Super-resolution reconstruction techniques entail the utilization of software algorithms to transform one or more sets of low-resolution images captured from the same scene into high-resolution images. Nevertheless, the extraction of image features and nonlinear mapping methods in the reconstruction process remain challenging for existing algorithms. These issues result in the network architecture being unable to effectively utilize the diverse range of information at different levels. The loss of high-frequency details is significant, and the final reconstructed image features are overly smooth, with a lack of fine texture details. This negatively impacts the subjective visual quality of the image. The objective is to recover high-quality, high-resolution images from low-resolution images. In this work, an enhanced deep convolutional neural network model is employed, comprising multiple convolutional layers, each of which is configured with specific filters and activation functions to effectively capture the diverse features of the image. Furthermore, a residual learning strategy is employed to accelerate training and enhance the convergence of the network, while sub-pixel convolutional layers are utilized to refine the high-frequency details and textures of the image. The experimental analysis demonstrates the superior performance of the proposed model on multiple public datasets when compared with the traditional bicubic interpolation method and several other resolution methods.
Signal detection, identification, and characterization are among the major challenges in aerial communication systems. The ability to detect and recognize signals using cognitive technologies is still under active development when addressing uncertainties regarding signal parameters, such as blank spaces available within the transmitted signal and the utilized bandwidth. This paper proposes a learning-based identification framework for heterogeneous signals with orthogonal frequency division multiplexing (OFDM) modulation as generated in a simulated environment at an a priori unknown frequency. The implemented region-based signal identification method utilizes cyclostationary features for robust signal detection. Signal characterization is performed using a purposely-built, lightweight, region-based convolutional neural network (R-CNN). It is shown that the proposed framework is robust in the presence of additive white Gaussian noise (AWGN) and, despite its simplicity, shows better performance compared with conventional popular network architectures, such as GoogLeNet, AlexNet, and VGG 16. The signal characterization performance is validated under two degraded environments that are unknown to the system: Doppler shifted and small-scale fading. High performance is demonstrated under both degraded conditions over a wide range of signal to noise ratios (SNRs) and it is shown that the detection probability for the proposed approach is improved over those for conventional energy detectors. It is found that the signal characterization performance deteriorates under extreme conditions, such as lower SNRs and higher Doppler shifts.
To enhance system reliability and mitigate the vulnerabilities of the Global Navigation Satellite Systems (GNSS), it is common to fuse the Inertial Measurement Unit (IMU) and visual sensors with the GNSS receiver in the navigation system design, effectively enabling compensations with absolute positions and reducing data gaps. To address the shortcomings of a traditional Kalman Filter (KF), such as sensor errors, an imperfect non-linear system model, and KF estimation errors, a GRU-aided ESKF architecture is proposed to enhance the positioning performance. This study conducts Failure Mode and Effect Analysis (FMEA) to prioritize and identify the potential faults in the urban environment, facilitating the design of improved fault-tolerant system architecture. The identified primary fault events are data association errors and navigation environment errors during fault conditions of feature mismatch, especially in the presence of multiple failure modes. A hybrid federated navigation system architecture is employed using a Gated Recurrent Unit (GRU) to predict state increments for updating the state vector in the Error Estate Kalman Filter (ESKF) measurement step. The proposed algorithm’s performance is evaluated in a simulation environment in MATLAB under multiple visually degraded conditions. Comparative results provide evidence that the GRU-aided ESKF outperforms standard ESKF and state-of-the-art solutions like VINS-Mono, End-to-End VIO, and Self-Supervised VIO, exhibiting accuracy improvement in complex environments in terms of root mean square errors (RMSEs) and maximum errors.
Communication in High-Frequency (HF) band allows for good-quality, low-cost, and long-distance data-link transmission over diverse landscapes in aerial communication systems. However, as limited frequency resources are allocated, HF band suffers from poor spectrum efficiency when the channel is congested with many users. To maintain the robustness of the data-link transmission, Automatic Link Establishment (ALE) is the worldwide standard for sustaining HF communication of voice, data, instant messaging, internet messaging, and image communications. Technologies, such as spectrum sensing, Dynamic Spectrum Access (DSA) are utilised in ALE with the primary step of automatic mode recognition based on cognitive radio. Conventional methods, such as Automatic Modulation Recognition (AMR) targets at the classification of single modulation, while modern communication systems require recognising multiple modes in combination of various number of tones, tone spacing, and tone interval. In this study, an approach that features filling the gap using deep learning is proposed. By characterising the common in-use mode formats in HF range, investigation shows that spectrogram diagram varies significantly, which necessitates the accurate characterisation and classification of multiple communication modes. Specifically, Convolutional Neural Network (CNN or ConvNet) is adopted for classification. The dataset is collected through USRP N210 with GNU Radio simulation. By reconstructing the communication in selected modes, the mode formats are classified. The performance result of recognition accuracy is displayed with confusion matrix. The confident classification of spectral characteristics, as well as accurate estimation, are established for practical communication scenarios.
3 February 2021 in Journal of Intelligent & Robotic Systems
Rapid increases in unmanned aerial vehicles (UAVs) applications are attributed to severe spectrum collision issues, especially when UAVs operate in spectrum scarce environments, such as urban areas. Dynamic air-to-ground (A2G) link solutions can mitigate this issue by utilizing programmable communication hardware in the air and real-time assignment of spectrum resources to achieve high-throughput and low-latency connectivity between UAVs and operators. To mitigate the high-computation issue among ground control station (GCS) networks and provide a broad communication coverage for large number of UAVs, we propose an advanced UAV A2G communication solution integrated with the dynamic spectrum management (DSM) and network function virtualization (NFV) technology to serve urban operations. The edge-cutting UAV communication technologies are surveyed. The proposed scheme is discussed in terms of the high-level system architecture, virtual network architecture, specific virtual functions (SVFs), and affiliated operation support databases. Some major research challenges are highlighted and the possible directions of future research are identified.
To enable an efficient dynamic power and channel allocation (DPCA) for users in the downlink multi-channel non-orthogonal multiple access (MC-NOMA) systems, this paper regards the optimization as the combinatorial problem, and proposes three heuristic solutions, i.e., stochastic algorithm, two-stage greedy randomized adaptive search (GRASP), and two-stage stochastic sample greedy (SSD). Additionally, multiple complicated constraints are taken into consideration according to practical scenarios, for instance, the capacity for per sub-channel, power budget for per sub-channel, power budget for users, minimum data rate, and the priority control during the allocation. The effectiveness of the algorithms is compared by demonstration, and the algorithm performance is compared by simulations. Stochastic solution is useful for the overwhelmed sub-channel resources, i.e., spectrum dense environment with less data rate requirement. With small sub-channel number, i.e., spectrum scarce environment, both GRASP and SSD outperform the stochastic algorithm in terms of bigger data rate (achieve more than six times higher data rate) while having a shorter running time. SSD shows benefits with more channels compared with GRASP due to the low computational complexity (saves 66% running time compared with GRASP while maintaining similar data rate outcomes). With a small sub-channel number, GRASP shows a better performance in terms of the average data rate, variance, and time consumption than SSG.