The explosive growth of unmanned aerial vehicles (UAVs)-based networks has accelerated in recent years. One of the crucial tasks of a UAV-based network is managing and allocating resources, including time, power, fly trajectory, and energy resources. We investigate a UAV-based network that gathers information from smart devices, sensor devices, and IoT devices (IDs) with respect to energy efficiency (EE) maximization. The EE of users served by the UAV over the time slots of a cycle is maximized through three categories: UAV trajectory optimization, power allocation, and time slot assignment. However, these are non-convex problems that are very difficult to solve. To solve the problem efficiently, we divide it step by step and convert the non-convex optimization problem into an equivalent convex optimization problem, optimizing each equivalent problem over each variable while other variables are fixed. Firstly, we perform a UAV trajectory optimization with a different number of ground users. Secondly, the Dinkelbach method is used to construct a non-convex fractional power allocation problem. In addition, we develop an algorithm to distribute time slots to all users, which continually raises the EE value. Eventually, a scheme is provided to sequentially update the method to each equivalent problem. The numerical results provide evidence that by solving the proposed sum-rate maximization problem, the performance of ground users has significantly improved with the support of the UAV-based network.

In a non-cooperative communication environment, automatic modulation classification (AMC) is an essential technology for analyzing signals and classifying different kinds of signal modulation before they are demodulated. Deep learning (DL)-based AMC has been proposed as an efficient method of achieving high classification performance. However, most current DL-AMC methods have limited generalization capabilities under varying noise conditions, especially at low signal-to-noise ratios (SNRs). Therefore, these methods can not be directly applied to practical systems. In this paper, we propose a threshold autoencoder denoiser convolutional neural network (TADCNN), which consists of a threshold autoencoder denoiser (TAD) and a convolutional neural network (CNN). TADs reduce noise power and clean input signals, which are then passed on to CNN for classification. The TAD network generally consists of three components: the batch normalization layer, the autoencoder, and the threshold denoise. The threshold denoise component uses an auto-learning threshold sub-network to compute thresholds automatically. According to experiments, AMC with TAD improved classification accuracy by 70% at low SNR compared with a model without a denoiser. Additionally, our model achieves an average accuracy of 66.64% on the RML2016.10A dataset, which is 6% to 18% higher than the current AMC model.