In the Mobile Edge Computing (MEC)-oriented Space-Air-Ground Integrated Network (SAGIN), High-Altitude Platforms (HAPs) and Unmanned Aerial Vehicles (UAVs) exhibit superior line-of-sight communication probability and enhanced mobility. Compared to terrestrial eavesdroppers (Eves), these aerial platforms operate under more advantageous conditions for eavesdropping, thereby increasing the susceptibility of user-offloaded data to interception. To effectively address this issue, we formulate an optimization framework for secure data offloading in SAGIN, particularly addressing the threat posed by aerial Eves. Within this architecture, Internet of Things (IoT) devices offload data to UAV-assisted aerial networks and Low Earth Orbit (LEO) satellite networks. To enhance system security, HAP and UAV can be equipped with sensing and ranging modules to locate the approximate direction and position of Eve. Meanwhile, HAP sends Artificial Noise (AN) signal to confuse potential Eve. By applying optimization techniques such as Block Coordinate Descent (BCD) and Successive Convex Approximation (SCA), the original non-convex problem is decoupled into four sub-problems, and a Joint Optimization of Transmit Power from IoT Devices to UAV, AN Power, and UAV Trajectory (JOPAQ) algorithm is proposed, which holistically optimizes IoT device transmit power, HAP interference power, UAV flight trajectory, and sensing offload variables under energy consumption and flight constraints. The simulation results demonstrate the effectiveness of the JOPAQ algorithm, significantly enhancing the confidentiality and data security of the SAGIN-MEC framework.

The development of Large AI Models (LAMs) for wireless communications, particularly for complex tasks like spectrum sensing, is critically dependent on the availability of vast, diverse, and realistic datasets. Addressing this need, this paper introduces the ChangShuoRadioData (CSRD) framework, an open-source, modular simulation platform designed for generating large-scale synthetic radio frequency (RF) data. CSRD simulates the end-to-end transmission and reception process, incorporating an extensive range of modulation schemes (100 types, including analog, digital, OFDM, and OTFS), configurable channel models featuring both statistical fading and site-specific ray tracing using OpenStreetMap data, and detailed modeling of realistic RF front-end impairments for various antenna configurations (SISO/MISO/MIMO). Using this framework, we characterize CSRD2025, a substantial dataset benchmark comprising over 25,000,000 frames (approx. 200TB), which is approximately 10,000 times larger than the widely used RML2018 dataset. CSRD2025 offers unprecedented signal diversity and complexity, specifically engineered to bridge the Sim2Real gap. Furthermore, we provide processing pipelines to convert IQ data into spectrograms annotated in COCO format, facilitating object detection approaches for time-frequency signal analysis. The dataset specification includes standardized 8:1:1 training, validation, and test splits (via frame indices) to ensure reproducible research. The CSRD framework is released at https://github.com/Singingkettle/ChangShuoRadioData1The dataset is designed to be fully reproducible using the provided framework, configurations, and configurable fixed random seeds to accelerate the advancement of AI-driven spectrum sensing and management.

This paper concentrates on a fault detection-based performance recovery control scheme using an actuator replacing method. The main procedure includes: 1) an integral sliding-mode (ISM) based event-triggered nominal controller; 2) a shifting-function-assisted fault detection design; and 3) an event-triggered reconfigurable controller. Unlike existing FTC schemes, no prior fault model is required. Minor actuator faults are directly attenuated via ISM, whereas significant faults are handled by switching to a backup actuator. For both nominal and reconfigurable phases, a modified Hamilton-Jacobi-Bellman (HJB) equation is solved with a single critic neural network, and an experience replay mechanism mitigates the persistence of excitation requirement while reducing computational complexity. A shifting function guarantees that post-replacement reconstructed states re-enter the prescribed performance bound within a finite window, and an actuator-oriented event-triggered strategy lowers update frequency while ensuring single-actuator operation. The effectiveness of the proposed scheme is validated by two simulation examples.

Integrated sensing and communication (ISAC) has emerged as a pivotal technology for next-generation wireless networks, enabling simultaneous high-speed communication and precise environmental awareness. This paper presents a novel ISAC imaging method, which leverages sparse multi-view channel state information (CSI) from existing communication infrastructure to reconstruct scattering fields, thereby achieving high-fidelity 3D imaging and environment reconstruction without the need for dedicated sensing hardware. A Neural Scattering Field Network (NSFNet) is designed to accomplish this task. The framework consists of two key components: 1) EdgeFusionNet, which extracts robust geometric features from sparse multi-view CSI using a multi-scale 3D CNN with edge-guided attention, and 2) MLP-based decoder that explicitly regresses view-dependent scattering coefficients, thereby addressing both the limited-view sampling challenge and the physical view-dependency of scattering. A self-supervised training strategy combining reconstruction loss and total variation regularization ensures accurate and smooth reconstructions. Experimental results demonstrate that NSFNet significantly outperforms compressed sensing and ablation deep learning baselines in complex scenarios, achieving superior performance in terms of F1-score (>0.83) and Chamfer Distance (<0.15 m). Furthermore, the method maintains stable performance under practical signal-to-noise ratio conditions and varying user equipment deployment densities, offering a scalable and hardware-efficient solution for ISAC-enabled environmental sensing. The proposed approach bridges the gap between sparse communication channel measurements and high-resolution 3D imaging, paving the way for seamless integration of sensing and communication in next-generation wireless networks.

Orthogonal Time Frequency Space (OTFS) enables reliable communication in high-speed mobility scenarios, making it ideal for Low Earth Orbit (LEO) satellite communication. Furthermore sparse Code Multiple Access (SCMA) supports massive connection in uplink mobile communications. This paper proposes an OTFS-SCMA scheme for LEO satellite communications and the corresponding cross domain detection algorithm. At the transmitter, users are grouped, and a practical codebook is employed. At the receiver, cross-domain detection is utilized to obtain initial estimates, which are then refined using the Message Passing Algorithm (MPA) for optimized detection. Comparative analysis with other baseline schemes demonstrates the performance gain.

For integrated sensing and communication (ISAC) systems in complicated propagation scenarios, achieving precise multi-target sensing still faces many challenges. In particular, dynamic target perception is severely affected by clutter generated from static environmental factors, making accurate target distinction difficult. To address this issue, we propose an effective ISAC scheme designed for clutter suppression and efficient multi-target sensing. Unlike conventional methods that depend on prior information and complex computations, the proposed method eliminates long-term dependencies and reduces computational complexity. Specifically, we first construct a hybrid channel model that jointly considers static environments and dynamic targets. Upon receiving the ISAC signal, the base station (BS) estimates the static channel and applies an efficient spatial-domain filter to extract the effective dynamic channel. Subsequently, a reduced-complexity perception algorithm is employed to estimate key parameters, including distance, velocity, and angle. Simulation results demonstrate the feasibility and performance benefits of the proposed method in detecting numerous targets within complex cluttered environments.

With the rapid development of communication technologies, the need for accurate and efficient complaint auxiliary analysis (CAA) among mobile network optimization personnel is growing. However, most existing research solutions focus only on text data and structured data, with few incorporating user complaint speech data. To address this, the authors propose a CAA scheme for mobile network based on multi-modal generative Large Language Model (LLM). By integrating speech, text, and other multi-modal data, the proposed scheme aims to accurately understand and efficiently analyze user complaint information. The proposed approach consists of three key components. First, the Whisper model is employed to transform user complaint speech data into structured textual representations. Subsequently, a self-constructed domain-specific dictionary is integrated with an Attention-based mechanism to train a Key Information Extraction (KIE) model, thereby enhancing its semantic comprehension performance. Finally, keyword-based knowledge derived from knowledge graphs is combined with expert-defined rules to train a Generative Programs and Recommendations (GPR) model, enabling the system to deliver more accurate and professional Customer Assistance Automation (CAA) solutions. Experimental evaluations demonstrate that the proposed framework exhibits superior generalization capabilities.

Interference cancellation has always been a crucial task in wireless communications, especially in the presence of nonlinear distortions caused by power amplifier. However, when considering the ultra-wideband (UWB) orthogonal frequency division multiplexing (OFDM) systems, this task becomes more challenging as the channel estimation will be severely impacted by the nonlinearity, thus leading to significant performance degradation. Driven by solving this problem, this paper proposed a novel nonlinear signal processing method, referred to as log-sum-minimization sparse channel estimation based nonlinearity cancellation (LSMSCE-NC). In detail, an optimization model based on the log-sum norm minimization and nonlinearity cancellation is first established and then its iterative solution is also presented. The numerical results reveal that the proposed LSMSCE-NC method achieves significant bit error rate (BER) and normalized mean square error (NMSE) advantages compared to the state-of-the-art algorithm.

Automatic Modulation Classification (AMC) using deep learning techniques has become a prominent area of research, showcasing considerable practical applications. However, the present AMC deep learning network, trained on a specific channel model, performs poorly in a new channel scenario. To deal with this, an adversarial semi-supervised domain adaptation method is proposed. Specifically, classification accuracy, cluster sensitivity, and distribution distance are optimized together, utilizing a three-stage iterative training approach. As a result, the proposed model has achieved robust performance with a limited amount of labeled data when shifting to a new channel.

Jamming attack is a critical technique in communication countermeasures. This paper proposes a novel jamming method that employs Denoising Diffusion Proba-bilistic Models (DDPM) to generate distorted signals, which effectively increase the bit error rate (BER). The underlying mechanism functions similarly to a parroting technique, where the attacker replicates the transmission behavior during the communication process and transmits nonsensical information to cause interference. Compared to additive white Gaussian noise (AWGN) jamming, the proposed method demonstrates a more favorable energy efficiency ratio.

ABSTRACT Introduction With the development and integration of satellite and terrestrial networks, mobile traffic prediction has become more important than before, which is the basis for service provision and resource scheduling when supporting various vertical applications. However, existing traffic prediction methods, especially deep learning‐based methods, require massive data for model training. Due to data privacy concerns, mobile traffic data are not easily shared among different parties, making it difficult to obtain a precise prediction model. Methods To mitigate the data leakage risk, a federated learning framework is proposed in this study for mobile traffic prediction in satellite‐terrestrial integrated networks to achieve a tradeoff between data privacy and prediction accuracy. In the proposed framework, local models are trained in base stations on the ground, and a global model is aggregated in the satellite edge server in space. Results A deep learning‐based prediction model with an adaptive graph convolutional network (AGCN) and long short‐term memory (LSTM) modules is proposed and validated in numerical experiments, which achieves the lowest prediction error with a real‐world traffic dataset when compared with other graph neural network (GNN) variants in the federated learning setting. Conclusion Numerical experiments with a real‐world mobile traffic dataset demonstrate the effectiveness of the proposed approach, which outperforms other GNN variants with lower prediction errors.

The increasing consumer application demands of unmanned aerial vehicles (UAVs) in low-altitude airspace lead to many challenges for traditional UAV monitoring in terms of small UAV target sensing probability, target imaging resolution, equipment cost, etc. To solve these problems cost-effectively, the integrated sensing and communication (ISAC) technology based on cellular networks shows potential communication and sensing capabilities by sharing the same hardware equipment. This paper proposes a novel UAV target imaging algorithm using the channel state information (CSI) based ISAC signal to improve the target sensing probability and achieve target imaging. An ISAC signal model is designed to reconstruct the target image from CSI based on the time-frequency transformation relationship between target motion and CSI. Considering the sparsity of target scattering points, a two-dimensional alternating direction method of multipliers joint autofocus (2D-ADMM-AF) algorithm is proposed using compressed sensing theory, which achieves super-resolution reconstruction of the target image while eliminating the high side lobe problem caused by the sparse aperture problem that may exist in the ISAC scene. Further, the image entropy is introduced in the reconstruction process to achieve target translation compensation, eliminating the image blur problem caused by the translational motion of the UAV target. The simulation results show that compared with the traditional algorithm, the image entropy reconstructed by the 2D-ADMM-AF is reduced by 27% and the operating efficiency is improved by 85.5%. In addition, the real data of the DJI M300 UAV is collected using the ISAC testbed to verify the ability to image targets with RCS=0.3m2, 0.7m×0.7m.

The power amplifier (PA) is an indispensable component in wireless communication systems, while the nonlinearity induced by PA can lead to significant performance degradation. The conventional nonlinearity equalization (NLE) method can effectively mitigate the nonlinear effects and provide superior BER performance but requires intensive computational complexity. To this end, we propose a novel NLE method in the time domain, which can significantly reduce the computational complexity without sacrificing the BER performance. Specifically, we first propose a novel PA representation of the sum of products (SPs), which is a unified time-domain representation for several typical memory and memoryless PA models. On this basis, the SPs-iterative least square equalizer (SPs-ILSE) method is proposed to mitigate the impact of both the memory and memoryless PA’s nonlinear distortions at the receiver side. The computational complexity of complex multiplication (CCCM) in the proposed method isO(NlogN) for the memoryless PA models andO(KN2) for the memory PA models. Moreover, considering the commonly utilized PA models, we also derive the closed-form expression for the achievable SINR of the SPs-ILSE method in the ideal conditions. Numerical results show that (i) the closed-form SINR expression is valid for both the memory and memoryless scenarios (ii) the proposed method exhibits the superior bit error rate (BER) performance in comparison to several relevant nonlinear signal processing methods such as digital pre-distortion (DPD), and power amplifier nonlinearity cancellation (PANC) (iii) the proposed NLE method achieves the same BER performance as the previous NLE method, i.e., reconstruction of distorted signals (RODS), while the CCCM of the proposed method is much lower.

In this paper, an adaptive fuzzy bipartite formation tracking control strategy of nonlinear multi-agent systems (MASs) is developed under a directed communication topology. Fuzzy logic systems are utilized to approximate the unknown nonlinear dynamics. Then, considering MASs with both cooperative and competitive relationships, the proposed bipartite formation tracking strategy allows each agent to operate according to their specific advantages. By employing a nonlinear state transformation function, the original constrained outputs of the MASs are converted into unconstrained ones. Meanwhile, a prescribed-time shifting function is introduced to handle the deferred output constraints, by which the singularity problem generated from the denominator of nonlinear state transformation functions is avoided. In addition, considering the communication bandwidth is limited, a distributed dynamic self-triggered control (DSTC) mechanism is constructed to improve the transmission efficiency. Different from the traditional self-triggered control strategy, the proposed DSTC strategy allows triggered intervals to be adjusted dynamically according to bipartite formation tracking errors, which enables the DSTC strategy to compromise communication burdern and system performances dynamically. Finally, two simulation examples are given to illustrate the validity of the proposed control scheme.

Radio frequency fingerprint identification (RFFI) is a physical layer security methodology to recognize individual devices by leveraging hardware imperfections inevitably induced in the manufacturing process. However, the performance degradation caused by the time-varying channel impacts and interferences has severely restricted the development of RFFI. To this end, we present a channel-agnostic RFFI system, which consists of three modules, i.e., signal preprocessing module, feature extraction module, and classification module. In the signal preprocessing module, we first propose a novel approach, referred to as limiter-based spectral circular shift bidirectional division (LB-SCSBD), to generate two parallel spectral quotient (SQ) sequences. Then, we define the spectral quotient constellation (SQC) symbols according to different modulation formats, and thereby transform the SQ sequences into four magnitude-based sequences in terms of two channel-robust signal representations, i.e., the SQ magnitude (SQM) and SQC error vector magnitude (SQC-EVM). In the feature extraction module, we present a moment-based statistical feature extractor (MB-SFE) to extract the device-specific information from the above four sequences. In the classification module, the extracted statistics are fed into the multi-class support vector machine (SVM) for training and testing. We take WiFi as a case study and evaluate the performance of the proposed RFFI system by classifying eight simulated device models and six universal software radio peripheral (USRP) transmitter radios. Experimental results show that (i) the proposed method achieves the accuracies of 99.84% and 98.26% with eight devices in QPSK and 16QAM cases, as well as the accuracy of 92.42% with six USRP devices (ii) the proposed method exhibits superior classification performance in comparison to some existing RFFI methods, leading to a significant accuracy improvement of at least 38.33%.

Automatic modulation classification (AMC) is a crucial technique for the design of intelligent transceivers and has received considerable research attention. Conventional feature-based (FB) methods have the advantage of low computational complexity. However, these methods are highly sensitive to the distribution shifts of the received signal caused by the variation of channel effects and have rarely been studied in orthogonal frequency division multiplexing (OFDM) systems under unseen synthetic channels with multipath fading effects, carrier frequency offset (CFO), phase offset (PO) and additive noise. To solve this problem, this paper proposes a novel FB method using the error vector magnitude (EVM) features for AMC tasks (termed as EVM-AMC), which can achieve reliable classification performance for the communication scenarios considering unseen synthetic channels in OFDM systems. Specifically, we first propose the axisymmetric mapping-based self-circulant differential division (AM-SCDD) algorithm to convert the received signal into the non-negative spectral quotient (NNSQ) sequence, deeply suppressing the synthetic channel effects. Subsequently, we derive the EVM features by analyzing the matched error vectors between the generated NNSQ sequence and the predefined NNSQ constellation symbol (NNSQCS) masks. During this process, a percentile-based filter is utilized to remove the outliers in each matched error vector. Finally, the feature samples collected from various channel conditions are sent to the multi-class support vector machine (SVM) classifiers for training and testing. Two candidate modulation type sets are employed to evaluate the performance of the proposed EVM-AMC method under both the constant and changing channel conditions. Our numerical results demonstrate that 1) the proposed method exhibits impressive robustness and generalization when dealing with unseen synthetic channels, 2) the proposed method yields the best classification performance when compared to the conventional FB AMC methods in the presence of channel effects.

Radio frequency fingerprint identification (RFFI) is a promising physical layer security technique that employs the hardware-introduced features extracted from the received signals for device identification. In this paper, we consider an RFFI problem in the presence of hybrid time-varying distortions (HTVDs) induced by multipath fading channel, carrier frequency offset (CFO), and phase offset. To solve this problem, an HTVDs-robust RFFI framework is proposed. Firstly, we derive that the residual HTVDs after CFO correction can be approximated as multiplicative interference in the frequency domain. Secondly, we define a novel signal analysis dimension named spectral quotient (SQ) representation and then present the spectral circular shift division (SCSD) method to generate the HTVDs-robust SQ signals, where the multiplicative interference can be suppressed. Thereafter, the statistics including root mean square (RMS), variance (VAR), skewness (SKE), and kurtosis (KUR) are extracted from the real and imaginary components of the SQ signals, respectively. Finally, the statistical features are used for the training and testing of the support vector machine (SVM) classifiers. To further enhance the performance of the proposed RFFI scheme, we also present the spectral circular multi-shift division (SCMSD) method, which increases the flexibility in the generation of the HTVDs-robust SQ signals. Given what we knew, this is the first time attempting to mitigate the HTVDs by leveraging the strong frequency correlation at the neighboring subcarriers in the multivariate hypothesis tasks. Compared to several handcraft feature-based RFFI methods, the proposed method exhibits superior identification accuracy and strong robustness. Experimental results show that the proposed RFFI scheme can achieve the accuracy of 91.3%with five devices and 86.4% with sixteen devices when the classifiers are trained with the additive white Gaussian noise but are tested with the Rayleigh channel.

Automatic modulation classification (AMC) is to identify the modulation type of a received signal, which plays a vital role to ensure the physical-layer security for Internet of Things (IoT) networks. Inspired by the great success of deep learning in pattern recognition, the convolutional neural network (CNN) and recurrent neural network (RNN) are introduced into the AMC. In general, there are two popular data formats used by AMC, which are the in-phase/quadrature (I/Q) representation and amplitude/phase (A/P) representation, respectively. However, most of AMC algorithms aim at structure innovations, while the differences and characteristics of I/Q and A/P are ignored to analyze. In this article, lots of popular AMC algorithms are reproduced and evaluated on the same data set, where the I/Q and A/P are used, respectively, for comparison. Based on the experimental results, it is found that: 1) CNN-RNN-like algorithms using A/P as input data are superior to those using I/Q at high signal-to-noise ratio (SNR), while it has an opposite result in low SNR and 2) the features extracted from I/Q and A/P are complementary to each other. Motivated by the aforementioned findings, a multitask learning-based deep neural network (MLDNN) is proposed, which effectively fuses I/Q and A/P. In addition, the MLDNN also has a novel backbone, which is made up of three blocks to extract discriminative features, and they are CNN block, bidirectional gated recurrent unit (BiGRU) block, and a step attention fusion network (SAFN) block. Different from most of CNN-RNN-like algorithms (i.e., they only use the last step outputs of RNN), all step outputs of BiGRU can be effectively utilized by MLDNN with the help of SAFN. Extensive simulations are conducted to verify that the proposed MLDNN achieves superior performance in the public benchmark.

The Internet of Things (IoT) permeates every aspect of our daily lives as billions of interconnected devices are deployed in the physical world. However, IoT networks operate in an untrusted environment and often suffer from many malicious active attacks. Automatic modulation classification (AMC), which can identify the modulation format of intercepted signals without prior knowledge, is a vital technology in countering physical-layer threats of IoT. However, most of the existing algorithms assume the channel is time invariant, and the AMC in time-varying channels is not been well studied. To deal with this dilemma, a novel AMC algorithm MCBLDN consisting of multiple convolutional neural networks (CNNs), a bidirectional long short-term memory network (BLSTM), and a deep neural network (DNN) is proposed. In MCBLDN, a multislot constellation diagram (CD) method is proposed to extract time-evolution characteristics for generating more discriminative features. Specifically, different grayscale subimages generated by slotted CDs are processed serially by their respective CNNs. Therefore, MCBLDN is overparameterized and time consuming. In addition, the frequency offset and phase offset caused by time-varying channels are neglected in MCBLDN, which is detrimental to the performance of AMC. To address the mentioned disadvantages, a lightweight MCBLDN with a spatial transformer network (SLCBDN) is proposed. First, the multiple CNNs in MCBLDN are pruned into a lightweight classification model, and the input data are rearranged to facilitate parallel processing by the lightweight CNN. Additionally, the spatial transformer network (STN) is utilized to reduce the influence of frequency offset and phase offset. Numerical results verify that the proposed method achieves superior performance and higher speed compared to the baseline algorithm MCBLDN.

Automatic modulation classification (AMC) identifies a received signal’s modulation scheme without prior knowledge of the intercepted signal, which enables significant applications in both the military and civilian domains. Inspired by the great success of deep learning (DL), lots of neural networks are introduced into AMC. To further improve classification performance, various complementary cues including in-phase/quadrature (I/Q), amplitude/phase (A/P), constellation, and other formats are used together to enhance the discrimination of the DL model, where only outputs of the last layer are used. In this paper, we find that different layers’ outputs in the DL model are also complementary to each other. As a result, a hierarchical classification head based convolutional gated deep neural network (HCGDNN) is proposed by utilizing different layers’ output, which only uses the I/Q cue. The proposed HCGDNN consists of three groups of convolutional neural networks (CNN) blocks, two groups of bidirectional gated recurrent units (BiGRU), and a hierarchical classification head. Compared to the long short-term memory (LSTM), the BiGRU has a smaller computational complexity and also releases the gradient dispersion and explosion in the training phase. With the help of the hierarchical classification head, three groups of modulation predictions are made for a received I/Q signal. After that, a novel nonlinear optimization fusion method is derived to generate fusion weights to fuse different groups, then a final classification decision is made. Compared to AMC methods using various cues, the proposed HCGDNN only uses I/Q cue and has low computational overhead. Numerical results suggest that the newly developed HCGDNN achieves superior performance on the public benchmark.To help other researchers, the source code will be uploaded to the github as long as the paper is published.

Aiming to improve the security in device authentication, many methods based on the physical layer signal are proposed. The essential characteristics of the physical layer signal captured during the wireless communications are called Radio Frequency Fingerprint (RFF). Constellation diagram is a form of two-dimensional signal representation besides the waveform, and different digital modulation types will generate different constellation patterns. Compared with the existing researches on the feature extraction or the circuit component impairments, we investigate the effect of different digital modulation types on RF fingerprints identification. Five bitsimilar USRP X310 and one calibrated USRP 2943R act, respectively, as the transmitting and receiving devices. In this experiment, we use ten digital modulation types with signal carrier to transmit the randomly generated bit sequences, and use the cable connection to reduce the impact of the channel. Then we observe the variation of recognition accuracy under different modulation types. Our experiment results show that except for the BPSK and OQPSK, the recognition accuracy of the rest modulation types is over 98% when SNR is more than 20dB.

The Internet of Things (IoT) pervades every aspect of our daily lives and industrial productions since billions of interconnected devices are deployed everywhere of the globe. However, the seamless IoT unveils a number of physical-layer threats, such as jamming and spoofing that decrease the communication performance and the reliability of the IoT systems. As the process of identifying the modulation format of signals corrupted by noise and fading, automatic modulation classification (AMC) plays a vital role in physical-layer security as it can detect and identify the pilot jamming, deceptive jamming, and sybil attacks. In this article, we propose a novel cyclic correntropy vector (CCV)-based AMC method using long short-term memory densely connected network (LSMD). Specifically, cyclic correntropy model-driven feature CCV is first extracted using the received signals as it contains both the second-order and the higher order characteristics of cyclostationary. Then, the extracted CCV feature is put into the data-driven LSMD which mainly consists of long short-term memory (LSTM) network and dense network (DenseNet). Moreover, an additive cosine loss is utilized to train the LSMD for maximizing the interclass feature differences and minimizing the intraclass feature variations. Simulations demonstrate that the proposed CCV-LSMD method yields superior performance than other recent schemes.

Robust channel estimation in time-varying channels is used to guarantee the quality of communication services, especially for Vehicle-to-Everything (V2X) scenarios. To improve the channel estimation accuracy and reduce the pilot overhead, multi-input multi-output (MIMO) radar is deployed to assist millimeter wave (mmWave) channel estimation. In this paper, we propose a MIMO radar aided channel estimation scheme using deep learning (DL) for the uplink mmWave multiuser (MU)-MIMO communications. To allocate pilot resources reasonably, we design a transmission frame structure of joint radar module and communication module, which divides the estimation scheme into two stages, i.e., the arrival/departure (AoA/AoDs) estimation stage and the gain estimation stage. In view of the imperfections of array elements in practice, we propose an AoA/AoDs estimation algorithm based on subspace reconstruction in the AoA/AoDs estimation stage named two-step angle estimation (TSAE) algorithm. In the gain estimation stage, a DL based channel gain estimator is designed. An autoencoder combined with residual structure named residual denoising autoencoder (RDAE) is proposed to eliminate the noise on wireless signals, which is passed into the least square (LS) estimation module to obtain gains. Simulation results demonstrate that the MIMO radar aided and DL-based channel estimator provides the efficient estimation performance of the high-mobility mmWave channel with fewer training resources.

The development of the Internet-of-Things (IoT) security is comparatively slower than the pace of the IoT innovations. The seamless IoT network operates in an untrusted environment and is exposed to many malicious active attacks. As the process of identifying the modulation format of signals is corrupted by noise and fading, automatic modulation classification (AMC) can be viewed as an effective approach to counter physical-layer threats for IoT as it can detect and identify the pilot jamming, deceptive jamming, and Sybil attacks. Nowadays, data-driven deep learning (DL) techniques, which are capable of extracting discriminative features and perform better robustness to channel and noise conditions, have drawn widespread attention. The deep residual network (ResNet) has a strong representative ability, which can learn latent information repeatedly from the received signals and improve the classification accuracy. Meanwhile, the gated recurrent unit (GRU), which is capable of exploiting temporal information of the received signal can expand the dimension of the signal features for satisfactory classification performance. Considering the advantages of the above networks, this article proposes a novel gated recurrent residual neural network (GrrNet) for feature-based AMC, where the amplitude and phase of the received signal are utilized as the inputs of GrrNet. In GrrNet, a ResNet extractor module is first designed to extract the highly representative features and then temporal information is obtained by the subsequent GRU module which is capable of processing the representative features with the arbitrary length for modulation classification. Moreover, extensive simulations are conducted to verify the classification performance and robustness of the proposed GrrNet and it is shown that GrrNet outperforms other recent DL-based AMC methods. Moreover, the influence of the network parameters, symbol length, and frequency offset on performance is also explored.

Automatic modulation classification (AMC) plays a crucial role in the cognitive radio networks, to which feature-based (FB) methods are the dominating solutions. However, the original features in FB methods are redundant, leading to the ambiguity of classification. To tackle this problem, this paper proposes a novel multi-objective modulation classification (MOMC) method. To reduce the redundant features, the original multi-features are recombined into a single feature by multiobjective genetic programming (MOGP) algorithm. Two quantitative objectives, the classification error rate and the variance for robustness, are then presented to jointly optimize the algorithm as two fitness functions. Furthermore, the single feature generated by MOGP is classified by logistic regression (LR) with low computational complexity. Simulation results verify the enhanced robustness and classification accuracy performance yielded by our proposed MOMC method compared to the existing classification methods.

Millimeter wave band (mmWave) integrates with a wide variety of signals under manifold communication standards due to its high-capacity feature, which enables mmWave beam sensing to serve a valuable function in discriminating different signals. In this paper, we propose a novel frame structure consisting of variant beam sensing process and data transmission process. In the beam sensing process, multi-power-level beam sensing method is conducted in every direction to discriminate multi-users under multiple standards. The sensing duration varies with the number of directions. Several performance metrics are correspondingly proposed to quantify the beam sensing for multiple mmWave users, such as the probability of correct detection and the false alarm probability. In the second process, the signal with the biggest received signal-to-noise ratio (SNR) is given priority to communicate. On this base, sensing-throughput tradeoff is analyzed to balance the time division between two processes for throughput maximization. Finally, numerical evaluations and simulations are conducted to verify the correctness of the proposed methods.

Channel tracking has been a promising technology to sustain the directional link in the millimeter wave (mmWave) communication. However, the complex dynamic environment alters the movement of the mobile station (MS) and increases the inaccuracy of tracking. To tackle this issue, this paper proposes a novel variational mobility oriented channel tracking (VMCT) algorithm for three-dimensional (3D) mmWave system, where the MS moves with variational direction and velocity. The corresponding angles of arrival (AoA) variation is modeled as a multiple linear regression (MLR) process, which contributes to acquire the temporal correlation of sequential AoA states. To further train the weight coefficients of the MLR model with small-scale dataset, an integration of maximum likelihood function and Bayesian conjugate prior distribution is exploited. Simulation results verify the enhanced mean square error (MSE) performance yielded by our proposed tracking algorithm compared to the existing tracking methods.

The automatic recognition of modulation type for a detected signal is a significant task in wireless communication, which is the intermediate step between signal detection and demodulation. There are two general methods adopted in modulation recognition, i.e., likelihood-based (LB) method and feature-based (FB) method. Both LB and FB approach do not perform well when the signals are incomplete and received from a very limited number of observations. Therefore, we adopt a method based on dictionary learning to identify the modulation type of incomplete signals. The orthogonal matching pursuit (OMP) method is used to obtain the sparse representation and the sequential generalization of K-means (SGK) method is used to update the dictionary set. The experimental result shows that the recognition accuracy of our method is much higher, compared with the method based on higher-order cumulant.

Automatic modulation classification (AMC) technology, used to identify the modulation type of the received signal, plays an important role in the radio detection and electronic warfare applications. In this paper, we propose a novel featurebased AMC method in MIMO system. Firstly, the independent component analysis (ICA) is applied to separate the mixed signals at the receiver side. Then, the multiple features based on higher-order cumulants are extracted, and are used to identify the modulation type of signals. In classification process, the identification operation is modeled as an optimization problem, and we adopt a Quasi-Newton method to solve it. Simulation results show that the proposed method can classify various modulation types, and implying the effectiveness of the proposed scheme. Moreover, the analysis based on measured data shows that our proposed scheme is practical and efficient. In excellent signal-to-noise (SNR) range, it proves that the proposed method outperforms the other classical feature based methods in terms of probability of correct identification.

A cognitive device-to-device (D2D) network with D2D transmitters (DTs) that harvest radio-frequency (RF) energy from the primary transmitters (PTs) is investigated. A novel D2D transmitter-assisted cooperative (DTAC) protocol is proposed, in which a group of DTs that have no transmission opportunity act as potential relays to improve the communications of the primary network. The primary network outage probability is characterized and used to make comparisons between the direct link and the cooperative link which adopts different combining techniques at the primary receivers. The active probability of the DTs is derived, and the D2D network throughput is maximized by seeking an optimal transmission power for the PTs. Simulation results are provided to validate the theoretical analysis.

In this paper, an energy harvesting scheme for a multi-user multiple-input-multiple-output secrecy channel with artificial noise (AN) transmission is investigated. Joint optimization of the transmit beamforming matrix, the AN covariance matrix, and the power splitting ratio is conducted to minimize the transmit power under the target secrecy rate, the total transmit power, and the harvested energy constraints. The original problem is shown to be non-convex, which is tackled by a two-layer decomposition approach. The inner layer problem is solved through semi-definite relaxation, and the outer problem, on the other hand, is shown to be a single-variable optimization that can be solved by 1-D line search. To reduce computational complexity, a sequential parametric convex approximation method is proposed to find a near-optimal solution. This paper is then extended to the imperfect channel state information case with norm-bounded channel errors. Furthermore, tightness of the relaxation for the proposed schemes is validated by showing that the optimal solution of the relaxed problem is rank-one. Simulation results demonstrate that the proposed SPCA method achieves the same performance as the scheme based on 1-D but with much lower complexity.

To reduce the huge consumption of traditional sensing, a multi-centers estimation based sensing scheme is proposed in this paper. Firstly, all potential channels are clustered into highly related groups with some channels selected as detecting channels (DCs) using an unsupervised algorithm. In each group, the states of other channels (estimated channels, ECs) are estimated according to their correlations with the DCs and the dependence on history to save sensing time. Specifically, number of groups (Ng) and number of DCs in each group (NDC) can be adjusted jointly to improve sensing performance. Moreover, two Hidden Markov Model (HMM) based estimation methods, namely joint estimation (JE) and cooperative estimation (CE), are formulated. In JE, the DCs are modeled as the observed vectors and utilized jointly to estimate ECs' states. While in CE, each DC estimates ECs' states separately and a weight-based cooperative algorithm is designed to merge their results. Tested with real-world measurement data, results show the reduced sensing consumption is considerable at the expense of slight sensing accuracy loss. On these bases, it is significant to note that NdC should be adjusted according to sensing consumption to optimize performance.

As traditional spectrum sensing approaches unable to deal with the contradiction between detection accuracy and complexity in cognitive radio network, a novel q-weighed sequential cooperative energy detection method for spectrum sensing in time varying channel is proposed in this paper to achieve better performance with lower complexity. By adding the q- weighted log likelihood ratio (LLR) of the past local observations from previous sensing slots to the current LLR sequentially, cognitive radio nodes can aggregate the current and previous received energy values to yield the improvement of sensing performance. Moreover, we pose a q-weighted K-out of-N voting rule at the fusion center to minimize the total error probability. For different probability of primary signal for turning its state from active to idle, we employ corresponding different weighted value q to make the sensing scheme more flexible and efficient.

Automatic modulation classification (AMC) conducted by a single receiver plays a crucial role in spectrum monitoring and signal interception. To improve the accuracy of feature based AMC, a novel multi-cumulant based modulation classification scheme using uniform linear array is proposed in this paper. Moreover, two methods are formulated to combine the signal from different antenna branches, i.e., DOAC (Direction of arrival estimation based Combination) and CC (Cooperative Combination). With an estimate of the incident angle of the signal, DOAC combines signals from different branches using maximum ratio combining. CC calculates the feature value of each branch independently and utilizes the average feature value of all branches for classification. Simulation results prove that using multiple cumulants yields performance gain over traditional methods using a single cumulant. Moreover, the influence of antenna number and sample length on performance is also explored.

For years, the increasing popularity of flash memory has been changing storage systems. Flash-based solid-state drives (SSDs) are widely used as a new cache tier on top of hard disk drives (HDDs) to speed up data-intensive applications. However, the endurance problem of flash memory remains a concern and is getting worse with the adoption of MLC and TLC flash. In this article, we propose a novel cache management algorithm for flash-based disk cache named Lazy Adaptive Replacement Cache (LARC). LARC adopts the idea of selective caching to filter out seldom accessed blocks and prevent them from entering cache. This avoids cache pollution and preserves popular blocks in cache for a longer period of time, leading to a higher hit rate. Meanwhile, by avoiding unnecessary cache replacements, LARC reduces the volume of data written to the SSD and yields an SSD-friendly access pattern. In this way, LARC improves the performance and endurance of the SSD at the same time. LARC is self-tuning and incurs little overhead. It has been extensively evaluated by both trace-driven simulations and synthetic benchmarks on a prototype implementation. Our experiments show that LARC outperforms state-of-art algorithms for different kinds of workloads and extends SSD lifetime by up to 15.7 times.

NAND flash memory has attracted wide attention in both academia and industry in recent years. Its high random access performance fills the gap between DRAM and hard disks. While MLC is endorsed for higher density and lower cost per bit, it suffers from poor performance and endurance. Dual-mode flash combines SLC and MLC in a single device and thus provides the opportunity to trade density for performance. In this paper, we propose the Scalable Flash Storage(SFS) abstraction layer to facilitate cache management on dual-mode flash. SFS exposes a virtualized address space to hide the variable density of the medium. A differentiated write interface is introduced, which allows the cache manager to explicitly send write requests to SLC for high performance. SFS dynamically scales the proportions of SLC and MLC to balance between cache capacity and performance. SFS provides partially persistent storage service. It allows the cache manager to manage the data persistence on flash so that critical data can be retained persistently. Non-persistent data are discarded during garbage collection to mitigate write amplification. Based on the SFS, a Dual-mode Flash Cache(DMFC) architecture is designed to utilize the configurable density and performance. Experimental results show that DMFC can significantly improve overall performance for various workloads.

To understand the usage of TV spectrum, a comprehensive measurement is conducted in Beijing, China. The measurement consists of two parts, i.e., fixed measurement and radio environment mapping (REM). In the fixed measurement, spectrum utilization is calculated considering not only time domain utilization but also specific Chinese TV standards. Thus the spectrum utilization is 7% higher than reported previously. To study the geographical distribution of signal strength, REM is constructed for a small area in the downtown. Since traditional systematic sampling may place sample positions in inaccessible areas, a novel sampling algorithm named simulated annealing assisted electron repulsion (SAER) is proposed. Results show SAER results in smaller error in REM than systematic sampling and signal strength variation can reach 30 dB in the considered area, which means spatial spectrum access opportunities may exist for cognitive radio.

The increasing popularity of flash memory has changed storage systems. Flash-based solid state drive(SSD) is now widely deployed as cache for magnetic hard disk drives(HDD) to speed up data intensive applications. However, existing cache algorithms focus exclusively on performance improvements and ignore the write endurance of SSD. In this paper, we proposed a novel cache management algorithm for flash-based disk cache, named Lazy Adaptive Replacement Cache(LARC). LARC can filter out seldom accessed blocks and prevent them from entering cache. This avoids cache pollution and keeps popular blocks in cache for a longer period of time, leading to higher hit rate. Meanwhile, LARC reduces the amount of cache replacements thus incurs less write traffics to SSD, especially for read dominant workloads. In this way, LARC improves performance and extends SSD lifetime at the same time. LARC is self-tuning and low overhead. It has been extensively evaluated by both trace-driven simulations and a prototype implementation in flashcache. Our experiments show that LARC outperforms state-of-art algorithms and reduces write traffics to SSD by up to 94.5% for read dominant workloads, 11.2-40.8% for write dominant workloads.