Dynamic spectrum access (DSA) is regarded as an effective and efficient technology to share radio spectrum among different networks. As a secondary user (SU), a DSA device will face two critical problems: avoiding causing harmful interference to primary users (PUs), and conducting effective interference coordination with other secondary users. These two problems become even more challenging for a distributed DSA network where there is no centralized controllers for SUs. In this paper, we investigate communication strategies of a distributive DSA network under the presence of spectrum sensing errors. To be specific, we apply the powerful machine learning tool, deep reinforcement learning (DRL), for SUs to learn “appropriate” spectrum access strategies in a distributed fashion assuming NO knowledge of the underlying system statistics. Furthermore, a special type of recurrent neural network (RNN), called the reservoir computing (RC), is utilized to realize DRL by taking advantage of the underlying temporal correlation of the DSA network. Using the introduced machine learning-based strategy, SUs could make spectrum access decisions distributedly relying only on their own current and past spectrum sensing outcomes. Through extensive experiments, our results suggest that the RC-based spectrum access strategy can help the SU to significantly reduce the chances of collision with PUs and other SUs. We also show that our scheme outperforms the myopic method which assumes the knowledge of system statistics, and converges faster than the Q-learning method when the number of channels is large.

Being a promising technology for fifth-generation (5G) communication systems, a novel railway communication system based on control/user (C/U) plane split heterogeneous networks can provide a high-quality broadband wireless service for passengers in high-speed railways with higher system capacity, better transmission reliability, and less cochannel interference. However, due to its special architecture where the C-plane and the U-plane must be split and supported by a macro Evolved Node B (eNB) and a phantom eNB, respectively, it would suffer more serious handover problem, particularly in intermacrocell handover, which directly degrades its applicability and availability in high-speed railways. Moreover, no technical specification has been released about this network architecture. Therefore, this paper focuses on redesigning and analyzing technical details and handover procedures based on Long-Term Evolution (LTE) specifications to guarantee the proposed system's practicability and generality and its analytical tractability. To resolve the handover problem, this paper proposes a handover trigger decision scheme based on GM(1, n) model of the grey system theory. By this scheme, the received signal quality from the (N + 1)th measurement report can be predicted from the Nth measurement period, and the predicted values can be then utilized to make the handover trigger decision. The simulation results show that our proposed scheme is capable of triggering handover in advance effectively and of enhancing handover success probability remarkably.

The shortage and congestion of lower spectra motivate the exploration of the broadband and underutilized mmWave to be used in future public mobile communications, and subsequently future HSR mobile communications, which has the potential to provide multi-gigabit rate radio access for train passengers. However, due to the inferior propagation characteristics of mmWave and particularity of HSR scenarios, there are many design challenges ahead. In this article, we tackle those challenges by developing technical solutions for mmWave broadband HSR systems. We first propose feasible multiple access techniques and frame structures based on OFDM and SC communications, respectively. We then present promising train-trackside network architectures based on different MIMO techniques, including BF and SM, which have been viewed as the key enabling technologies to realize mmWave communications in outdoor environments. Moreover, we discuss the inherent defects of each architecture and offer the corresponding solutions or recommendations. Finally, we conduct performance evaluation, and discuss the advantages and disadvantages of the proposed architectures. We hope this article will stimulate further research on the innovative use of mmWave in high-speed rail systems.

In industrial sensor networks, complex industrial environments may be encountered leading to a mix of signals of different types. To address this problem, a signal classification method using feature fusion is proposed for industrial Internet of things in this paper. In the proposed method, the received signals of nodes are processed by frequency reduction and sampling pretreatment, based on which intelligent representations of signals are obtained. Using federated learning, the data samples are trained with the feature fusion network. Moreover, the trained deep learning network is used on each sensor node to classify signals, the results of which will be transmitted to aggregation center. In the aggregation center, the improved evidence theory method is used to aggregate the recognition results of each sensor node to achieve the final classification. Simulation shows that the proposed method has excellent classification performances and effectively protects the privacy of industrial information.

Dynamic spectrum access (DSA) has the great potential to alleviate spectrum shortage and promote network capacity. However, two fundamental technical issues have to be addressed, namely, interference coordination between DSA users and interference suppression for primary users (PUs). These two issues are very challenging since generally there is no powerful infrastructures in DSA networks to support centralized control. As a result, DSA users have to perform spectrum management individually, including spectrum access and power allocation, without accurate channel state information and centralized control. In this article, a novel spectrum management framework is proposed, in which Q-learning, a type of reinforcement learning, is utilized to enable DSA users to carry out effective spectrum management individually and intelligently. For more efficient process, neural networks (NNs) are employed to implement Q-learning processes, so-called deep Q-network (DQN). Furthermore, we also investigate the optimal way to construct DQN considering both the performance of wireless communications and the difficulty of NN training. Finally, extensive simulation studies are conducted to demonstrate the effectiveness of the proposed spectrum management framework.

For distributed sensor systems using neural networks, each sub-network has a different electromagnetic environment, and these recognition accuracy is also different. In this paper, we propose a distributed sensor system using incremental learning to solve the problem of radio frequency fingerprint identification. First, the intelligent representation of the received signal is linearly fused into a four-channel image. Then, convolutional neural network is trained by using the existing data to obtain the preliminary model of the network, and decision fusion is used to solve the problem in the distributed system. Finally, using new data, instead of retraining the model, we employ incremental learning by fine-tuning the preliminary model. The proposed method can significantly reduce the training time and is adaptive to streaming data. Extensive experiments show that the proposed method is computationally efficient, and also has satisfactory recognition accuracy, especially at low signal-to-noise ratio (SNR) regime.