- Wireless Signal Modulation Classification
- Wireless Communication Networks Research
- Advanced Wireless Communication Techniques
- Ultra-Wideband Communications Technology
- Radar Systems and Signal Processing
- PAPR reduction in OFDM
- Power Line Communications and Noise
- Antenna Design and Analysis
- Millimeter-Wave Propagation and Modeling
- Machine Fault Diagnosis Techniques
- Internet Traffic Analysis and Secure E-voting
- Target Tracking and Data Fusion in Sensor Networks
- Blind Source Separation Techniques
- Cognitive Radio Networks and Spectrum Sensing
- Advanced MIMO Systems Optimization
- Fault Detection and Control Systems
- Network Security and Intrusion Detection
- Advanced Adaptive Filtering Techniques
- Anomaly Detection Techniques and Applications
- Cooperative Communication and Network Coding
- Advanced Measurement and Detection Methods
- Advanced SAR Imaging Techniques
- Diverse Aspects of Tourism Research
- Error Correcting Code Techniques
- Artificial Intelligence in Games
Harbin Engineering University
2014-2024
University of International Business and Economics
2022-2023
New York University
2023
Dalian University of Technology
2022
Advanced Micro Devices (United States)
2022
Hong Kong Polytechnic University
2020
Wuxi Vocational Institute of Commerce
2018
Naval Aeronautical and Astronautical University
2014-2017
Zhengzhou University
2007-2008
Zhongyuan University of Technology
2007
Nowadays, network intrusions have brought greater impact in a large scale. Intrusion Detection Systems (IDS) been recent research hotspot for both the industry and academic. However, due to dynamic characteristics of traffic, it is challenging extract significant features identify traffic types. This paper focuses on applying deep learning methods feature extraction. Specifically, an IDS model proposed based autoencoder long short-term memory (LSTM) cell. The overall architecture intrusion...
Nowadays, fifth generation (5G) network and unmanned aerial vehicle (UAV) are more important in the civil military field. Only communicating correctly 5G between UAVs, they can play a role real world. Modulation classification is premise to ensure communication UAVs correctly. However, effects of multipath fading always exists environment UAV channel, which leads severe modulation performance degradation. In order resolve this problem, we proposed novel algorithm that classify signals...
In this paper, information entropy and ensemble learning based signal recognition theory algorithms have been proposed. We extracted 16 kinds of features out 9 types modulated signals. The used are numerous, including Rényi energy on S Transform Generalized Transform. three feature selection algorithms, sequence forward (SFS), floating (SFFS) RELIEF-F to select the optimal subset from features. use five classifiers, k-nearest neighbor (KNN), support vector machine (SVM), Adaboost, Gradient...
Modulation identification shows great significance for any receiver that has little knowledge of the modulation scheme received signal. In this paper, we compare performance a deep autoencoder network and three shallow algorithms including SVM, Naive Bayes BP neural in field communication signal recognition. Firstly, cyclic spectrum is used to pre-process simulation signals, which are at various SNR (from -10dB 10dB). Then, established approximate internal properties from amount data. A...
To address multi‐sensor real‐time track‐to‐track association problem of aircraft platforms in a complex environment, where sensor biases are time‐varied, targets distributed closely and different sensors report targets, an anti‐bias algorithm based on distance detection is proposed according to the statistical characteristics Gaussian random vectors. First, vector between homologous tracks derived its feature analysed; second, rough minimum average refined χ 2 distribution illustrated...
Signal modulation recognition is widely utilized in the field of spectrum detection, channel estimation, and interference recognition. With development artificial intelligence, substantial advances signal utilizing deep learning approaches have been achieved. However, a huge amount data required for learning. increasing focus on privacy security, barriers between sources are sometimes difficult to break. This limits renders them weak, so that not sufficient. Federated can be viable way...
In the battlefield, we often don't know parameters about enemy wireless communication system. Therefore, need to use electronic reconnaissance equipment search, intercept, identify and analyze signal. However, exciting methods can only detect signal layer such as carrier frequency bandwidth, cannot obtain more information. order improve ability, propose a novel protocol classification algorithm based on long short-term memory (LSTM) deep belief network (DBN). We first introduce DBN, then...
The relay is harvesting energy for data transmission and eavesdropping attacks will occur. Considering carried security of relay, this letter studies selection in multi-hop D2D networks with an eavesdropper. To improve the secure connectivity performance higher bound source-to-destination secrecy probability (SCP) derived. Then, we derive factor directly related to largest SCP. a algorithm proposed find path. Our theoretical derivation are verified by Monte Carlo simulation results. results...
Graph neural networks (GNNs) have demonstrated a remarkable ability in the task of semi-supervised node classification. However, most existing GNNs suffer from nonrobustness issues, which poses great challenge for applying into sensitive scenarios. Some researchers concentrate on constructing an ensemble model to mitigate issues. Nevertheless, these methods ignore interaction among base models, leading similar graph representations. Moreover, due deterministic propagation applied GNNs, each...
Ultra Wide Bandwidth (UWB) system based on Parallel Combinatory Spread Spectrum (PCSS-UWB), combined low interception Probability of UWB with high data transmission efficiency System, can fulfill the needs modern communication in and security. Through simulation, it shows that PCSS-UWB has a better bit error rate (BER) Performance under additive white Gaussian noise (AWGN) background. Furthermore, BER performance by using different pseudo-noise sequences is analyzed result proves related to...
A flexible mathematical framework for adaptive wireless communication waveform design is of importance the implement cognitive radio-based software defined radio (CR-based SDR). As one popular models, "spectrally modulated spectrally encoded" (SMSE) was proposed to tackle this problem but it cannot be trivially applied (massive) multiple input output (MIMO) systems. In paper, we extend useful SMSE model into MIMO Inspired by tensor technique in signal processing and machine learning,...
Generalized Frequency Division Multiplexing (GFD-M) is a promising candidate waveform for the air interface of fifth generation (5G) communication networks. The combination GFDM and time-reversal space-time coding (TR-STC) can improve bit error rate (BER) performance. In this paper, we employ Walsh-Hadamard Transform (WHT) in TR-STC-GFDM systems to further enhance system performance over frequency-selective time-variant channels. Simulation results show that with slight increase complexity,...
The adaptive antenna-array based OFDM system always uses beamforming to choose the channel with largest Eigenvalue transmit block symbol. Such selection diversity is a simple way. However, it does not make full use of capacity promise. With goal maximizing throughput limited power for target BER, novel dynamic subcarrier and allocation scheme on SVD decomposition proposed in this paper. We investigate algorithm three backgrounds: impact SNR, antenna number user number. Numerical simulation...
A new ultra wideband (UWB) communication system based on parallel combinatory spread spectrum (PCSS) is proposed in this paper. The have the advantages both of PCSS and UWB system, such as high efficiency information transmission excellent secret property. Thus, can meet needs security same time modern system. simulation model provided performance analyzed. results indicate that has a better bit error rate (BER) performance.
This paper proposes a novel three-dimensional (3D) geometry-based stochastic model (GBSM) for multiple-input multiple-output (MIMO) vehicle-to-vehicle (V2V) wideband channels. The proposed GBSM combines line-of-sight (LoS) components, two-sphere model, and multiple confocal ellipsoidal models which represent the tapped delay line (TDL) structure. is sufficiently generic, flexible adaptable to various V2V scenarios. Based on we compare statistical properties of different taps in several...
With the rapid growth of wireless communication services, spectrum resources are increasingly scarce. Cognitive radio and dynamic access technologies valued for their ability to improve efficiency. This paper first gives classification technology, then analyzes research status shortcomings finally proposes a non-cooperative distributed model based on deep Q-Learning. Compared with existing model, this new not only adopts overlay access, but also combines underlay recognizes that users can...
The successive cancellation decoder is the first decoding algorithm for polar codes which can achieve binary memoryless symmetric channels' capacity. However, SC does not perform well. A method called List proposed and one of best algorithms in terms balance between bits error rate computation complexity. In this paper, we find that CRC-aided SCL improves effect codes. This scheme CRC play an inner outer role a concatenation Simulation shows Frame length N=1024 under with L=2 8 bit better...
In this paper, we have applied Spectrally Modulated Encoded (SMSE) framework to design multiple input output (MIMO) waveform. The existing SMSE can be used generate different multi-carrier waveforms by appropriately designing variables including data, coding, orthogonality, windowing, availability of frequency and components. It has been approved a generic for cognitive radio due its flexibility. However, spacial diversity not considered in the framework. Hence, extend with MIMO capability...
The 7nm node is the first generation where EUV lithography has been employed to replace a few multi-patterning immersion layers in high-volume manufacturing. insertion of can simplify production process, reduce cycle time, improve performance, and enhance yield. However, order fully take advantage these benefits, we need overcome new challenges introduced by technology. In this paper, present how integrate 193nm optimize critical steps process. Our product was initially taped out with full...
Encrypted traffic classification can effectively supervise and manage the data transmitted in network. Most deep learning-based encrypted models focus on modeling only one characteristic of network, but network has both spatial temporal characteristics, part research adopts recurrent neural training method to grasp characteristics traffic, there is a low efficiency problem when model sequential data. The not efficient. In this paper, we propose ASTNet, an based spatio-temporal features....