Automatic modulation recognition (AMR) is an essential and challenging topic in the development of the cognitive radio (CR), and it is a cornerstone of CR adaptive modulation and demodulation capabilities to sense and learn environments and make corresponding adjustments. AMR is essentially a classification problem, and deep learning achieves outstanding performances in various classification tasks. So, this paper proposes a deep learning-based method, combined with two convolutional neural networks (CNNs) trained on different datasets, to achieve higher-accuracy AMR. A CNN is trained on samples composed of in-phase and quadrature component signals, otherwise known as in-phase and quadrature (IQ) samples, to distinguish modulation modes that are relatively easy to identify. We adopt dropout instead of pooling operation to achieve higher recognition accuracy. A CNN based on constellation diagrams is also designed to recognize modulation modes that are difficult to distinguish in the former CNN, such as 16QAM and 64QAM, demonstrating the ability to classify QAM signals even in scenarios with a low signal-to-noise ratio.

Millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) has been regarded to be an emerging solution for the next generation of communications, in which hybrid analog and digital precoding is an important method for reducing the hardware complexity and energy consumption associated with mixed signal components. However, the fundamental limitations of the existing hybrid precoding schemes is that they have high computational complexity and fail to fully exploit the spatial information. To overcome these limitations, this paper proposes a deep-learning-enabled mmWave massive MIMO framework for effective hybrid precoding, in which each selection of the precoders for obtaining the optimized decoder is regarded as a mapping relation in the deep neural network (DNN). Specifically, the hybrid precoder is selected through training based on the DNN for optimizing precoding process of the mmWave massive MIMO. Additionally, we present extensive simulation results to validate the excellent performance of the proposed scheme. The results exhibit that the DNN-based approach is capable of minimizing the bit error ratio (BER) and enhancing spectrum efficiency of the mmWave massive MIMO, which achieves better performance in hybrid precoding compared with conventional schemes while substantially reducing the required computational complexity.

Accurate flight delay prediction is fundamental to establish the more efficient airline business. Recent studies have been focused on applying machine learning methods to predict the flight delay. Most of the previous prediction methods are conducted in a single route or airport. This paper explores a broader scope of factors which may potentially influence the flight delay, and compares several machine learning-based models in designed generalized flight delay prediction tasks. To build a dataset for the proposed scheme, automatic dependent surveillance-broadcast (ADS-B) messages are received, pre-processed, and integrated with other information such as weather condition, flight schedule, and airport information. The designed prediction tasks contain different classification tasks and a regression task. Experimental results show that long short-term memory (LSTM) is capable of handling the obtained aviation sequence data, but overfitting problem occurs in our limited dataset. Compared with the previous schemes, the proposed random forest-based model can obtain higher prediction accuracy (90.2\% for the binary classification) and can overcome the overfitting problem.Accurate flight delay prediction is fundamental to establish the more efficient airline business. Recent studies have been focused on applying machine learning methods to predict the flight delay. Most of the previous prediction methods are conducted in a single route or airport. This paper explores a broader scope of factors which may potentially influence the flight delay, and compares several machine learning-based models in designed generalized flight delay prediction tasks. To build a dataset for the proposed scheme, automatic dependent surveillance-broadcast (ADS-B) messages are received, pre-processed, and integrated with other information such as weather condition, flight schedule, and airport information. The designed prediction tasks contain different classification tasks and a regression task. Experimental results show that long short-term memory (LSTM) is capable of handling the obtained aviation sequence data, but overfitting problem occurs in our limited dataset. Compared with the previous schemes, the proposed random forest-based model can obtain higher prediction accuracy (90.2\% for the binary classification) and can overcome the overfitting problem.

Beamforming is considered one of the most important techniques for designing advanced multiple-input and multiple-output (MIMO) systems. Among these design criterions, sum rate maximization (SRM) under a total power constraint is a challenge issue due to its nonconvexity. Existing techniques for the SRM problem only obtain suboptimal solutions but require huge amount of computation due to their complex matrix operation and iterations. Unlike these conventional methods, we propose a deep learning based fast beamforming design method without complex matrix operation and iterations. Specifically, we first derive a heuristic solution structure of the downlink beamforming through the virtual equivalent uplink channel based on optimum MMSE receiver which separates the problem into power allocation and virtual uplink beamforming (VUB) design. Next, beamforming prediction network (BPNet) is designed to perform the joint optimization of power allocation and VUB. Moreover, the BPNet is trained offline using two-step training strategy. Simulation results demonstrate that our proposed method is fast while obtains the comparable performance to the state-of-the-art method.

Automatic modulation classification (AMC) is an promising technology for non-cooperative communication systems in both military and civilian scenarios. Recently, deep learning (DL) based AMC methods have been proposed with outstanding performances. However, both high computing cost and large model sizes are the biggest hinders for deployment of the conventional DL based methods, particularly in the application of internet-of-things (IoT) networks and unmanned aerial vehicle (UAV)-aided systems. In this correspondence, a novel DL based lightweight AMC (LightAMC) method is proposed with smaller model sizes and faster computational speed. We first introduce a scaling factor for each neuron in convolutional neural network (CNN) and enforce scaling factors sparsity via compressive sensing. It can give an assist to screen out redundant neurons and then these neurons are pruned. Experimental results show that the proposed LightAMC method can effectively reduce model sizes and accelerate computation with the slight performance loss.

Characterization and linearization of RF power amplifiers (PAs) are key issues of fifth-generation wireless communication systems, especially when high peak-to-average ratio waveforms are introduced. Recently, deep learning methods have achieved great success in numerous domains including wireless physical-layer. However, there has been limited work in using deep learning for PAs behavioral modeling and linearization. In this paper, we make a bridge between memory effects of the nonlinear PAs and memory of bidirectional long short-term memory (BiLSTM) neural networks. We then build a BiLSTM-based behavioral modeling architecture and its accompanying digital predistortion (DPD) model by reconciling a non-causality concern. Next, an additional model is proposed in this study to mitigate uncertainty of the tested PA when transforming phases. The experimental results demonstrate the effectiveness of the proposed scheme, in which the adequately trained networks are capable of characterizing the PA, and the artificial intelligence-based DPD shows promising linearization performance when considering the tested PAs inherent unpredictability.