Automatic modulation recognition (AMR) plays an important role in modern wireless communication. In this letter, a novel framework for AMR is proposed. The ResNeXt network serves as the backbone, and four proposed adaptive attention mechanism modules are incorporated. The time-frequency representations of the received signals are utilized as the inputs of the proposed deep learning (DL) network, and a transfer learning strategy is adopted based on the pre-trained ResNeXt weakly supervised learning (WSL) model. The comparisons with several state-of-the-art techniques on the RadioML2016.10B and RadioML2018.01A datasets show that the proposed framework converges quickly and can achieve higher robustness and 2% to 8% higher accuracy than other state-of-the-art techniques.

Automatic modulation recognition (AMR) of radio signals is becoming increasingly important due to its key role in wireless communication system management, monitoring, and control. In this paper, we propose an end-to-end AMR framework based on deep learning (DL), named CCNN-Atten. First, the complex-valued convolutional neural network (CCNN) extracts the features of the radio signal, and the feature calibration (FC) module selectively enhances the important features and suppresses irrelevant features. Then, a temporal context capture (TCC) module uses a modified multi-head attention mechanism (MHA) to capture the temporal dependence in the extracted features. The improved MHA mechanism, as a kind of self-attention mechanism, deploys causal convolutions to encode the temporal information of the input features and captures their local temporal relationship. In addition, due to the limited hardware resources in the real scenarios, we also considered a good compromise between recognition accuracy and computational complexity. Experiments were performed with the RadioML2016.10B, RadioML2016.10A, and RadioML2018.01A datasets, demonstrating the ability of the proposed CCNN-Atten to learn more robust features than other state-of-the-art (SOTA) techniques with 1 ~ 10% higher accuracy and a lower computational complexity than the SOTA models. The experimental results also show that CCNN-Atten achieves outstanding performance in dealing with radio signals with a lower sampling rate and small signal observation window.

Wireless body area networks (WBANs), as an important part of the Internet of things and sensor networks, are becoming more important in health monitoring and entertainment services. The sensor nodes in these applications need to transmit information data to personal station and receive control information from personal station to adjust working modes. However, unlike an easily rechargeable personal station, body sensor nodes suffer from energy shortages, which affect the nodes' lifetimes. To solve this problem, we design a novel medium access control (MAC) layer protocol to enhance the energy efficiency and extend the lifetime of body sensor nodes in WBANs. The proposed protocol utilizes a hybrid scheme by taking advantage of carrier-sense multiple accesses with the collision avoidance and time division multiple access schemes. In addition, we allocate the main transmission overhead at the personal station side and design a novel awaiting order state for sensor nodes to improve energy efficiency. Also, the analysis and simulation results demonstrate that the proposed protocol can provide a 6% to 15% decrease in energy consumption from other protocols.

Fog-cloud computing is one of cores techniques in wireless networks, where fog and cloud nodes provide high-speed and large-scale computing services to mobile users through a collaborative method. For the conventional fog-cloud computing schemes, the computing abilities of nodes and the computation task offloading methods are the main factors affecting the latency and energy consumption. However, when computing abilities of nodes reach saturation, the latency and energy consumption caused by task data transmission are approximate to those caused by computation. In order to reduce latency and energy consumption of data transmission and computation, this paper proposes an opportunistic access fog-cloud computing network (OFCN), where each mobile user selects fog node through opportunistic access method. Then, we formulate an optimization problem, considering both resource allocation and computation offloading under the constraints of users’ quality of service (QoS) requirements. Due to the difficulty of solving, we divide the original problem into four suboptimal problems and develop an iterative algorithm to approach the global optimal solution. Numerical results show that the proposed OFCN can achieve lower latency and energy consumption than the conventional fog-cloud computing networks.

It is challenging to covert communication under channel distribution information (CDI) uncertainty caused by limited feedback and rapid-changing channels. This letter investigates covert communication in uplink non-orthogonal multiple access (NOMA) systems under CDI uncertainty, where a covert user (CU) and reliable user (RU) transmit messages to a legitimate receiver (Bob) threatened by a warden (Willie). RU not only transmits its public messages but also helps CU to send private messages against Willie’s detection. To protect CU’s transmission with imperfect CDI, we first study the worst case for NOMA users and obtain a lower bound of the worst-case detection error probability (DEP). In addition, both closed-form expressions for RU and CU’s connection outage probabilities are derived. Constrained by the DEP and connection outage probability, we maximize the expected covert rate (ECR) and derive the optimal power settings for RU and CU under imperfect CDI. Numerical results show that the proposed robust power setting scheme achieves a positive maximum ECR despite CDI uncertainty causing the reduction of covertness performance.

Industrial Internet of Things (IIoT), regarded as an important technology for Industry 4.0, has the capability to connect massive IoT devices anywhere and at anytime in manufacturing industry. Enabling such a huge network requires message delivering among sensors, actuators, controllers, and the remote control to be seamless and reliable. However, IIoT wireless environment typically faces challenges such as blockage caused by IoT obstacles. To tackle the above issue, the unmanned aerial vehicle (UAV) and the reconfigurable intelligent surface (RIS) are exploited in this paper, which can provide favorable air-to-ground links and further rebuild the wireless channels. Moreover, the device-to-device (D2D) communication technique is introduced to enable direct information exchange between IoT devices. Specifically, we consider both the communication between the UAV and the cellular users (e.g., fixed IoT infrastructures) as well as the communication between D2D users (e.g., mobile IoT devices). Instead of only considering throughput, we focus on energy efficiency optimization for D2D users while guaranteeing the quality of service for cellular users, since energy-efficient transmission or green communication is important for IIoT scenarios. The transmit power, channel allocation parameters, and RIS’s reflection coefficients are jointly optimized to maximize energy efficiency for D2D users. To solve the formulated optimization problem, both centralized and distributed optimization algorithms based on deep neural networks are provided. Simulation results show that the introduction of RIS can significantly improve system performance. Moreover, the proposed algorithms can approximate the optimal solutions without the need of exhaustive search.