This paper explores the challenges of user equipment (UE) association and downlink power control in scalable cell-free massive multiple-input multiple-output (MIMO) systems. Initially, we develop a scalable UE association algorithm, which ensures that each UE is connected to only a small set of access points (APs). This approach effectively reduces both fronthaul requirements and the computational load on the APs. The algorithm employs a competition-based mechanism to ensure that APs efficiently distribute workloads while satisfying the quality of service (QoS) requirements of UEs, preventing them from losing network connectivity. Second, we introduce a distributed downlink power control method based on deep neural networks (DNNs) to improve the long-term downlink spectral efficiency (SE) of the entire network. This method relies solely on locally collected large-scale fading information as DNN input, making it adaptable to dynamic scenarios involving varying numbers of associated UEs. To further enhance the training efficiency of the DNN, we design a distributed training framework that fully leverages the computational resources of distributed edge processors (EPs). Simulation results show that the proposed UE association algorithm and downlink power control method exhibit significant advantages over the benchmarks.

Automatic modulation recognition plays a critical role in military applications, particularly in electronic warfare, spectrum surveillance, and secure communication systems. The precise identification of signal modulation modes is crucial for ensuring the efficacy, security, and efficiency of communications. Given the problems of limited feature extraction capability and performance degradation when dealing with low signal-to-noise ratio signals, this work proposes a novel model architecture that combines the Transformer with relative position encoding and separable convolution in the complex domain. The network can directly process the complex representation of signals to capture features in the time-frequency domain and enhance the ability to recognize complex signals. This method introduces relative position encoding in the complex domain into the Transformer framework, which uses complex attention mechanisms and adaptive position encoding to enhance the model’s long-range modeling capability. At the same time, it improves computational efficiency and local feature extraction capability by introducing separable convolution layers. Then, we construct an attention-driven feature fusion module, which can automatically adjust the weight ratio between features to achieve the optimal combination of features. The model shows excellent classification performance under various signal-to-noise ratio conditions in RadioML2016.10a and RadioML2016.10b datasets, especially in low signal-to-noise ratio environments, which is significantly improved compared to other methods. The research not only provides a new solution for AMR tasks but also expands new ideas for applying the Transformer model in signal processing.

The sun glint poses a significant challenge in optical remote sensing of the ocean using unmanned aerial vehicles (UAVs). It contaminates the oceanographic information within the images, not only obscuring underwater seafloor features but also increasing the radiance reflectance within the images, which is detrimental for the study and monitoring of the marine environment. Currently, there are two main approaches for sun glint correction in optical imagers (in this context, referring to visible light images captured in RGB channels): methods based on sea surface statistical model and methods based on deep learning. To assess the potential application of sun glint correction methods on UAV RGB imagery, this paper reviews and summarizes both types of methods. Through experimental evaluation, our examines the usability of these methods in UAV-based ocean remote sensing, providing a scientific foundation for obtaining high-quality marine monitoring images. The experimental results indicate that traditional sea surface statistical model-based methods developed for satellite imagery face challenges when transitioning to high-resolution UAV imagery. Conversely, deep learning-based methods show promise for sun glint correction in UAV RGB imagery. Although these methods are still in the early exploration stage, they hold the potential to offer innovative and efficient solutions for mitigating sun glint effects in high-resolution UAV RGB imagery.

Existing automatic 3D image segmentation methods usually fail to meet the clinic use. Many studies have explored an interactive strategy to improve the image segmentation performance by iteratively incorporating user hints. However, the dynamic process for successive interactions is largely ignored. We here propose to model the dynamic process of iterative interactive image segmentation as a Markov decision process (MDP) and solve it with reinforcement learning (RL). Unfortunately, it is intractable to use single-agent RL for voxel-wise prediction due to the large exploration space. To reduce the exploration space to a tractable size, we treat each voxel as an agent with a shared voxel-level behavior strategy so that it can be solved with multi-agent reinforcement learning. An additional advantage of this multi-agent model is to capture the dependency among voxels for segmentation task. Meanwhile, to enrich the information of previous segmentations, we reserve the prediction uncertainty in the state space of MDP and derive an adjustment action space leading to a more precise and finer segmentation. In addition, to improve the efficiency of exploration, we design a relative cross-entropy gain-based reward to update the policy in a constrained direction. Experimental results on various medical datasets have shown that our method significantly outperforms existing state-of-the-art methods, with the advantage of less interactions and a faster convergence.

Recent improvements in technology and computational power have increased interest in the application of data driven modeling (DDM) in petroleum industry. Recovery process evaluation using numerical reservoir simulators are always time consuming and computational intensive with many assumptions and uncertainty involved and inefficient for fast decision making. Thus, DDM have been adopted as an alternative tool to predict production performance under waterflooding which is one of the most important techniques for improving oil recovery. A synthetic waterflooding dataset including production profile, operational parameters, reservoir properties and well locations is constructed using the numerical reservoir simulator. Exploratory data analysis provides several insights into the nonintuitive factors in building the reservoir model. K-means clustering analysis is performed to identify internal groupings among producers. Artificial neural network (ANN) and support vector regression (SVR) are used to decipher the nonlinear relationships between input attributes and waterflooding production. The trained models are subsequently used to predict cumulative oil and watercut on the unseen samples. Clustering analysis reveal that distance to the free water level has a dominant effect and the clustering assignment is controlled by the interplay among input attributes characterizing reservoir properties and relative well locations. Good agreements between predicted outputs from models and simulation targets present the satisfactory generalization performance and predictive capabilities of ANN and SVR methods. ANN model with one output provides the most accurate prediction result on the test data. SVR models provide similar but slightly worse forecast than ANN models. Proposed methodologies in this work can be utilized as a surrogate or complementary model to analyze and predict recovery process in other reservoirs fast and efficiently.

RGB-D images captured by consumer camera can provide pair-wise color and depth information but depth image usually contains strong noise and large holes. Due to different modalities of RGB-D, the intensity-guided depth enhancement easily causes artifacts and blurred edges. To solve this problem, we propose the multi-scale mutual feature convolutional neural network (MSMF-CNN) to learn essential mutual features of RGB-D. This end-to-end framework has two stages, i.e., mutual feature learning and depth regeneration. Firstly, we design two parallel separated subnetworks as intensity and depth mutual feature generator. Specifically, we design a multi-scale mutual feature generator to enforce the depth mutual feature learning and incorporate various structure characteristic. Secondly, depth transformation sub-network combines two mutual features to recover clean depth image. We tested our model on two well-known datasets in terms of PSNR, visual effect and computing speed. Compared with state-of-art algorithms, our method provides better performance of all three criteria.

Deep Packet Inspection (DPI) is a widely used network protocol identification technology. Currently, Data flows on Internet are growing in every minute. Therefore, it's urgently demanded to improve the processing performance of the DPI engine. In this paper, we proposed an optimized solution based on regular expressions. We used an optimized algorithm to reduce the number of states of a deterministic finite automaton, in order to reduce memory space requirement of the DPI engine. Meanwhile, we improved the matching performance of the DPI engine by adopting a hybrid packet matching pattern.