In this paper, a novel fusion network for automatic modulation classification (AMC) is proposed in underwater acoustic communication, which consists of a Transformer and depth-wise convolution (DWC) network. Transformer breaks the limitation of sequential signal input and establishes the connection between different modulations in a parallel manner. Its attention mechanism can improve the modulation recognition ability by focusing on the key information. DWC is regularly inserted in the Transformer network to constitute a spatial–temporal structure, which can enhance the classification results at lower signal-to-noise ratios (SNRs). The proposed method can obtain more deep features of underwater acoustic signals. The experiment results achieve an average of 92.1% at −4 dB ≤ SNR ≤ 0 dB, which exceed other state-of-the-art neural networks.

This paper studies the resource allocation method of the heterogeneous cellular network. Aiming at the heterogeneous cellular network that supports user direct-connection communication, a wireless network communication technology based on SIDR is proposed to configure the user's transmit power for direct-connection communication. The method first defines the ratio of the power of the valuable signal received by the cellular communication node to the power of the interference from the direct-connected communication users. In this paper, SIDR can predict the relationship between the increase of outage probability of cellular communication nodes and the transmit power of direct-connected communication users after the introduction of direct-connected communication. A simulation example verifies that the method has strong practicability in practical applications, especially in multi-source fusion systems with limited communication.

Since the C2N-h2D crystal was efficiently synthesized, this study aims to investigate bandgap modulation of nanoribbons and nanotubes. Appling Density Functional Theory (DFT), the band-gap modulation of C2N-h2D nanomaterials is researched under elastic strains. The results of the current study indicate that the band gap of C2N-h2D nanoribbons and nanotubes can be tuned along two directions, namely, stretching or compressing nanoribbons and nanotubes when ɛ is changed from -10% to 10% in zigzag and armchair, respectively. This study also finds that the band gap of the C2N-h2D nanoribbons and nanotubes change with increase of widths or the radii of nanotubes. Therefore, the great potential applications of the C2N-h2D nanomaterials have been predicted in strain sensor and optical electronics at nanoscale.

Among many two-dimensional materials, hexagonal boron nitride (h-BN) nanomaterials have attracted great attention due to their unique properties and potential applications. In this work, BN-related nanosheets, nanoribbons, and nanotubes are studied by density functional theory. Using energy band engineering, external strain and stacking are introduced into the above materials to tune the band gap. The modulation of the band gap of the BN sheet is linear under strain, and is reduced from 5.456 eV to 3.915 eV. At the same width of zigzag nanoribbons, the change in band gap is close to 3 eV when strain is applied, which will have potential applications in electronics, optoelectronics, stress sensors, and so on.

Wind disturbance may significantly reduce the flight control performance of quadrotors when flying. In order to meet high performance flight control requirements, this paper presents a quadrotor anti-wind model (QAWM) and a flight control scheme for quadrotors. The QAWM takes the influence of aerodynamic effects on the propeller, gyroscopic effect and wind disturbance into account. The model can represent more flight states including not only hovering but also flight in windy conditions even maneuver flight. A cascade control scheme with an improved extended state observer (IESO) is adopted to decompose the quadrotor flight control problem into position control loop and attitude control loop. In the attitude control loop, the IESO is used to estimate the unmodeled dynamics, parameter uncertainties and external disturbances, as well as compensate the angular velocity control. In addition, the stability of IESO and the closed loop system are proved. Simulation and experimental results show the effectiveness of the nonlinear quadrotor model and control scheme, and the control scheme can effectively improve the flight performance of quadrotors under wind disturbance even maneuver flight.

Based on machine vision, this paper uses Matlab to perform image processing and image recognition on the image signals of mechanical parts collected, and transmits the control signals wirelessly through Zigbee to realize on-line detection and sorting of mechanical parts. Taking a complex workpiece with multiple circular holes as an example, the structure and implementation method of the system are introduced. The system has the advantages of low cost, high precision and easy installation. Its non-contact, real-time, flexibility and accuracy can effectively solve the problems of traditional detection methods.