Many machine learning algorithms and almost all deep learning architectures are incapable of processing plain texts in their raw form. This means that their input to the algorithms must be numerical in order to solve classification or regression problems. Hence, it is necessary to encode these categorical variables into numerical values using encoding techniques. Categorical features are common and often of high cardinality. One-hot encoding in such circumstances leads to very high dimensional vector representations, raising memory and computability concerns for machine learning models. This paper proposes a deep-learned embedding technique for categorical features encoding on categorical datasets. Our technique is a distributed representation for categorical features where each category is mapped to a distinct vector, and the properties of the vector are learned while training a neural network. First, we create a data vocabulary that includes only categorical data, and then we use word tokenization to make each categorical data a single word. After that, feature learning is introduced to map all of the categorical data from the vocabulary to word vectors. Three different datasets provided by the University of California Irvine (UCI) are used for training. The experimental results show that the proposed deep-learned embedding technique for categorical data provides a higher F1 score of 89% than 71% of one-hot encoding, in the case of the Long short-term memory (LSTM) model. Moreover, the deep-learned embedding technique uses less memory and generates fewer features than one-hot encoding.

The non-terrestrial network (NTN) is a network that uses radio frequency (RF) resources mounted on satellites and includes satellite-based communications networks, high altitude platform systems (HAPS), and air-to-ground networks. The fifth generation (5G) and NTN may be crucial in utilizing communication infrastructure to provide 5G services in the future, anytime and anywhere. Based on the outcome of the Rel-16 study, the 3rd generation partnership project (3GPP) decided to start a work item on an NTN in 5G new radio (NR) Rel-17, and the focus of the study was on mobility management procedures, due to the movements of NTN platforms; especially, low earth orbit (LEO) satellites. Handover enhancements were discussed to tackle the frequent handover due to the fast satellite movement. Therefore, the major problem of handover in LEO satellite systems was the signaling storm created by handing over all user equipment (UE) in a cell to a new cell because when the UE crosses the boundary between the neighboring cell of a satellite, an intra-satellite or cell handover occurs; thus, all users in a cell are expected to experience a change of cell due to handover every few seconds. In addition, UE location is not easy to define due to moving cell/beam situations. In this study, we propose machine learning-based solutions for handover decisions in non-terrestrial networks for cell handovers or intra-satellite handovers to reduce signaling storms during handovers where the handover requests will be executed by clustered users. First, the dataset was generated by the simulator that simulates communication between users and satellites. Second, we preprocessed the data, and also used the feature creation technique to create the distance feature using the Haversine formula, and then applied clustering and classification algorithms. The experimental results show that the distance between a user and its cell center is an important parameter for handover decisions in NTN, and the random forest outperforms all models with a higher accuracy of 99% along with a better F1-score of 0.9961.

Image processing is one of the most quickly evolving technologies today and it is an approach for applying operations on an image to improve it or extract relevant information from it. It is a critical research field in engineering and computer science. However, analyzing a high number of variables demands a lot of memory and processing resources, and it can cause a classification algorithm to overfit on training samples and underfit on test samples. As a result, various strategies, such as extraction can be used to reduce the number of features in a dataset by producing new ones from old ones. In this paper, we firstly propose a deep learning-based feature extraction approach with a modular neural network where we employ a pre-trained network, neural architecture search net (NASNet), as a feature extractor on a custom dataset of raw minerals of copper and cobalt images. It allows the input image to be feed-forwarded while performing feature learning and feature map, then stopping at a pooling layer before the fully connected (FC) layer in the NASNet to extract and save the outputs of that layer in dumped files. Secondly, the extracted features are then used as training data to build a deep neural network and machine learning algorithms for the image classification of copper and cobalt raw minerals. The experimental results show that the NASNet extracts the features efficiently, and the proposed modular neural network performs well with the boosting-decision tree as a classifier, which gives higher accuracy of 91% than 90% of the deep neural network; moreover, the precision is 1 higher than 0.98 for the deep neural network.

Wireless-powered communication networks (WPCNs) consist of wireless devices (WDs) that transmit information to the hybrid access point (HAP). In this situation, there is interference among WDs that is considered to be noise and causes information loss because of adjacent signals. Moreover, power is limited and can be lost if transmission distance is long. This paper studies sum-throughput maximization with sectored cells for WPCN. We designed a downlink (DL) energy beamforming by sector based on the hybrid space division multiple access (SDMA) and nonorthogonal multiple access (NOMA) approach to maximize the sum throughput. First, a cell is divided into several sectors, and signals from each sector are transmitted to each antenna of the HAP, so that the signals are not adjacent. Further, the HAP decodes the overlapping information of each sector. Next, power allocation is optimized by sector. To optimize power allocation, a constrained optimization problem is formulated and then converted into a nonconstraint optimization problem using the interior penalty method. The optimal solution derives the maximal value to the problem. Power for each sector is optimally allocated according to this optimal solution. Under this consideration, sum-throughput maximization is performed by optimally allocating DL energy beamforming by sector. We analyzed sum throughput and fairness, and then compared them according to the number of sectors. Performance results show that the proposed optimal power allocation by sector using hybrid SDMA/NOMA outperforms the existing equal power allocation by sector in terms of the sum throughput while fairness is also maintained. Moreover, the performance difference between the hybrid approach and SDMA, which optimally allocates power by sector, was about 1.4 times that of sum throughput on average, and the hybrid approach was dominant. There was also no difference in fairness performance.

Recently, various new attempts are being made to improve the quality of media content according to the expansion of the media market. Pre-visualization is one of those attempts, and the behavior of characters (agents) in virtual space is essential for pre-visualization. In this paper, a study was conducted to automatically generate behaviors of virtual characters for more efficient visualization in pre-visualization. In particular, we propose a method to automatically produce an appropriate behavior by detecting the state of the surrounding environment with a deep reinforcement learning technique. A virtual environment is created using a game engine to configure space for reinforcement learning, and a reinforcement learning model of the training environment is configured with Python and PyTorch. The virtual environment and the model training environment are communicated with the ML-agents toolkit. In the virtual environment, the character basically moves in a straight line, and three obstacles appear at random locations in front of the character. The character senses 9 states and allows 5 actions. After that, a reward is offered according to the action to proceed with learning. For performance evaluation, reinforcement learning training was conducted using the Proximal Policy Optimization (PPO) algorithm and Soft Actor-Critic (SAC) algorithm, and performance comparisons were also conducted according to the batch size. As a result, we are able to secure a reinforcement learning model with obstacle avoidance capability. Applying the model to the character proved that the character can automatically animate according to the state of the surrounding environment without explicit programming.

Wireless Powered Communication Network (WPCN) consists of Hybrid Access Point (HAP) that performs power transmission and data collection at the same time, and multiple nodes that can transmit data. In WPCN, depending on the wireless communication environment, the nodes cannot be able to transmit data because they can fail to receive power. Hence, increasing the transmission rate under a given resource is one of the very important issues. In ordinary mobile communications, a cell is divided into several sectors and the data is collected through multiple antennas to increase the transmission rate using SDMA. As a result, if the number of nodes in the one sector increases, the interference between nodes increases, and the transmission rate may decrease. Accordingly, in order to maximize performance, the number of nodes that can exist in a sector must be limited. The transmission rate between nodes according to the distance difference may not be fair because the nodes far from the HAP charge a small amount of power by attenuation of the signal, and the nodes close to the HAP charge a relatively large amount of power. Therefore, we propose Hybrid SDMA and Non-Orthogonal Multiple Access (NOMA) as a way to maximize the performance in term of both Sum-Throughput and Fairness. Also, we prove that there is a tradeoff between Sum-Throughput and Fairness according to the number of sectors. The simulation results show that the Hybrid SDMA and NOMA improves the performance substantially compared to the conventional SDMA.