Recently, the frequent occurrence of the misuse and intrusion of UAVs has made it a research challenge to identify and detect them effectively, and relatively high bandwidth and pressure on data transmission and real-time processing exist when sampling UAV communication signals using the RF detection method. In this paper, firstly, for data sampling, we chose a compressed sensing technique to replace the traditional sampling theorem and used a multi-channel random demodulator to sample the signal; secondly, for the detection and identification of the presence, type, and flight pattern of UAVs, a multi-stage deep learning-based UAV identification and detection method was proposed by exploiting the difference in communication signals between UAVs and controllers under different circumstances. The data samples are first passed by detectors that detect the presence of UAVs, then classifiers are used to identify the type of UAVs, and finally flight patterns are judged by the corresponding classifiers, for which two neural network structures (DNN and CNN) are constructed by deep learning algorithms and evaluated and validated by a 10-fold cross-validation method, with the DNN network used for detectors and the CNN network for subsequent type and flying mode classification. The experimental results demonstrate, first, the effectiveness of using compressed sensing for sampling the communication signals of UAVs and controllers; and second, the detecting method with multi-stage DL detects higher efficiency and accuracy compared with existing detecting methods, detecting the presence, type, and flight model of UAVs with an accuracy of over 99%.

Vehicle-to-everything (V2X) communications, which provide wireless connectivity among vehicles, road sides drivers, passengers, and pedestrians, are attracting great interest. In this paper, we propose an energy-efficient relay assisted transmission scheme based on V2X communications in the uplink cellular networks for delay insensitive applications. We aim to maximize energy efficiency (EE) of the uplink while considering both the circuit power and transmit power as well as the delay constraint. The optimal resource allocation for direct transmission mode and V2X-enabled transmission mode are derived. For the V2X-enabled transmission mode, we first consider the transmission without the delay constraint and obtain the corresponding optimal power allocation. Based on the obtained result, we further propose the optimal resource allocation policy that satisfies the delay constraint and minimizes the energy consumption. Simulation results show that there exists an energy-delay tradeoff and the proposed V2X-enabled scheme can achieve significant energy saving with an acceptable delay compared with the traditional direct scheme, the decoding-and-forward (DF)-based transmission scheme, and the opportunistic store-carry and forward (SCF)-based transmission scheme, especially when the cell radius is not small.

In recent years, a lot of power allocation algorithms have been proposed to maximize spectral efficiency (SE) and energy efficiency (EE) for the distributed antenna systems (DAS). However, the traditional iterative power allocation algorithms are difficult to be implemented in reality because of their high computational complexity. With the development of machine learning algorithms, it has been proved that the machine learning method has excellent learning ability and low computational complexity, which can approximate the traditional iterative power allocation well and be easily to be implemented in reality. In this paper, we propose a new deep neural network (DNN) model for DAS. From the perspective of machine learning, traditional iterative algorithms can be regarded as a nonlinear mapping between user channel realizations and optimal power allocation schemes. Therefore, we train the DNN to learn the nonlinear mapping between the user channel realizations and the corresponding power allocation schemes based on the traditional iterative algorithm. Then, a power allocation schemes based on DNN method is developed to maximize SE and EE for DAS. The simulation results show that the proposed scheme can not only obtain the almost similar performance as the traditional iterative algorithm, but also reduce much online computational time.

In this paper, we consider the combination of machine learning (ML) and wireless communication. We design a machine learning generated clusters model in distributed antenna system (DAS), which is constructed by two different ML clustering algorithms, i.e., k-means algorithm and gaussian mixture model-based (GMM) algorithm. Under the communication scenario of DAS with ML generated clusters model, we investigate two different power allocation optimization problems with interference of maximizing spectral efficiency (SE) and energy efficiency (EE) in DAS, respectively. We compare the SE and EE of DAS with ML generated clusters model and the conventional model. Simulation results verify the effectiveness of DAS with ML generated clusters model, which can obtain the much better performance of SE and EE compared with the conventional communication model in DAS.

We investigate worst-case robust beamforming for simultaneous wireless information and power transfer (SWIPT) in a multiuser beamspace massive multiple-input multiple-output (mMIMO) system. The objective is to minimize the transmit power of the base station (BS) subject to the individual signal-to-interference-plus-noise ratio (SINR) and the energy-harvesting constraints under imperfect channel state information (CSI). Instead of directly resorting to semi-definite relaxation (SDR), we convert the initial non-convex optimization to a power allocation problem, which greatly reduces the computational complexity. The beamforming vectors are proved as scaled versions of the estimated channels, and then the optimal scaling factors are derived in closed-form. Simulations demonstrate that the proposed robust beamforming method achieves the globally optimal point for the initial design when the channel estimation errors are small, and leads to satisfactory performance when the channel estimation errors are large.

Irregular reconfigurable intelligent surfaces (IRIS) represents an advanced technology compared with traditional RIS, which can effectively enhance the energy efficiency (EE) performance of the system. This letter explores the problem of maximizing EE in a multiple-input single-output (MISO) network assisted by IRIS. To address the coupled optimization variables in this problem, we employ the block coordinate descent algorithm for efficient decoupling of the objective problem. Furthermore, quadratic transform theory, successive refinement method and an adaptive topology design method based on the tabu search algorithm are used to solve the decoupled subproblems. Simulation results confirm the substantial benefits of IRIS on the system’s EE performance.