To meet the rapid growth of mobile communications services, the continuously upgraded wireless networks significantly increase the demand for radio spectrum resources.However, the spectrum supply is limited, thus triggering the spectrum demand-supply gap.Consequently, dynamic spectrum sharing emerges as a promising paradigm of enhancing spectrum usage, which has been widely recognized and adopted in commercial wireless standards,such as the 4G LTE, 5G NR, and so on.As part of 3GPP Release 15, operators allocate portions of the 4G LTE spectrum that they are already using to 5G NR – this means that 4G LTE and 5G NR users can coexist in the same frequency band/channel at the same time.A software upgrade makes it possible to use existing RAN to deliver 5G services to unleash the potential of artificial intelligence in mobile networks.

Due to the open nature of the dynamic spectrum sharing architecture, various unknown malicious devices could participate in the spectrum sensing,access and management, leading legitimate users to be exposed much more internal and external attacks than ever before.It can be observed that the conventional security paradigms (e.g., cryptography and authentication) are insufficient for securing wireless communications, which motivates researchers and scientists to seek new means to complement the traditional security mechanisms and to improve the security of spectrum sharing systems.

The primary goal of this feature topic is to present the state-of-the-art original research and latest advances and innovations in key architectures, techniques, schemes, applications, and solutions of wireless security for dynamic spectrum sharing systems.All of the submitted papers were evaluated according to the standard reviewing process of China Communications.Following a rigorous peer-review process, nine articles were accepted in this special issue.The accepted papers cover various topics for enabling wireless security in dynamic spectrum sharing systems, including deep learning, relay, frequency-hopping, localization, identification, radio map,spectrum monitoring, and physical layer security.We hope this special issue will open up many exciting and critical future research activities in related fields.

In the face of new service demands of communication in the future, such as super-heterogeneous network, multiple communication scenarios, a large number of antenna elements and large bandwidth,new theories and technologies of intelligent communication have been widely studied, among which Deep Learning (DL) is a powerful technology in artificial intelligence (AI).It can be trained to continuously learn to update the optimal parameters.The first paper, “An Overview of Wireless Communication Technology Using Deep Learning” by Jiao et al., reviews the latest research progress of DL in intelligent communication and emphatically introduces four scenarios, including Cognitive Radio(CR), Edge Computing (EC), Channel Measurement(CM) and Visible Light Communication (VLC).The prospect and challenges of further research and development in the future are also discussed.

The article by Chen et al., “Relay-Assisted Secure Short-Packet Transmission in Cognitive IoT with Spectrum Sensing,” proposes a relay-assisted maximum ratio combining/zero-forcing beamforming(MRC/ZFB) scheme to guarantee the secrecy performance of dual-hop short-packet communications in cognitive Internet of Things (loT).The authors analyze the average secrecy throughput of the system and further investigate two asymptotic scenarios with the high signal-to-noise ratio (SNR) regime and the infinite block length.In addition, the Fibonacci-based alternating optimization method is adopted to jointly optimize the spectrum sensing block length and transmission block length to maximize the average secrecy throughput.The numerical results verify the impact of the system parameters on the tradeoff between the spectrum sensing block length and transmission block length under a secrecy constraint.It is shown that the proposed scheme achieves better secrecy performance than other benchmark schemes.The continuous change of communication frequency brings difficulties to the reconnaissance and prediction of non-cooperative communication networks.Since the frequency-hopping (FH) sequence is usually generated by a certain model with a certain regularity, the FH frequency is thus predictable.In the article entitled“Frequency-hopping Frequency Reconnaissance and Prediction for Non-cooperative Communication Network,” Li et al.investigate the FH frequency reconnaissance and prediction of a non-cooperative communication network by effective FH signal detection,time-frequency (TF) analysis, wavelet detection and frequency estimation.With the intercepted massive FH signal data, long short-term memory (LSTM)neural network model is constructed for FH frequency prediction.Simulation results show that their parameter estimation methods could estimate frequency accurately in the presence of certain noise.Moreover, the LSTM-based scheme can effectively predict FH frequency and frequency interval.

The article by Li et al., “Passive Localization of Multiple Sources Using Joint RSS and AOA Measurements in Spectrum Sharing System,” proposes a centralized localization scheme to estimate the positions of multiple sources.The authors develop two computationally light methods to handle the unknown RSS-AOA measurements-source association problem.One method utilizes spatial coordinates conversion to compute the minimum Euclidean distance summation of measurements.Another method exploits the long short-term memory (LSTM) network to classify the measurement sequence.Then, the authors propose a weighted least squares (WLS) approach to obtain the closed-form estimation of the positions by linearizing the non-convex localization problem.Numerical results demonstrate that the proposed scheme could gain sufficient localization accuracy under adversarial scenarios where the sources are in close proximity,and the measurement noise is strong.

The article by Tang et al., “Specific Emitter Identification for IoT Devices Based on Deep Residual Shrinkage Networks,” proposes a deep residual shrinkage network (DRSN) for specific emitter identification, particularly in the low SNR.The learnable soft threshold can preserve more key features to improve performance, and an identity shortcut can speed up the training process.The authors collect signals via the receiver to create a dataset in the actual environments.The DRSN is trained to extract features and implement the classification of emitters automatically.Experimental results show that DRSN obtains the best accuracy under different SNRs and has less running time, which demonstrates the effectiveness of DRSN in identifying specific emitters.

The article by Yang et al., “Primary User Adversarial Attacks on Deep Learning-based Spectrum Sensing and the Defense Method,” proposes a primary user adversarial attack (PUAA) to verify the robustness of the deep learning based spectrum sensing model.The authors design three PUAA methods in black box scenario.To defend against PUAA, they propose a defense method based on an autoencoder named DeepFilter.They apply the long short-term memory network and the convolutional neural network together to DeepFilter to extract the temporal and local features of the input signal at the same time to achieve effective defense.Results show that the three PUAA methods designed can greatly reduce the probability of detecting the deep learning-based spectrum sensing model.

The article by Zhang et al., “Intelligent Spectrum Management Based on Radio Map for Cloud-Based Satellite and Terrestrial Spectrum Shared Networks,”proposes an analytical framework that considers the inter-component interferences induced by spectrum sharing (SS).An intelligent SS scheme based on radio map (RM) consisting of LSTM-based beam prediction, transfer learning-based spectrum prediction, and joint non-preemptive priority and preemptive priority(J-NPAP)-based proportional fair spectrum allocation proposed.The simulation result shows improved spectrum utilization rate of cloud-based satellite and terrestrial spectrum shared networks (CB-STSSN).The proposed scheme decreases user blocking rate and waiting probability.

The article by Zhang et al., “Proactive Spectrum Monitoring to Suspicious Wireless Powered Communications in Dynamic Spectrum Sharing Networks,”studies the proactive spectrum monitoring with one half-duplex spectrum monitor to cope with the potential suspicious wireless powered communications(SWPC) in dynamic spectrum sharing networks.Simulation results show that the jamming-assisted spectrum monitoring schemes via spectrum monitoring data (SMD) transmission achieve much better performance than conventional passive spectrum monitoring, since the proposed schemes can obtain more accurate and effective spectrum characteristic parameters, which provide basic support for fine-grained spectrum management and a solution for spectrum security in dynamic spectrum sharing network.

Last but not least, the article by Xu et al., “Coalitional Game Based Joint Beamforming and Power Control for Physical Layer Security Enhancement in Cognitive IoT Networks,” proposes a coalitional game based joint beamforming and power control scheme to improve the achievable security of cognitive internet-of-thing devices.Specifically, the secondary network consisting of a multi-antenna secondary transmitter and multiple secondary users (SUs) is allowed to access the licensed spectrum resource of primary user (PU) with underlay approach in the presence of an unauthorized eavesdropper.Based on the Merge-Split-Rule, coalitional game is formulated among distributed secondary users with cooperative receive beamforming.Then, an alternative optimization method is used to obtain the optimized beamforming and power allocation schemes by applying the up-downlink duality.The simulation results demonstrate the effectiveness of the proposed scheme in improving the SU's secrecy rate and system utility while guaranteeing PU's interference threshold.

In conclusion, the Guest Editors of this feature topic would like to thank all the authors for their valuable contributions and the anonymous reviewers for their constructive comments and suggestions.We also would like to acknowledge the guidance from the editorial team of China Communications.