square extreme eigenvalue (MSEE) and Maximum-minimum eigenvalue (MME) methods, the proposed algorithm exhibits better SS performance. When the noise is Colored Gaussian noise or the SNR is lower, the proposed algorithm demonstrates prominent SS capabilities.

In GNSS-denied environments, Signals of Opportunity (SOP) offer an efficient and passive solution for navigation and positioning by utilizing ambient signals. Nevertheless, conventional SOP techniques face significant challenges in real-time processing, especially under sub-Nyquist sampling conditions, due to high data acquisition rates and off-grid errors. To address this, this paper proposes the Signal Reconstruction and Kernel Sparse Encoding (SRKSE) model, a novel general framework for high-precision parameter estimation. By combining compressed sensing with a deep unfolding network, the SRKSE model not only achieves robust signal reconstruction but also effectively reduces quantization errors. Key innovations of SRKSE include dual cross-attention mechanisms for enhanced feature extraction, sinc sparse kernel encoding to minimize quantization errors, and a custom loss function for balanced optimization. With these advancements, SRKSE achieves up to a 650-fold improvement in Time of Arrival (TOA) estimation accuracy while operating at just 1% of the Nyquist sampling rate. The SRKSE surpasses both conventional and deep learning-based techniques in accuracy and efficiency, especially when operating under sub-Nyquist sampling conditions. Simulations and real-world experiments confirm the reliability and potential of SRKSE for real-time applications in IoT and wireless communication.

Extremely large-scale multiple-input multiple-output (XL-MIMO) is regarded as a promising technology for next-generation communication systems. However, this will expand the near-field (NF) range, rendering more users more likely to be located in the NF region. In this paper, we aim to answer two questions: What are the new characteristics of the NF channel? Is it necessary to develop new transciver techniques to maintain system performance within the NF region? To this end, we first review current NF channel models and analyze the differences between the existing 3GPP TR 38.901 channel model and the NF channel model, including the spherical wavefront and spatially non-stationarity. Then, we provide examples on how these differences affect the XL-MIMO system performance in terms of beamforming gain and achievable rate. Simulation results demonstrate that, when using far-field (FF) technique under the NF channel, the maximum normalized beam gain loss is less than 3 dB for most users in the NF region defined by Rayleigh distance. Moreover, the achievable rate loss of beam training is less than 3% compared to that realized by NF technique. Finally, we demonstrate the necessity of employing NF transceiver techniques based on simulation results.

In the wake of major natural disasters or human-made disasters, the communication infrastructure within disaster-stricken areas is frequently damaged. Unmanned aerial vehicles (UAVs), thanks to their merits such as rapid deployment and high mobility, are commonly regarded as an ideal option for constructing temporary communication networks. Considering the limited computing capability and battery power of UAVs, this paper proposes a two-layer UAV cooperative computing offloading strategy for emergency disaster relief scenarios. The multi-agent twin delayed deep deterministic policy gradient (MATD3) algorithm integrated with prioritized experience replay (PER) is utilized to jointly optimize the scheduling strategies of UAVs, task offloading ratios, and their mobility, aiming to diminish the energy consumption and delay of the system to the minimum. In order to address the aforementioned non-convex optimization issue, a Markov decision process (MDP) has been established. The results of simulation experiments demonstrate that, compared with the other four baseline algorithms, the algorithm introduced in this paper exhibits better convergence performance, verifying its feasibility and efficacy.

In the Internet of Vehicles (IoV), collaboration of vehicles, roadside units (RSUs), and cloud servers enables promising Artificial Intelligence (AI) applications for Intelligent Connected Vehicles (ICVs). To address data insufficiency and privacy concerns in AI tasks, Federated Learning (FL) has been introduced. However, without appropriate incentives, users may be reluctant to contribute their computing resources to FL. This paper proposes a three-layer Stackelberg game-based federated learning incentive mechanism, termed TLSG-FL, which designs rational reward and penalty strategies to motivate participants to contribute data and complete local training while preserving privacy. Furthermore, unlike conventional two-layer frameworks, the proposed three-layer architecture incorporates a cloud server for global aggregation, pricing policy management, and cross-region model coordination, overcoming the limitations of localized RSU aggregation and enhancing system-wide consistency. Simulation results demonstrate that TLSG-FL significantly improves the convergence speed and accuracy of the global model while achieving fair and efficient profit allocation among all participants.

This paper investigates a communication network composed of directional multi-beam unmanned aerial vehicles (UAVs), which operates under a TDMA mode. The timeslot allocation problem targets to maximize the communication capacity of the network by optimizing the communication states of all UAVs in each timeslot, under the constraint of maximum timeslot interval between any two UAVs. The formulated optimization problem is a linear 0-1 integer programming, which is difficult to solve, especially in large-scale networks. To solve it, we construct an undirected graph based on the concept of communication coding and a linear 0-1 integer programming model based on communication coding is then established for the timeslot resource allocation problem. The equivalence between the two resource allocation models is theoretically proven. To solve a general timeslot allocation problem with differentiated communication weights of UAVs, a variable neighborhood search algorithm is proposed. Finally, numerical examples validate the proposed model and algorithm.

To enhance weak target detection performance in existing Multiple Input Multiple Output (MIMO) Dual-function radar-communication (DFRC) systems, we propose a joint design scheme for transmit waveform and receive filter to achieve low auto-correlation and cross-correlation properties. With the aid of reconfigurable intelligent surfaces (RIS), which provide only direct assistance to communication, an optimization problem is proposed to achieve a trade-off between communication and detection performance by minimizing the weighted sum of the multi-user interference (MUI) and weighted integrated sidelobe level (WISL), along with a penalty term considered to control the pulse compression peak loss. Energy budget and shift constraints are introduced to meet the practical requirement. Since the problem is non-convex, the alternating optimization (AO) framework is employed to iteratively optimize the variables using methods such as the Riemannian compositional gradient (RCG) algorithm, low complexity-golden section search (LC-GSS) algorithm, and majorization-minimization (MM) algorithm. Numerical simulations demonstrate that the proposed scheme significantly improves both communication performance and target detection capabilities compared to conventional methods.

With the emergence of reconfigurable intelligent surface (RIS), its potential applications in wireless communication have garnered significant attention. This paper presents an innovative architecture that integrates sweeping robot-assisted RIS (SR-RIS) to function as mobile passive relays, thereby enhancing communication rates between wireless access points (APs) and multiple users. We formulate a maximization problem aimed at enhancing the communication rate by jointly optimizing the movement trajectory of the sweeping robot, service indicators, SR-RIS phase shift, and the transmit power of APs. To solve this complex optimization problem, we propose an alternating optimization algorithm and an improved genetic algorithm (IGA). Through iterative optimization, we can efficiently search for the optimal solution. Simulations validate that the proposed method significantly enhances the communication rate of the considered system compared with conventional alternatives.

Semantic communication, an emerging paradigm in advanced communication systems, focuses on transmitting the semantic meaning of data rather than exact, bit-by-bit replication. While this shift enhances communication efficiency, it introduces challenges in verifying the integrity and authenticity of the transmitted content, as traditional methods often fail to detect semantic-level distortions. To address these issues, this paper presents a multi-task learning framework specifically designed for robust semantic communication in image transmission. The framework utilizes a shared encoder to extract essential semantic features from the Region of Interest (ROI), followed by two task-specific decoders: one for image reconstruction and another for perceptual hash generation to enable semantic-level tamper detection. This dual-objective design improves transmission quality while enhancing resilience to content-preserving distortions and malicious manipulations. Experimental results demonstrate that the proposed framework achieves efficient transmission, high reconstruction fidelity, and robust semantic integrity verification, offering a practical solution for semantic integrity-preserving communication in visual data scenarios.

Low earth orbit (LEO) satellite communications have become one of the key technologies for the sixth generation network because of the multiple advantages over medium- and high-orbit satellites, such as low costs and low latency. How to utilize the LEO satellite infrastructure to develop more functionality and realize the integration of sensing and communication (ISAC) is a promising research area. In this paper, we investigate the LEO satellite ISAC technology in ocean scenarios. In particular, we model the sensing echo signals based on both signal-dependent and spatial-angle-dependent sea clutter, and formulate a joint multiuser beamforming and satellite selection optimization problem, aiming at maximizing the sum rate constrained on the target sensing requirement of the minimum signal-to-clutter-plus-noise power ratio (SCNR). We optimize the transceiver beamforming vectors via the sequential convex approximation method and the alternating optimization method. Moreover, we propose an efficient satellite selection method by solving an SCNR maximization problem. We provide various simulation results to verify that the proposed optimization algorithms could achieve a favorable trade-off between the communication and sensing (CS) performance in ocean scenarios. Results also show that mainly due to the sea clutter effect, the CS trade-off and the satellite selection are significantly influenced by the locations of different satellites and different sea conditions.

Time-sensitive networking (TSN) has emerged as a promising communication technique for hard real-time embedded systems. TSN generally comprises a control plane (i.e., planner) and a data plane (e.g., switch chip). While the planner generates policies that map critical traffics into the resources temporally and spatially, the switch chip forwards packets according to pre-scheduled results without any deviation. However, we observed that practical forwarding behaviors are inconsistent with the prescheduled results from the time dimension. We formalized and named the above phenomena an inconsistent problem, which leads to non-deterministic latency and losses of critical packets during packet forwarding. To address this problem, we propose the dual guard band (D-GB) mechanism to dissect the inconsistent problem. D-GB abstracts cross-queue and cross-chip guard bands based on the fine-grained latency analysis of the general forwarding model. Existing TSN planners integrating the D-GB can guarantee that no inconsistent problem occurs when critical packets are forwarded. We build an industrial sensing topology based on real-life TSN switch chips to verify our analysis and proposed mechanism. The results demonstrate that the D-GB mechanism solves the inconsistent problem.

Addressing the joint optimization problem of dual-function radar-communication (DFRC) waveforms, this paper proposes an enhanced waveform design method for multiple-input multiple output (MIMO) radar-communication systems. This approach is designed to minimize multi-user interference (MUI) energy in the downlink while adhering to constraints on total power, peak-to-average power ratio (PAPR), and the Cram′er-Rao Bound (CRB) for angle of arrival estimation. First, the study solves the optimal solution for both omnidirectional and directional beampattern design problems. Subsequently, a flexible trade-off optimization between radar and communication performance is conducted under the given constraints. Furthermore, this study introduces a joint optimization algorithm combining the golden section search and the alternating direction method of multipliers (ADMM), capable of solving for optimal solutions with low computational complexity for multiple variables. Finally, the Branch-and-Bound (BnB) algorithm is updated to address the constant modulus design problem, thereby significantly improving the algorithm’s convergence speed. Simulation experiments demonstrate the impact of the proposed constraints on the performance of the dual-functional system and verify that the designed waveforms are able to achieve an effective trade-off between MIMO radar and communication performance.

Time synchronization is a prerequisite for ensuring determinism in Time-Sensitive Networking. While time synchronization errors cannot be overlooked, pursuing minimal time errors may incur unnecessary costs. Using complex network theory, this study proposes a hierarchy for TSN and introduces the concept of bounded time error. A coupling model between traffic scheduling and time synchronization is established, deriving functional relationships among end-to-end delay, delay jitter, gate window, and time error. These relationships illustrate that time errors can trigger jumps in delay and delay jitter. To evaluate different time errors impact on traffic scheduling performance, an end-to-end transmission experiment scheme is designed, along with the construction of a TSN test platform implementing two representative cases. Case A is a closed TSN domain scenario with pure TSN switches emulating closed factory floor network. Case B depicts remote factory interconnection where TSN domains link via non-TSN domains composed of OpenFlow switches. Results from Case A show that delay and delay jitter on a single node are most significantly affected by time errors, up to one gating cycle. End-to-end delay jitter tends to increase with the number of hops. When the ratio of time error boundary to window exceeds 10%, the number of schedulable traffic flows decreases rapidly. Case B reveals that when time error is below 1 μs, the number of schedulable traffic flows begins to increase significantly, approaching full schedulability at errors below 0.6 μs.

With the development of new Internet services with computation-intensive and delay-sensitive tasks, the traditional ``Best Effort" network communication mode has been greatly challenged. The network system is urgently required to provide end-to-end communication determinacy and computing determinacy for new applications to ensure the safe and efficient operation of services. Based on the research of the convergence of computing and networking, a new network paradigm named deterministic computing power networking (Det-CPN) is proposed. In this article, we firstly introduce the research advance of computing power networking. And then the motivations and scenarios of Det-CPN are analyzed. Following that, we present the system architecture, technological capabilities, workflow as well as key technologies for Det-CPN. Moreover, performance evaluation and simulation results are presented to illustrate the performance of the proposed scheme. Finally, the challenges and future trends of Det-CPN are analyzed and discussed.

Machine learning-based localization techniques have demonstrated effectiveness in complex indoor environments. However, they often suffer from overfitting issues due to oversimplified mappings between received signals and estimated positions. This paper presents a deep variational model-based indoor localization system for 5G networks. The proposed approach characterizes channel state information by constructing both location-dependent and location-independent latent variables. By extracting location-dependent features, our deep variational model jointly mitigates time-difference-of-arrival and angle-of-arrival estimation errors, thereby significantly improving localization accuracy. We validated the proposed system in complex indoor environments using a 3GPP-compliant 5G testbed, demonstrating its superior performance over traditional machine learning methods.

The high mobility of UAVs poses significant challenges for millimeter-wave beam prediction. Traditional beam training methods suffer from intolerable signaling overhead and face difficulties in establishing a reliable link within the channel coherence time. To address this, we propose a Beam Prediction Model (MIS) based on an Intelligent Multimodal Information Screening Mechanism. This model fuses UAV multimodal information and uses Recursive Feature Elimination (RFE) to dynamically identify the key modalities for the task. We designed a lightweight early fusion network architecture and combined it with a multilayer perceptron (MLP) to establish robust associations between key UAV information and the optimal beam index, enabling fast and accurate beam prediction. Experimental results on the Deepsense6G real-world dataset show that the MIS model can achieve 79.63% top-1 prediction accuracy and 98.41% top-3 prediction accuracy without the need for beam search. Compared to traditional position-assisted beam prediction models, it improves top-1 accuracy by 19.66%, and reduces measurement overhead by 69.14% compared to local beam search methods. This research provides an engineering-feasible beam prediction solution for high-mobility UAV communication systems.

Federated Learning (FL) is an innovative machine learning paradigm that eliminates the need to upload user data to a central server, enabling efficient collaborative learning across multiple distributed devices while ensuring data privacy and compliance with data localization regulations. However, in real-world scenarios, the inherent unreliability of wireless communication, such as network latency, packet loss, and connection interruptions, can cause unstable transmission of model updates, greatly hindering its effectiveness. This paper analyzes the influence of transmission errors on the convergence rate of the FL algorithm and formulates an adaptive power control problem to maximize the convergence rate. By solving this problem, we dynamically adjust the transmit power of each client according to the network condition and speed up the convergence process of the global model. Through extensive experiments, we not only prove the convergence of the adaptive power control algorithm but also demonstrate its significant performance improvement in the FL scenario characterized by unreliable wireless links.

Compared with one stage predistortion, the multistage predistortion that adopts a cascade of several models can further compensate the strong nonlinear distortions. However, the multistage method requires more complex identification process. To simplify the identification process and improve the generalization performance, we propose a novel two stage DPD based on generalized advanced neural network (GANN) model. In this framework, GANN is inserted into the DPD architecture as the first stage model. The residual nonlinearity of the power amplifier (PA) is compensated by a simplified GMP model in the second stage. By injecting the embedding vector into GANN, the GANN model is able to adapt to PAs with different states such as different bandwidths, power back-offs and other conditions. The proposed method is performed on a Doherty PA driven by the orthogonal frequency division multiplexing (OFDM) signals for experimental validation. The proposed two stage DPD based on GANN is validated to compensate strong nonlinear distortions and achieve the similar good linearization as the other multistage method while having low complexity of the identification process under different bandwidths and power conditions.

Integrated Sensing And Communication (ISAC) is one of the key technologies of 6G networks. With the development of large-scale array technology, target positioning estimated by Direction of Arrival (DOA)is indispensable in ISAC application scenarios. To address the limitations of traditional uniform linear arrays in expanding array aperture effectively due to spatial sampling theorem constraints, a novel atomic norm minimization algorithm based on generalized coprime Arrays is proposed. Firstly, an optimized co-prime array model is constructed by compressing inter-element spacing and introducing sub-array dis placement. Then, by utilizing array interpolation to f ill the holes, a virtual uniform linear array is con structed to fully utilize virtual array elements. Finally, the properties of virtual domain atomic norms are an alyzed. The interpolation virtual array covariance ma trix is reconstructed in an off-grid manner to over come the basis mismatch problem. This is combined with the Multiple Signal Classification (MUSIC) al gorithm for precise DOA estimation. Simulation re sults demonstrate that the proposed method achieves higher estimation accuracy and greater degrees of free dom compared to traditional methods. USing M+N 1 physical elements can provide O(MN) degrees of freedom and achieve 1° resolution.

Efficient energy utilization in covert communication sustains covertness while assuring communication quality and efficiency. This paper investigates covert communication energy efficiency (EE) in direct uplink satellite-ground communications, focusing on enhancing system EE via optimized transmit beamforming and satellite orbit altitude selection. This paper first establishes an optimization problem to maximize system EE in a direct uplink satellite-ground covert communication scenario. To solve this non-convex optimization problem, it is decomposed into two subproblems and solved using the successive convex approximation (SCA) method. Based on the above methods, this paper proposes an overall iterative optimization algorithm. Simulation results demonstrate that the proposed algorithm surpasses the baseline in terms of system EE. Furthermore, they elucidate the correlation between the amount of information received by the receiver and the variations in the satellite's orbital altitude.

In IPv6 (Internet Protocol Version 6) networks, the Duplicate Address Detection (DAD) protocol plays a critical role in verifying that an address is unique before it is assigned to an interface. This protocol operates using ICMPv6 (Internet Control Message Protocol Version 6) messages specifically, Neighbor Solicitation (NS) and Neighbor Advertisement (NA) messages for sending requests and receiving responses, respectively. However, DAD’s lack of authentication for ICMPv6 messages makes it susceptible to Denial-of-Service (DoS) attacks. Many existing solutions attempt to address these vulnerabilities, but they often fall short due to factors like high computational complexity, increased system demands, or reliance on centralized structures, all of which can hinder practical use. This paper introduces an alternative method that employs extended headers to enable the creation of multiple addresses for multiple Solicited Node Multicast Addresses (SNMA). By enhancing the efficiency of address configuration, this approach outperforms the traditional single address DAD. The evaluation of the proposed solution against the standard DAD protocol considers several performance metrics, including processing time, bandwidth usage, CPU and RAM utilization, and network overhead. Although this method may use slightly more resources, the resulting improvements in security, efficiency, and accuracy in unique IP address assignment make it a viable and effective solution.

Modulation signals are likely to experience channel fading when transmitted over wireless channels, and channel blind equalization is a powerful method to combat fading and guarantee high transmission quality and efficiency because no pilot information is required. In this paper, we propose a dualbranch blind equalization autoencoder with hybrid attention mechanisms, referred to as DBeA-HA, to restore various modulation signal waveforms. In the main branch, we employ multi-layer depthwise separable convolutions (DSC) with rich residual connections and squeeze-and-excitation (SE) mechanism in our encoder to extract channel features of faded signals at different resolutions, while reducing complexity. The auxiliary branch is constructed by incorporating a lightweight residual temporal dilated convolution to capture temporal correlations of faded signals. Additionally, the convolutional block attention module (CBAM) and multi-head self-attention (MHSA) mechanisms are applied within and between branches, respectively, to further enhance feature extraction and fusion capabilities. By integrating two branches effectively, the proposed method achieves high equalization performance and keeps low complexity. Experimental results reveal that in signal-to-noise ratio (SNR) ranging from -12 dB to 8 dB with six-path fading, our DBeA-HA effectively compensates for the effects of fading channels and surpasses existing equalization methods. For demodulation, compared with“ResNet+De” and the least mean square (LMS) methods, our DBeA-HA reduces the BER of QPSK by an average of 32.66% and 72.98%, and reduces the BER of 16QAM by an average of 19.41% and 55.33%, respectively, with a moderate level of complexity.

As the demand for ocean exploration and exploitation increases, acoustic communication holds paramount importance and finds extensive applications ranging from scientific data collection at seabed stations to marine geological research. However, data transmission rate of underwater acoustic communication is constrained by the narrow bandwidth due to severe attenuation of high-frequency sounds underwater. By introducing orbital angular momentum (OAM) into the realm of acoustics, there arises a significant enhancement in the transmission capacity and spectral efficiency of underwater acoustic communication systems. This paper aims to study the concept and evolution of orbital angular momentum, elucidate its impact on the field of underwater acoustic communication, and offers insights into future development trends.

Previous methods of mitigating Non-Line_x005fof-Sight (NLOS) errors usually pre-assumed that the NLOS errors satisfy the interval set. However, these methods exhibit notable shortcoming: The interval set is too conservative to exploit fully the NLOS error properties. To address this shortcoming, we consider using the data-driven method to construct error sets. Precisely, We use the k-means method to cluster the ranging errors of the training points in the target region, thus constructing the data-driven error sets.Based on the data-driven error sets, we formulate the source localization problem as a positive semi-definite matrix form with uncertainty. Consider the number of training points that can determine the data-driven error set bounds. To ensure high-accuracy localization even with a few training points. Subsequently,we use robust optimization concepts and some mathematical methods to eliminate the uncertainty. Finally, we obtain a solvable robust optimization problem. The experiments demonstrate that the proposed method achieves better localization performance even with fewer training points.

Congested link detection (CLD) has attracted more and more attention due to the rapid development of data traffic and services. In this work, we proposed a novel compressive sensing (CS) aided deep learning CLD scheme. Firstly, considering the network tomography structure, we proposed a CS-based preliminary CLD process to estimate the congestion probabilities for each link by utilizing the sparsity of congestions, it enables a decrease in the number of monitors needed, which in turn, enhances the flexibility and practicality of our scheme in various real-world scenarios. Then, based on the CS aided preliminary estimation, a long-short term memory network (LSTMN) is exploited to extract the time relationship of the congested states for each link, which can improve the accuracy of CLD. Moreover, since LSTMN utilizes the CS-aided preliminary estimation results to extract time relationships, our proposed scheme can reduce the monitoring cost and improve CLD accuracy. Ultimately, the simulation results substantiate the efficacy of our proposed scheme.

Hierarchical Federated Learning (HFL) is a distributed machine learning paradigm tailored for multi-tiered computation architectures, which supports massive access to devices' models simultaneously. To enable efficient HFL, it is crucial to design suitable incentive mechanisms to ensure that devices actively participate in local training. In this paper, we design two-level incentive mechanisms based on game theory for HFL in a device-edge-cloud coordinating architecture, aiming at encouraging the participation of entities in each level. To be specific, in the design of the lower-level incentive mechanism, we propose a coalition formation game to optimize the device-edge association and bandwidth allocation. We first develop efficient coalition partitions based on preference rules for optimizing device-edge association, which can be proven to be stable by constructing a potential function. Then, we develop a gradient projection method to optimally allocate bandwidth among the coalitions. In the upper-level game, we design a Stackelberg game algorithm to jointly maximize the utilities of the cloud and each edge server. The Stackelberg game algorithm is able to determine the optimal number of aggregations at each edge server, as well as the reward provided by the cloud server to each edge server for the performance improvement due to the edge aggregations. Numerical results indicate that the proposed method outperforms recent benchmark schemes, including the Accelerated device coalition formation (ADCF) algorithm and the Max-Q mechanism. Furthermore, our proposed method demonstrates significant accuracy improvements on real datasets, such as CIFAR-10, compared to the three benchmark schemes.

This paper centres on achieving the maximization of weighted throughput (WTP) in a multiuser cell-free massive multiple-input multiple-output (mMIMO) system with both finite blocklength (FBL) and infinite blocklength (INFBL), which is conducted against the backdrop of constrained time-frequency resources. We aim to ensure a quality of service (QoS) for all users, particularly in the FBL scenario, maintaining an acceptable latency and block error rate (BLER). To counteract the impact of reduced DoF of channel matrix due to a large number of users access, which leads to decreased system performance, we strive to optimize WTP by scheduling multiple users to different resource elements (RE) and applying precoding operation accordingly, subject to the limitations imposed by total power consumption per time slot and requisite QoS parameters. Simulation results demonstrate the superiority of the proposed multiuser processing (MUP) scheme over both single-user processing (SUP) and alluser processing (AUP) alternatives, and the proposed iterative algorithm based on genetic algorithm (GA) achieves up to 49.36% system performance gains compared to the benchmark algorithms. This substantiates the efficacy of our method in enhancing network performance and user satisfaction.

Non-orthogonal multiple access (NOMA) techniques assisted mobile edge computing (MEC) has attracted attention due to its advantages in reducing the energy consumption and the latency of MEC offloading. One of the critical challenges is how to efficiently allocate limited resources and reasonably group users to enhance system performance. In this work, we propose a hybrid NOMA assisted MEC offloading scheme based on user grouping. Our objective is to minimize the overall system energy consumption by jointly optimizing user grouping, transmit power, transmit time and the offloading ratio. The optimization problem is solved in two stages: user grouping and resource optimization. In the first stage, users are initially grouped based on channel gains, and then regrouped according to their latency requirements to reduce inter-user interference, thereby achieving more efficient resource allocation. In the second stage, the problem is simplified to a single-group energy minimization problem. By applying KKT conditions, a suboptimal solution for each user’s transmission power, transmission time, and offloading ratio is derived. Simulation results indicate that, under the same conditions, the proposed scheme achieves lower system energy consumption compared to other benchmark schemes.

The complexity of underwater environments renders underwater acoustic signals vulnerable to various forms of noise during transmission, creating significant challenges for signal demodulation tasks. This paper presents a novel demodulation method for multi-class single-carrier underwater acoustic signals. Our approach employs two innovative structures for modeling in the time and frequency domains, integrating these features for comprehensive discrimination. Specifically, we introduce a High-Efficient Convolution (HEC) Block to extract time-domain waveform features and a Local-Global Attention (LGA) structure for time-frequency features, utilizing cross-attention to fuse these features. This method enables the network to learn hidden frequency, phase, and amplitude characteristics within high-dimensional features, effectively capturing both fine-grained local and long-distance global features. A classifier is then constructed to categorize multi-class modulation signals, completing the demodulation process. Simulation results highlight the method's exceptional performance: in a Gaussian channel with a Signal-to-Noise Ratio (SNR) of 0 dB, the demodulation error rates for 2PSK, 2FSK, 2ASK, 4PSK, 4FSK, and 8PSK signals are all below 0.01, while the error rate for 16QAM modulated signals is less than 0.1. Additionally, validation using BELLHOP simulation data and real-world data collected from the Yellow Sea further demonstrates the proposed method's remarkable noise resistance and demodulation capabilities.

High penetration, high-resolution ranging, high efficiency, low cost, low latency, high security, and anti-multipath nature of ultra-wideband technology make it a good candidate for precise ranging and communication. In the existing iterative reference enhancement (IRE) techniques, the erroneous detection of a symbol in the previous iteration significantly deteriorates its detection. In this paper, the effect of the symbol on its own detection in the next iteration is called the self-effect. We show that if the symbol is incorrectly detected, the self-effect appears as a detrimental bias term in its decision statistic causing a significant performance loss. In this paper, we propose a novel IRE technique to exclude the self-effect and achieve a considerable performance gain in both detection and time-of-arrival (ToA) estimation. The proposed IRE technique outperforms the existing IRE technique over IEEE 802.15.3a channel model 4 by at least $2.5$ dB in detection and $6$ dB in ToA estimation with less complexity, higher speed, and less processing burden. Moreover, it achieves the optimal localization performance for $E_b/N_0>20$ dB. Furthermore, we develop an accurate analysis framework to calculate the bit-error probability of the proposed IRE method in the presence of monobit quantization, noise, and multipath.

In the sixth generation mobile communication (6G) system, Non-Terrestrial Networks (NTN), as a supplement to terrestrial network, can meet the requirements of wide area intelligent connection and global ubiquitous seamless access, establish intelligent connection for wide area objects, and provide intelligent services. Due to issues such as massive access, doppler shift, and limited spectrum resources in NTN, research on resource management is crucial for optimizing NTN performance. In this paper, a comprehensive survey of multi-pattern heterogeneous NTN resource management is provided. Firstly, the key technologies involved in NTN resource management is summarized. Secondly, NTN resource management is discussed from network pattern and resource pattern. The network pattern focuses on the application of different optimization methods to different network dimension communication resource management, and the resource type pattern focuses on the research and application of multi-domain resource management such as computation, cache, communication and sensing. Finally, future research directions and challenges of 6G NTN resource management are discussed.

Time synchronization is the basis for ensuring coordinated linkage among various subsystems of the train communication networks (TCN), and it is directly related to the safety of train operation. To enhance the reliability of time synchronization, multiple master clocks are typically used for redundancy. However, during the switching of grandmaster (GM) clock, significant time difference will be caused, making it difficult to ensure reliable synchronization. This paper proposes a reliable synchronization solution for the next-generation TCN (NG-TCN). Specifically, the solution adopts a networking architecture based on the SDN paradigm, integrating a synchronization controller for synchronization configuration and compensation management. Furthermore, a reliable synchronization method based on intelligent prediction is proposed for the compensation of time difference during GM switching. This paper builds a network testbed based on the GMs and TSN switches, and evaluates the reliable synchronization performance according to the 1-PPS outputs from the clocks. Experimental results show that the proposed solution can maintain the performance of synchronization accuracy within 1 us during the GM switching.

Currently, with the development of satellite communication, there are many challenges faced by satellite communication. Scholars have attempted to use the modern mathematical theory of differential geometry to solve problems in the fields of information and communication, and this emerging theoretical system is called information geometry, which is considered a revolutionary theory of information. Its application in the field of satellite communication has become a research hotspot and future development direction.This paper provides a detailed overview of the application of information geometry in satellite communication systems. Firstly, the feasibility of geometric analysis of satellite communication signals is explored . Subsequently, the basic concepts of information geometry are introduced. Then, the paper reviews the applications of information geometry in channel coding, on-board filtering, satellite spectrum sensing, interference signal detection techniques, and incoming wave direction estimation. In addition, this paper presents the information geometry representation of polar code BP decoding, the on-board nonlinear filtering process based on information geometry, and the satellite interference signal detection process based on information geometry. Finally, a summary and outlook are provided.

Multiple UAVs cooperative target search has been widely used in various environments, such as emergency rescue and traffic monitoring. However, uncertain communication network among UAVs exhibits unstable links and rapid topological fluctuations due to mission complexity and unpredictable environmental states. This limitation hinders timely information sharing and insightful path decisions for UAVs, resulting in inefficient or even failed collaborative search. Aiming at this issue, this paper proposes a multi-UAV cooperative search strategy by developing a real-time trajectory decision that incorporates autonomous connectivity to reinforce multi-UAV collaboration and achieve search acceleration in uncertain search environments. Specifically, an autonomous connectivity strategy based on node cognitive information and network states is introduced to enable effective message transmission and adapt to the dynamic network environment. Based on the fused information, we formalize the trajectory planning as a multi-objective optimization problem by jointly considering search performance and UAV energy harnessing. A multi-agent deep reinforcement learning based algorithm is proposed to solve it, where the reward-guided real-time path is determined to achieve an energy efficient search. Finally, extensive experimental results show that the proposed algorithm outperforms.