Due to the strong error correction ability for short messages, Polar code is applied in 5G system for control channels, and is also applied in space communications. In space communication systems, polar decoder can be efficiently implemented by SRAM-FPGA. However, SRAM-FPGA is very sensitive to the soft errors caused by cosmic particles, and Single Event Upsets (SEUs) is the most common effect. In particular, SEUs on the configuration memory of SRAM-FPGA will change the circuit functionality. Therefore, protection of SRAM-FPGA based polar decoders against SEUs is of great significance. In this paper, we first evaluate the reliability of the FPGA based polar decoder by the fault injection experiments. Then, an Enhanced Duplication with Comparison (E-DWC) based protection scheme is proposed to protect the decoder from SEUs on configuration memory. The hardware evaluation results indicate that the proposed scheme can almost completely eliminate the impact of SEUs on the decoder, and the hardware overhead is 2.11 times that of the unprotected decoder.

Large-scale antennas array is one of the key technologies in future wireless communication, which results in the users operating in the mixed far- and near-filed regions. In this work, we investigate the potential of beam focusing and steering in the mixed field to improve the performance of the simultane- ous wireless information and power transfer (SWIPT) in the intelligent reflecting surface (IRS) aided cell- free systems, where our goal is to maximize the har- vested energy of all receivers by jointly optimizing the transmit and reflective precoding vectors/matrices for the full digital (FD) and hybrid (HY) architec- tures. The formulated non-convex optimization prob- lem with unit-modulus constraints and coupled con- straints is challenging to solve. To address the issue, we transform the transmit and reflective precoding op- timization problem into two convex subproblems for the FD architecture. Then, based on the obtained so- lutions in FD architecture, we propose a block coor- dinate descent for subspace decomposition (BCD-SD) algorithm to design digital and analog precoding vec- tors/matrices for the HY architecture, where the per- formance of HY architecture is close to that of the FD architecture. Numerical results reveal that the pro- posed optimization schemes are more effective than the conventional schemes

This paper proposes a genetic optimization method for the construction of non-binary quasi-cyclic low-density parity-check (NB-QC-LDPC) codes with short block lengths. In our scheme, the initial template base matrices and the corresponding non-binary replacement matrices are constructed by the progressive edge growth algorithm and randomly generated, respectively. The genetic algorithm is then utilized to optimize the base matrices and the replacement ones. The simulation results show that the NB-QC-LDPC codes constructed by the proposed method achieve better decoding performance and lower implementation complexity compared to the existing NB-LDPC codes such as consultative committee for space data system and BeiDou satellite navigation system.

In this paper, an interference cancellation based neural receiver for superimposed pilot (SIP) in multi-layer transmission is proposed, where the data and pilot are non-orthogonally superimposed in the same time-frequency resource. Specifically, to deal with the intra-layer and inter-layer interference of SIP under multi-layer transmission, the interference cancellation with superimposed symbol aided channel estimation is leveraged in the neural receiver, accompanied by the pre-design of pilot code-division orthogonal mechanism at transmitter. In addition, to address the complexity issue for inter-vendor collaboration and the generalization problem in practical deployments, respectively, this paper also provides a fixed SIP (F-SIP) design based on constant pilot power ratio and scalable mechanisms for different modulation and coding schemes (MCSs) and transmission layers. Simulation results demonstrate the superiority of the proposed schemes on the performance of block error rate and throughput compared with existing counterparts.

Parallel concatenated codes~(PCC) are widely used in practice. Spatial coupling is useful for enhancing the performance of coding techniques. However, existing constructions of spatially coupled PCC~(SC-PCC) have inferior performance in either the waterfall or error-floor region. Recently, a novel construction of SC-PCC, called hybrid coupled PCCs~(HC-PCCs), was proposed. The preliminary results showed the advantages of HC-PCC in terms of iterative decoding thresholds. In this paper, we intend to give a further investigation on HC-PCC. First, we present the HC-PCC. Second, we derive the density evolution~(DE) equations of HC-PCCs over the binary erasure channels~(BECs). Third, we study the construction of rate-compatible HC-PCCs~(RC-HC-PCCs) via DE analysis. Fourth, we show that we can lower bound for the performance of an HC-PCC with a hybrid concatenated code~(HCC) ensemble. We then derive the weight enumerator of the resulting HCC, which is used in conjunction with the union bound to estimate the error-floor of HC-PCC. Finally, we show numerical results to demonstrate the impact of various parameters on the performance of HC-PCCs. By selecting the parameters appropriately, the performance of HC-PCCs can be significantly enhanced. Furthermore, the simulation results show that HC-PCCs outperform GSC-PCCs and HC-SCCs in the waterfall region.

Over the years and nowadays container technology is widely used among the communities for application deployment and maintenance. Especially, comparing with virtual machines, containers are lightweight and occupy fewer hardware resources since they share the kernel with the host system. However, due to the weak isolation of container mechanisms, the host system or other containers are vulnerable when a container is attacked exploiting the kernel vulnerabilities. Besides, unexpected control of container engines and tainted images are also threats to containers. Thus, it is important to know about the security patterns of these issues toward enhancing security. To this, in this paper, we present an empirical study of container escape, which is a kind of risk to gain permissions to take control of other containers or the host from one container. Basically, our study includes the common patterns, root causes, exploits, possible fixes, and many more. Particularly, we study nine (09) related vulnerabilities discovered in recent years by analyzing their root causes, deploying environments to simulate the attack and comparing the official patches. For some of these vulnerabilities, we also present alternative defense or fix methods. Additionally, we summarize our learning outcomes for each vulnerability, and propose further analysis using these experiences.

Spectrum sensing is crucial for enabling opportunistic spectrum access (OSA) in cognitive radio (CR). However, the accuracy of spectrum sensing can be compromised by several malicious attacks, such as primary user emulation attack (PUEA). A PUEA represents that the attacker sends imitated primary signal to mislead spectrum sensing results and thereby prevent the secondary users from accessing the idle spectrum band, thereby leading to a low spectrum utilization. In this paper, we propose a novel intelligent reflecting surface (IRS)-enhanced cooperative spectrum sensing (CSS) scheme and exploit IRS to suppress PUEA. The maximal ratio combining (MRC) is adopted as the fusion rule for CSS. To this end, an optimization problem is formulated to maximize the probability of detection by designing the weight coefficients of MRC at a fusion center (FC) and the phase shifts of the elements in IRS. To solve the non-convex problem with coupled variables, an efficient alternating optimization (AO) algorithm is proposed. In particular, the optimal weight coefficients are obtained by adopting the transformation of Rayleigh quotient. Then, semi-definite relaxation (SDR), Charnes-Cooper transformation and Gaussian Randomization are exploited to obtain the solutions of IRS phase shifts. Simulation results demonstrate that the proposed scheme is computationally efficient and can improve the sensing performance while reducing the impact of PUEA.

Reconfigurable intelligent surface (RIS) is an economically effective solution for improving the spectral efficiency and energy efficiency. The sum capacity of an RIS-aided multiple-input multiple-output (MIMO) broadcast channel (BC) depends on the phase shifts. This motivates us to investigate the problem of phase shift optimization in an RIS-aided MIMO BC via sum capacity maximization in this letter. More concretely, we consider a common scenario where there is only a finite number of discrete phase shifts available for each element of the RIS. By exploiting the duality between the MIMO BC and the MIMO multiple access channel, the RIS phase shifts are optimized along with the users' input covariance matrices through maximizing the sum capacity using an alternating optimization algorithm. In order to deal with the optimization subproblems, an accelerated algorithm based on water-filling with fast convergence is proposed to optimize the input covariance matrices, and a channel separation technique with reduced computational complexity is exploited for RIS phase shift optimization. Simulation results are illustrated to verify the effectiveness and superiority of the proposed method.

As a common marine atmospheric structure, evaporation duct can provide favorable conditions for the long-range propagation of terahertz waves. To further study and predict the propagation characteristics of terahertz waves in evaporation duct, a new method is proposed to calculate the variable atmospheric absorption loss. Combined with the parabolic equation, the propagation model is constructed. The propagation characteristics of terahertz waves under different propagation distances, frequencies and evaporation duct heights (EDHs) are studied. The results show that the propagation factor of terahertz waves is mainly manifested by atmospheric absorption, and the absorption loss increases with the increase of frequency. The effect of EDH on propagation factor is related to frequency and propagation distance, and the change of EDH is accompanied by the jitter of propagation factor. The higher the frequency and the greater the propagation distance, the greater the jitter amplitude of propagation factor caused by changing EDH. The influence of the heights of transmitting and receiving antennas is further discussed, and the optimal setting heights of antennas for 140 GHz terahertz wave at different distances are given, which provides theoretical basis and numerical reference for effective planning and propagation experiment of terahertz wireless communication systems on the sea.

With the rapid development of Intelligent Cyber-Physical Systems (ICPS), diverse services with varying Quality of Service (QoS) requirements have brought great challenges to traditional network resource allocation. Furthermore, given the open environment and a multitude of devices, enhancing the security of ICPS is an urgent concern. To address these issues, this paper proposes a novel trusted virtual network embedding (T-VNE) approach for ICPS based combining blockchain and edge computing technologies. Additionally, the proposed algorithm leverages a Deep Reinforcement Learning (DRL) model to optimize decision-making processes. It employs the policy-gradient-based agent to compute candidate embedding nodes and utilizes a breadth-first search (BFS) algorithm to determine the optimal embedding paths. Finally, through simulation experiments, the efficacy of the proposed method was validated, demonstrating outstanding performance in terms of security, revenue generation, and Virtual Network Request (VNR) acceptance rate.

The rapid increase in the number of Low Earth orbit (LEO) satellites launched into space has brought unprecedented spectrum pressure. In order to obtain more flexible available bandwidth, current LEO operators are seeking to share spectrum with existing GEO satellites. In order to improve the capacity of the LEO system in spectrum sharing scenarios without interfering with the GEO system, the resources allocation problem of the LEO satellite system is studied in this paper. By modeling the problem as a mixed integer nonlinear programming problem, and solving three sub-problems decomposed by it, a frequency multiplexing based beam hopping algorithm is proposed to reasonably allocate space, frequency and power resources of the LEO system. The simulation results show that the proposed algorithm not only ensures the signal quality of GEO users, but also effectively improves the LEO system capacity.

Differential pulse-position modulation (DPPM) can achieve a good compromise between power and bandwidth requirements. However, the output sequence has undetectable insertions and deletions. This paper proposes a successive cancellation (SC) decoding scheme based on the weighted Levenshtein distance (WLD) of polar codes for correcting insertions/deletions in DPPM systems. In this method, the WLD is used to calculate the transfer probabilities recursively to obtain likelihood ratios, and the low-complexity SC decoding method is built according to the error characteristics to match the DPPM system. Additionally, the proposed SC decoding scheme is extended to list decoding, which can further improve error correction performance. Simulation results show that the proposed scheme can effectively correct insertions/deletions in the DPPM system, which enhances its reliability and performance.

To address the contradiction between the explosive growth of wireless data and the limited spectrum resources, semantic communication has been emerging as a promising communication paradigm. In this paper, we thus design a speech semantic coded communication system, referred to as Deep-STS (i.e., Deep-learning based Speech To Speech), for the low-bandwidth speech communication. Specifically, we first deeply compress the speech data through extracting the textual information from the speech based on the conformer encoder and connectionist temporal classification decoder at the transmitter side of Deep-STS. In order to facilitate the final speech timbre recovery, we also extract the short-term timbre feature of speech signals only for the starting 2s duration by the long short-term memory network. Then, the Reed-Solomon coding and hybrid automatic repeat request protocol are applied to improve the reliability of transmitting the extracted text and timbre feature over the wireless channel. Third, we reconstruct the speech signal by the mel spectrogram prediction network and vocoder, when the extracted text is received along with the timbre feature at the receiver of Deep-STS. Finally, we develop the demo system based on the USRP and GNU radio for the performance evaluation of Deep-STS. Numerical results show that the accuracy of text extraction approaches 95%, and the mel cepstral distortion between the recovered speech signal and the original one in the spectrum domain is less than 10. Furthermore, the experimental results show that the proposed Deep-STS system can reduce the total delay of speech communication by 85% on average compared to the G.723 coding at the transmission rate of 5.4 kbps. More importantly, the coding rate of the proposed Deep-STS system is extremely low, only 0.2 kbps for continuous speech communication. It is worth noting that Deep-STS with lower coding rate can support the low-zero-power speech communication, unveiling a new era in ultra-efficient coded communications.

Data confidentiality relies on cryptographic algorithms to a great extent. The critical nonlinear building block of cryptography is the S-box which plays a role in the design of various cryptographic algorithms to improve security and complexity. Therefore, a kind of newly dynamic and involutory S-box (DINV-S) was proposed in this paper. DINV-S is comprised of two-round Feistel structure. The proposed method has the capability to generate 256 parameter-dependent dynamic S-boxes within the F-function of each round in the Feistel structure by linearly reformulating any 8 × 8 static S-box. Theoretical analysis shows that differential uniformity and nonlinearity of DINV-S can be consistent with the static S-box. Additionally, specific experiments are carried out using AES S-box in conjunction with this method. The generated S-boxes in the F-function are found to outperform AES S-box in terms of average iteration cycle. The dynamic S-box derived by AES S-box has the optimal differential probability (DP=4/256), the optimal linear probability (LP=16/256), and the highest algebraic degree (def(S)=7).

Given the increasing demand for secure transmission and advancements in steganalysis techniques, it is necessary to design a steganography algorithm with enhanced steganography performance to ensure information security. In this paper, we preposed a semantic-weighted image steganography algorithm (SISA) based on convolutional neural network (CNN) . Firstly, SISA is proposed to trade off steganography capacity and performance. SISA adopts an autoencoder consisting of a semantic-based preparation-network, a hiding-network, and a reveal-network. Subsequently, a semantic-weighted and secure-enhanced image steganography algorithm (SSISA) is further proposed to trade off steganography capacity, performance and security. SSISA adopts a generative adversarial network (GAN) consisting of a generative network and a discrimination network. SSISA makes minimal sacrifices in steganography performance while significantly improving security. Experimental results show that two proposed steganography algorithms not only realize large-capacity image steganography, but also achieve state-of-the-art steganography performance. The average pixel deviation (APD) between the embedded images and the cover images is reduced by up to about 36.73% using the proposed SISA, compared with the existing algorithms that do not consider semantic weighting. The steganalysis detection rate is reduced by up to about 6.75% using the proposed SSISA, compared with the existing algorithms that do not consider security enhancement.

The quality of spectrum sensing plays a significant role in determining the outage probability during the data transmission phase in an interweave cognitive radio network. If the secondary user (SU) fails to detect the primary user (PU) activity, it can result in interference that limits the system performance. Additionally, since the wireless medium is broadcast in nature, there is a risk of eavesdroppers intercepting the cognitive users' data. Therefore, it is crucial to consider secrecy in the system analysis. In this paper, we analyze the secrecy outage probability (SOP) at the secondary receiver and derive the secret diversity gain for an interweave cognitive multiple-input-multiple-output (MIMO) fading channel in the presence of an eavesdropper. Our study takes into account the effects of the fading channel, the PU interference, and the eavesdropper on both spectrum sensing and data transmission phases. We demonstrate that utilizing all the antennas for sensing eliminates the limiting effects of missed detection probability and PU interference on the secret diversity gain. As a result, the cognitive user can achieve the same level of secret diversity gain as a conventional non-cognitive system (CNCS). Our analytical results are further validated through simulations.

Virtualization is an indispensable part of the cloud for the objective of deploying different virtual servers over the same physical layer. However, the increase in the number of applications executing on the repositories results in increased overload due to the adoption of cloud services. Moreover, the migration of applications on the cloud with optimized resource allocation is a herculean task even though it is employed for minimizing the dilemma of allocating resources. In this paper, a Fire hawk Optimization-Enabled Deep Learning Scheme (FHOEDLS) is proposed for minimizing the overload and optimizing the resource allocation on the hybrid cloud container architecture for migrating interoperability-based applications This FHOEDLS achieves the load prediction through the utilization of deep CNN-GRU-AM model for attaining resource allocation and better migration of applications. It specifically adopted the Fire Hawk Optimization Algorithm (FHOA) for optimizing the parameters that influence the factors that aid in better interoperable application migration with improved resource allocation and minimized overhead. It considered the factors of resource capacity, transmission cost, demand, and predicted load into account during the formulation of the objective function utilized for resource allocation and application migration. The cloud simulation of this FHOEDLS is achieved using a container, Virtual Machine (VM), and Physical Machine (PM). The results of this proposed FHOEDLS confirmed a better resource capability of 0.418 and a minimized load of 0.0061.

This paper studies the problem of jamming decision-making for dynamic multiple communication links in wireless communication networks (WCNs). We propose a novel jamming channel allocation and power decision-making (JCAPD) approach based on multi-agent deep reinforcement learning (MADRL). In high-dynamic and multi-target aviation communication environments, the rapid changes in channels make it difficult for sensors to accurately capture instantaneous channel state information. This poses a challenge to make centralized jamming decisions with single-agent deep reinforcement learning (DRL) approaches. In response, we design a distributed multi-agent decision architecture (DMADA). We formulate multi-jammer resource allocation as a multi-agent Markov decision process (MDP) and propose a fingerprint-based double deep Q-Network (FBDDQN) algorithm for solving it. Each jammer functions as an agent that interacts with the environment in this framework. Through the design of a reasonable reward and training mechanism, our approach enables jammers to achieve distributed cooperation, significantly improving the jamming success rate while considering jamming power cost, and reducing the transmission rate of links. Our experimental results show the FBDDQN algorithm is superior to the baseline methods.

The telecommunications industry is becoming increasingly aware of potential subscriber churn as a result of the growing popularity of smartphones in the mobile Internet era, the quick development of telecommunications services, the implementation of the number portability policy, and the intensifying competition among operators. At the same time, users’ consumption preferences and choices are evolving. Excellent churn prediction models must be created in order to accurately predict the churn tendency, since keeping existing customers is far less expensive than acquiring new ones. But conventional or learning-based algorithms can only go so far into a single subscriber’s data; they cannot take into consideration changes in a subscriber’s subscription and ignore the coupling and correlation between various features. Additionally, the current churn prediction models have a high computational burden, a fuzzy weight distribution, and significant resource economic costs. The prediction algorithms involving network models currently in use primarily take into account the private information shared between users with text and pictures, ignoring the reference value supplied by other users with the same package. This work suggests a user churn prediction model based on Graph Attention Convolutional Neural Network (GAT-CNN) to address the aforementioned issues. The main contributions of this paper are as follows: Firstly, we present a three-tiered hierarchical cloud-edge cooperative framework that increases the volume of user feature input by means of two aggregations at the device, edge, and cloud layers. Second, we extend the use of users’ own data by introducing self-attention and graph convolution models to track the relative changes of both users and packages simultaneously. Lastly, we build an integrated offline-online system for churn prediction based on the strengths of the two models, and we experimentally validate the efficacy of cloud-side collaborative training and inference. In summary, the churn prediction model based on Graph Attention Convolutional Neural Network presented in this paper can effectively address the drawbacks of conventional algorithms and offer telecom operators crucial decision support in developing subscriber retention strategies and cutting operational expenses.

Progressive pitch proposed by OneWeb is a promising solution to prevent huge non-geostationary orbit (NGSO) satellite constellations from interfering with geostationary orbit (GSO) satellite systems. However, the traditional progressive pitch primarily focuses on the protection of GSO systems, while the energy efficiency of NGSO systems is not optimized. In this paper, we propose an energy-efficient progressive pitch strategy for NGSO satellites in the spectral coexistence scenario between GSO and NGSO systems. By combining power control and spatial isolation, this strategy can not only protect the GSO system from interference, but also reduce power consumption and improve the quality of service for the NGSO system. Specifically, we formulate a joint pitch angle and beam power optimization problem to maximize energy efficiency of the NGSO system. Due to the fractional form of the energy efficiency function, we reformulate it into an approximately equivalent convex one by quadratic transform. Besides, the non-convex interference constraint can be solved by a low-complexity binary search method. Lastly, a joint power and pitch angle alternating optimization (JPPA-AO) algorithm is proposed to deal with this multivariate optimization problem. Simulation results demonstrate the proposed strategy outperforms conventional methods in communication capacity and energy reduction for the NGSO system.

This paper investigates the Reconfigurable Intelligent Surface (RIS)-aided MIMO covert communications in high-speed railway (HSR) scenario. In the scenario, RIS controls the phases of reflection elements dynamically to send the signal in the desired direction, which facilitates the covert communication between base station (BS) and train mobile relay (MR) in the existence of a watchful warden (Willie). To protect the desired transmission, it is assumed that MR sends out jamming signals with a varying power to confuse the Willie. Considering the Doppler spread caused by the time-varying wireless channel, the joint optimization problem of the BS beamforming matrix, MR beamforming matrix, and the RIS phase shifts is established to maximize the covert throughput. An alternating optimization (AO) method for handling non convex problems is proposed based on coupling effects and the constraints of constant modulus, and a semidefinite relaxation method is provided. Finally, we achieve the optimal solutions of the multi-objective optimization problem by interior-point method. The simulation results demonstrate that the proposed algorithm exhibits the superior robustness and covert performances in high-speed railway scenarios.

Mobile edge networks (MENs) based on integrated computing and communication are the main trends in the future with the emergence of mobile edge computing. This development brings computing, storage, and network resources closer to the edge, improving the quality of experience for users. However, the large-scale deployment of MENs has led to increased energy consumption and resource synergy issues due to network densification and service diversification. Therefore, improving the resource synergy efficiency of MENs for energy efficiency (EE) is a key issue for future networks. In this paper, we first review EE optimization techniques in traditional cellular networks and discuss the opportunities and challenges these techniques present for MENs. Furthermore, we examine the evolution and role of typical resource sharing scenarios and key technologies in MENs. Furthermore, we explore ways to optimize EE for typical mobile scenarios. Lastly, we discuss the future challenges and directions in EE optimization research within MENs.

Deep learning techniques have been extensively validated and recognized in the field of modulation identification. However, in many practical communication scenarios, such as underwater acoustic communication systems and radar communication systems with various channel variations, it is difficult to obtain a large number of labeled signal samples. In such scenarios, deep learning-based modulation identification methods are prone to overfitting, leading to poor recognition performance. Therefore, improving the modulation identification performance in environments with limited signal samples is a challenging problem. For solving this problem, this paper proposes a method based on data enhancement and migration learning automatic modulation recognition. This method first uses a data enhancement mechanism proposed in this paper to achieve signal sample expansion. Then, it combines the reconstructed signal with the original signal to achieve data enhancement. Secondly, a transfer learning model called Residual Dense Long and Short Term Memory Networks (ResLDNN) is designed for modulation recognition. This model is trained on AMR2016.10a data as the source domain data and fine-tuned on a small number of signal samples as the target domain, ultimately achieving modulation recognition for small sample signals. In order to validate the effectiveness of the proposed data augmentation method, experiments were conducted on both the original dataset and the augmented dataset. The results of the experiments demonstrate that the data augmentation method significantly improves the modulation recognition rate of the model. Additionally, experiments were conducted on datasets with different numbers of signal samples to verify the recognition performance of the ResLDNN model in a small sample environment. The experimental results show that the proposed ResLDNN model achieves higher recognition accuracy than the baseline model in a small sample environment, further confirming that the proposed method based on data augmentation and transfer learning effectively addresses the modulation recognition problem in a small sample environment.

This paper considers a multi-antenna access point (AP) transmitting secrecy message to a single-antenna user in the presence of a single-antenna illegal eavesdropper (Eve) and proposes a double active reconfigurable intelligent surfaces (DARISs) assisted physical layer security (PLS) scheme denoted by DARISs-PLS to protect the secrecy message transmission. We formulate a secrecy rate maximization problem for the proposed DARISs-PLS scheme by considering a power budget constraint for the two active reconfigurable intelligent surfaces (ARISs) and AP. To address the formulated optimization problem, we jointly optimize the reflecting coefficients for the two ARISs and the beamforming at the AP in an iterative manner by applying Dinkelbach based alternating optimization (AO) algorithm and a customized iterative algorithm together with the semidefinite relaxation (SDR). Numerical results reveal that the proposed DARISs-PLS scheme outperforms the double passive reconfigurable intelligent surfaces-assisted PLS method (DPRISs-PLS) and single ARIS-assisted PLS method (SARIS-PLS) in terms of the secrecy rate.

In this paper, we examine an illegal wireless communication network consisting of an illegal user receiving illegal signals from an illegal station and propose an active reconfigurable intelligent surface (ARIS)-assisted multi-antenna jamming (MAJ) scheme denoted by ARIS-MAJ to interfere with the illegal signal transmission. In order to strike a balance between the jamming performance and the energy consumption, we consider a so-called jamming energy efficiency (JEE) which is defined as the ratio of achievable rate reduced by the jamming system to the corresponding power consumption. We formulate an optimization problem to maximize the JEE for the proposed ARIS-MAJ scheme by jointly optimizing the jammer's beamforming vector and ARIS's reflecting coefficients under the constraint that the jamming power received at the illegal user is lower than the illegal user's detection threshold. To address the non-convex optimization problem, we propose the Dinkelbach based alternating optimization (AO) algorithm by applying the semidefinite relaxation (SDR) algorithm with Gaussian randomization method. Numerical results validate that the proposed ARIS-MAJ scheme outperforms the passive reconfigurable intelligent surface (PRIS)-assisted multi-antenna jamming (PRIS-MAJ) scheme and the conventional multi-antenna jamming scheme without RIS (NRIS-MAJ) in terms of the JEE.

Integrated sensing and communication(ISAC) is one of the key technologies for 6G networks. Its application scenarios include localization,recognition, environment perception and reconstruction,etc. Array signal processing is an important method for achieving target localization and recognition.However, the ideal array used in traditional localization methods cannot meet the requirements of ISAC scenarios. In practical scenarios, there may be sensor position errors in the array. Therefore, studying the performance of mixed-field signal source parameter estimation under imperfect arrays is crucial. This paper firstly applies the propagator method (PM) to the ideal linear and nonlinear array model of mixed field scenarios. Then, it constructs the signal model and corresponding parameter estimation methods for coprime arrays with sensor position errors. Finally, it thoroughly analyzes the parameter estimation performance for both ideal arrays and imperfect arrays. Experimental simulations validate the feasibility and effectiveness of the PM-multiple signal classification(MUSIC) algorithm for mixed-field signal source localization. The impact of snapshot number and perturbation factor on the mean squared error of parameter estimation is also analyzed. The experimental results demonstrate that increasing the snapshot number or reducing the perturbation factor can significantly improve the performance of the algorithm for parameter estimation.

Pre-trained language models (PLMs) that use semi-supervised learning have recently become much more widely used in few-shot learning because they exhibit remarkable capabilities. The self-training techniques and introduction of a data-efficient few-shot learner of language model (SFLM) is a recent method for few-shot learning using semi-supervised learning. However, fine-tuning the SFLM model is problematic because it requires repetitive updating of all model parameters with very little data. Using a parameter-efficient fine-tuning method can reduce the difficulty of fine-tuning by reducing the number of parameters during fine-tuning. However, its use will inevitably cause the problem of over-fitting the model due to the increased learning rate and affect the fine-tuning results. In this paper, we focus on combining parameter-efficient fine-tuning methods with semi-supervised few-shot learning to reduce the problem of influencing the model results caused by increasing the learning rate of the combined model. By modifying the goal of SFLM to provide Denser-supervision Loss and use the AdapterBias model to improve fine-tuning efficiency under the model fine-tuning. We trained on twelve datasets and achieved higher accuracy on nine datasets. Moreover, we also prove that without semi-supervised data, our Ada-SFLM model still has higher overall accuracy than some self-supervised models like LM-BFF.

(Quasi-)closed-form results for the statistical properties of unmanned aerial vehicle (UAV) air-to-ground channels are derived for the first time using a novel spatial-vector-based method from a three-dimensional (3-D) arbitrary-elevation one-cylinder model. The derived results include a closed-form expression for the space-time correlation function and some quasi-closed-form ones for the space-Doppler power spectrum density, the level crossing rate, and the average fading duration, which are shown to be the generalizations of those previously obtained from the two-dimensional (2-D) one-ring model and the 3-D low-elevation one-cylinder model for terrestrial mobile-to-mobile channels. The close agreements between the theoretical results and the simulations as well as the measurements validate the utility of the derived channel statistics. Based on the derived expressions, the impacts of some parameters on the channel characteristics are investigated in an effective, efficient, and explicable way, which leads to a general guideline on the manual parameter estimation from the measurement description.

In response to the current gaps in effective proactive defense methods within application security and the limited integration of security components with applications, this paper proposes a biomimetic security model, called NeuroShield, specifically designed for web applications. Inspired by the "perception-strategy-effect-feedback" mechanism of the human nervous control system, the model integrates biomimetic elements akin of neural receptors and effectors into applications. This integration facilitates a multifaceted approach to security: enabling data introspection for detailed perception and regulation of application behavior, providing proactive defense capabilities to detect and block security risks in real-time, and incorporating feedback optimization to continuously adjust and enhance security strategies based on prevailing conditions. Experimental results affirm the efficacy of this neural control mechanism-based biomimetic security model, demonstrating a proactive defense success rate exceeding 95\%, thereby offering a theoretical and structural foundation for biomimetic immunity in web applications.

?A terrestrial relay-aided reconfigurable intelligent surface (RIS) system with decode?, ?re-encode and forward (DRF) relaying scheme is presented where the RIS effectively contributes to both source-to-destination and relay-to-destination signaling?. ?While in the conventional decode and forward (DF) relaying scheme?, ?the source signal is merely duplicated in the relay and the time intervals are equally allocated to the source and relay nodes?, ?this paper considers DRF relaying scheme where versatile time-sharing is adopted for the source and relay nodes which can be optimized based on the relative coordinates of the involved nodes?. ?Two protocols namely unidirectional connection (UC) and bidirectional connection (BC) are proposed based on the source awareness from the relay's successful reception?. ?The outage probability (OP) performance for both protocols and both DF and DRF relaying schemes is analyzed and tight approximations are obtained?. ?The numerical results show the out-performance of the DRF over the DF relaying scheme in the both UC and BC protocols?. ?Equipped with the obtained system OP?, ?the system throughput is defined and the optimum system throughput is obtained by optimizing the system rate and the time-sharing between the source and the relay?. ?Analytical results are corroborated in the numerical examples?.

In this paper, we propose a sub-6GHz channel assisted hybrid beamforming for mmWave system under both line-of-sight (LOS) and non-line-of-sight (NLOS) scenarios without mmWave channel estimation. Meanwhile, we resort to the self-supervised approach to eliminate the need for labels, thus avoiding the accompanied high cost of data collection and annotation. We first construct the dense connection network (DCnet) with three modules: the feature extraction module for extracting channel characteristic from a large amount of channel data, the feature fusion module for combining multidimensional features, and the prediction module for generating the HBF matrices. Next, we establish a lightweight network architecture, named as LDnet, to reduce the number of model parameters and computational complexity. The proposed sub-6GHz assisted approach eliminates mmWave pilot resources compared to the method using mmWave channel information directly. The simulation results indicate that the proposed DCnet and LDnet can achieve the spectral efficiency that is superior to the traditional orthogonal matching pursuit (OMP) algorithm by 13.66% and 10.44% under LOS scenarios and by 32.35% and 27.75% under NLOS scenarios, respectively. Moreover, the LDnet achieves 98.52% reduction in the number of model parameters and 22.93% reduction in computational complexity compared to DCnet.

In integrated platforms, due to overlapping frequency bands between devices, co-site interference (CI) emitted by high-power transmitter (Tx) phased arrays interferes with co-sited receiver (Rx) arrays. To combat CI, we propose a CI cancellation (CIC) method based on an auxiliary transmit array, where the auxiliary array is placed near the receiver (Rx) array and transmits a reconstructed CI signal to cancel CI at the Rx antenna interface. First, a system model is established and the beamforming vector of the auxiliary array is designed for CIC. Then, the closed-form expression of minimum residual interference (RI) power is derived to study the relation between CIC performance and channel conditions. According to this relation, locations of arrays are obtained to achieve two objectives, i.e., effective CIC and minimizing the effect of the auxiliary array on the original function of the local Tx. The results show that the proposed method can cancel the CI close to the noise floor and has almost no effect on the beampattern of the Tx array in a proper location arrangement. Specifically, the proposed method can increase 7.9 dB beamforming gain when compared with the CIC method by extending the original local Tx array under the same hardware cost.

Covert underwater acoustic communication (CUAC) has always played an extremely important role in the secure transmission of underwater information.Based on the fact that the eavesdropper will filter out the marine ambient noises after detecting them, this paper proposes a camouflage CUAC method. This method uses audio digital watermarking technology to embed the digital watermarking information containing secret information into the discrete cosine transform (DCT) domain of the original ship-radiated noise. Then the communication signals with high similarity to the original ship-radiated noise are generated by inverse discrete cosine transform (IDCT). That induces the eavesdropper, even if it detects the communication signals, to misjudge the communication signals as marine ambient noises and filter them out. Due to the scarcity of available underwater acoustic channel bandwidth, this paper proposes a grouping M-ary spread spectrum modulation technology based on multi-chaotic orthogonal combination sequences (MCOCS) to modulate secret information into digital watermark information, which can improve the frequency band utilization. Two objective evaluation methods, namely normalized correlation coefffcient and perceptual speech quality evaluation method based on PESQ (Perceptual evaluation of speech quality) algorithm, are used to evaluate the camouflage effect from two perspectives of waveform and auditory. Simulations validate the reliability, effectiveness, and covertness of the proposed CUAC method. A reliable communication rate of 87.4 bps was achieved in the simulations and the bit error rate is lower than 10?3.

Web data extraction has become a key technology for extracting valuable data from websites. At present, most extraction methods based on rule learning, visual pattern or tree matching have limited performance on complex web pages. Through analyzing various statistical characteristics of HTML elements in web documents, this paper proposes, based on statistical features, an unsupervised web data extraction method---traversing the HTML DOM parse tree at first, calculating and generating the statistical matrix of the elements, and then locating data records by clustering method and heuristic rules that reveal inherent links between the visual characteristics of the data recording areas and the statistical characteristics of the HTML nodes---which is both suitable for data records extraction of single-page and multi-pages, and it has strong generality and needs no training. The experiments show that the accuracy and efficiency of this method are equally better than the current data extraction method.

The proliferation of phone communications has escalated the risk of call security, prompting extensive research into phone tapping. Existing optical and trojan based techniques encounter challenges in conditions such as darkness and resistance from antivirus software. In this paper, we present mmwTapper, a system leveraging mmWave signals to measure the subtle vibrations of a phone caused by its speaker, enabling the tapping of the speech and identity of a remote anonymous caller. Due to the extremely weak vibration caused by the speaker, the signal-to-noise ratio of the detected caller speech is very low. To enhance caller speech clarity, we propose a multi-stage learning based caller speech enhancement network to sequentially denoise the amplitude and phase of the speech. Moreover, collecting samples is extremely challenging in tapping applications. To ensure accurate identification with a limited number of samples, we design a contrastive learning based caller identification network, which can be pre-trained with a substantial number of public samples to extract invariant speech features. We extensively evaluate the performance of our proposed mmwTapper under various conditions and the results demonstrate its effectiveness in tapping caller speech and identity, even with a limited number of samples.

Multimedia semantic communication has been receiving increasing attention due to its significant enhancement of communication efficiency. Semantic coding, which is oriented towards extracting and encoding the key semantics of video for transmission, is a key aspect in the framework of multimedia semantic communication. In this paper, we propose a facial video semantic coding method with low bitrate based on the temporal continuity of video semantics. At the sender's end, we selectively transmit facial keypoints and deformation information, allocating distinct bitrates to different keypoints across frames. Compressive techniques involving sampling and quantization are employed to reduce the bitrate while retaining facial key semantic information. At the receiver's end, a GAN-based generative network is utilized for reconstruction, effectively mitigating block artifacts and buffering problems present in traditional codec algorithms under low bitrates. The performance of the proposed approach is validated on multiple datasets, such as VoxCeleb and TalkingHead-1kH, employing metrics such as LPIPS, DISTS, and AKD for assessment. Experimental results demonstrate significant advantages over traditional codec methods, achieving up to approximately 10-fold bitrate reduction in prolonged, stable head pose scenarios across diverse conversational video settings.

Memory-unsafe programming languages, such as C/C++, are often used to develop system programs, rendering the programs susceptible to a variety of memory corruption attacks. Among these threats, just-in-time return-oriented programming (JIT-ROP) stands out as an advanced method for conducting code-reuse attacks, effectively circumventing code randomization safeguards. JIT-ROP leverages memory disclosure vulnerabilities to obtain reusable code fragments dynamically and assemble malicious payloads dynamically. In response to JIT-ROP attacks, several re-randomization implementations have been developed to prevent the use of disclosed code. However, existing re-randomization methods require recurrent re-randomization during program runtime according to fixed time windows or specific events such as system calls, incurring significant runtime overhead. In this paper, we present the design and implementation of PtrProxy, an efficient re-randomization approach on the AArch64 platform. Unlike previous methods that necessitate frequent runtime re-randomization or reply on unreliable triggering conditions, this approach triggers the re-randomization process by detecting the code page harvest operation, which is a fundamental operation of the JIT-ROP attacks, making our method more efficient and reliable than previous approaches. We evaluate PtrProxy on benchmarks and real-world applications. The evaluation results show that our approach can effectively protect programs from JIT-ROP attacks while introducing marginal runtime overhead.

In this paper, we integrate the 5G low-density parity-check (LDPC) coded modulation systems with hybrid shaping, where the centroid-based geometric shaping is implemented to remedy the performance loss of the many-to-one probabilistic shaping. Taking into account the fact that the 5G parity-check matrices have an uneven density in different parts, we elaborately design a simple row-column interleaver for the bit-interleaved coded modulation with iterative decoding (BICM-ID) system to allocate the ambiguous bits caused by the many-to-one mapping to the sparser parity part,resulting in the hybrid shaping for BICM-ID (HS-BICM-ID) system.For the multi-layer coding (MLC) system, we propose a cyclic shift mapping scheme to allocate evenly the ambiguous bits to each layer, resulting in the hybrid shaping for MLC(HS-MLC) system. By treating the multiple binary codes of the HS-MLC system as a non-binary LDPC code,we present a joint decoding algorithm, which does not require iterative processing between demapping and decoding. Numerical results show that: 1) the HS-BICM-ID can obtain shaping gains of about 0.2 dB and 1.0 dB compared to the constant composition distribution matching (CCDM) shaping and the geometric shaping, respectively, while it can obtain a shaping gain of about 1.2 dB compared to the scheme with uniform input; 2) for the HS-MLC, the performance of the nonbinary joint decoding algorithm (without iterative demapping) is slightly better than the iterative MSD algorithm (with iterative demapping); 3) the HS-MLC has a significantly better performance compared to the time-division HS-BICM-ID over fast fading channels.

In this paper, we analyze the physical layer security (PLS) performance of a free-space optical (FSO) communication system composed of a transmitting satellite and ground users. Specifically, the FSO fading channels follow the Málaga distribution. Further, we scrutinize the influence of non-zero boresight pointing errors and angle-of-arrival fluctuations on the PLS performance for the first time. We derived the probability density function and cumulative density function of the FSO link, followed by the closed-form expressions of the secrecy outage probability (SOP) and the probability of strictly positive secrecy capacity (SPSC). The asymptotic SOP expression at the high signal-to-noise ratio regime and diversity order are also provided to reveal the physical mechanism of the PLS of the considered system. Finally, Monte Carlo simulation results are presented to verify the correctness of the analytical expressions. The results afford helpful insights for the future design of satellite FSO communication systems.

Inference using convolutional neural networks (CNN) stands as a cornerstone technology in artificial intelligence (AI), providing a potent solution for intelligent applications at the edge of the Internet of Things (IoT). However, amidst the dynamic operational landscape, resource constraints, and intricate inference tasks within the IoT, ensuring both low latency and high accuracy in inference poses formidable challenges. In this study, we introduce an edge-end collaboration solution for accelerating inference (EECI), which dynamically distributes inference tasks between end devices and edge servers. Our objective is to mitigate inference delays through optimized task allocation and model segmentation points, while upholding inference accuracy in dynamic network environments. To strike a balance between computation and communication, we propose an inference delay prediction model and an inference offloading decision model. These models adaptively guide inference decisions based on bandwidth conditions, thereby maximizing the utilization of computing resources. Simulation results demonstrate the outstanding performance of our solution on the hardware testbed evaluation. Compared to other benchmark solutions, ours exhibits a remarkable improvement in inference acceleration ranging from 21.2% to 73.6%, alongside a increase in the utilization of edge server computing resources.