Massive MIMO is one of the most promising technologies for the fifth generation (5G) mobile communication systems. In order to better assess the system performance, it is essential to build a corresponding channel model accurately. In this paper, a three-dimension (3D) two-cylinder regular-shaped geometry-based stochastic model (GBSM) for non-isotropic scattering massive MIMO channels is proposed. Based on geometric method, all the scatters are distributed on the surface of a cylinder as equivalent scatters. Non-stationary property is that one antenna has its own visible area of scatters by using a virtual sphere. The proposed channel model is evaluated by comparing with the 3GPP 3D channel model [1]. The statistical properties are investigated. Simulation results show that close agreements are achieved between the characteristics of the proposed channel model and those of the 3GPP channel model, which justify the correctness of the proposed model. The model has advantages such as good applicability

Since single satellite capacity is limited for hotspot areas, we propose a dual satellite (DS) cooperative transmission network to increase the number of served cells and improve the quality of service (QoS) in this article. We study the global beam hopping and power allocation in the proposed network, where the conflict between cell priority and latency is balanced. Due to the beam power impacting both cochannel interference and capacity, the two multiobjective problems are coupled. The plural method is introduced to simplify the coupled problems, which obtains a better QoS than the fixed-weight method. As the dimensionality of the beam hopping pattern increases with the number of cells, global beam hopping design is a dimensional explosion problem. The demand-weighted power (DWP)-adaptive particle marching (APM) algorithm includes roulette, stochastic exploration, and nonlinear weights strategies. It is proposed to handle high-dimensional problems and converge fast to better optimal solutions. The simulation results show that the DWP-APM algorithm performs better with a few particles than the four baseline algorithms. Compared with baseline algorithms, the DWP-APM algorithm reduces the computation time by 57% and improves the QoS level by 13.6% at least.

The number of base stations (BSs) in remote areas is poor, and seamless coverage cannot be achieved. This paper investigates a cognitive satellite-terrestrial network, including the low earth orbit-beam hopping (BH) satellite and terrestrial systems. Besides, the weighted capacity-request ratio is regarded as the quality of service (QoS) in the satellite system, and the weight indicates channel quality and service priority. Similarly, the capacity-request ratio is introduced as the QoS in the terrestrial system. By establishing a relationship between beams and terminals, the solution space of the BH pattern and beam power is shrunk, and the power allocation competition algorithm is proposed. The proposed algorithm promptly obtains a better solution than the particle swarm optimization and genetic algorithms. As the secondary system, the terrestrial system should obtain an accurate sense of the satellite system. We design a dynamic-ratio threshold spectrum detection mechanism, which causes less miss detection than energy detection under co-channel interference (CCI). Moreover, based on CCI suppression and quick selection of greedy thought, an adaptive resource adjustment algorithm is determined for the BS pattern and transmit power. Finally, the efficiency of the proposed algorithms and spectrum detection mechanism is demonstrated in simulations.

Automatic modulation recognition (AMR) stands as a crucial core technology within the realm of signal processing and perception, playing a significant part in harsh electromagnetic environments. The time–frequency image (TFI) of communication signals can manifest modulation characteristics and serve as a foundation for signal modulation recognition and classification. However, under the influence of the electromagnetic environment, communication signals are exposed to varying degrees of interference, which poses a challenge to the recognition of modulation types. Taking into account the effects of interference and channel fading, this paper introduces a communication signal modulation recognition algorithm based on deep learning (DL) and time–frequency analysis. This approach employs short-time Fourier transform (STFT) to generate time–frequency diagrams from time-domain signals. Subsequently, it binarizes the image and feeds it as input data to the neural network. Our research presents a composite deep convolutional neural network (CNN) architecture known as the composite dense-residual neural network (CDRNN). This architecture focuses on enhancing the feature extraction and identification, aiming to achieve accurate recognition of modulation types in harsh electromagnetic environments. Finally, simulation results validate that the proposed deep learning algorithm holds remarkable advantages in boosting the accuracy of modulation type recognition with better adaptability. The algorithm shows better performance even in harsh electromagnetic environments. When the signal-to-noise ratio (SNR) is 18 dB, the recognition accuracy can reach 92.1%.

The current physical layer security designs in integrated sensing and communication (ISAC) system commonly rely on the prior information about the eavesdropper, such as location and channel state information (CSI). However, for the covert eavesdropper, the absence of prior information significantly degrades security performance. To deal with this issue, an artificial noise (AN)-assisted ISAC secure beamforming scheme is proposed to counter the covert eavesdropper while considering the fairness of uses’ achievable secrecy rate. This scheme is more robust because it does not require any prior information of eavesdropper. We formulate an optimization framework to jointly design transmitted precoding matrix and AN vector. This complex non-convex multi-proportional fractional programming is solved by the proposed algorithm framework based on quadratic transformation-successive convex approximation (QT-SCA). Simulation results demonstrate the effectiveness and robustness of the proposed scheme for countering the covert eavesdropper.