An Indoor Location-Based Augmented Reality Framework.

This paper proposes an indoor location-based augmented reality framework (ILARF) for the development of indoor augmented-reality (AR) systems. ILARF integrates an indoor localization unit (ILU), a secure context-aware message exchange unit (SCAMEU), and an AR visualization and interaction unit (ARVIU). The ILU runs on a mobile device such as a smartphone and utilizes visible markers (e.g., images and text), invisible markers (e.g., Wi-Fi, Bluetooth Low Energy, and NFC signals), and device sensors (e.g., accelerometers, gyroscopes, and magnetometers) to determine the device location and direction. The SCAMEU utilizes a message queuing telemetry transport (MQTT) server to exchange ambient sensor data (e.g., temperature, light, and humidity readings) and user data (e.g., user location and user speed) for context-awareness. The unit also employs a web server to manage user profiles and settings. The ARVIU uses AR creation tools to handle user interaction and display context-aware information in appropriate areas of the device's screen. One prototype AR app for use in gyms, Gym Augmented Reality (GAR), was developed based on ILARF. Users can register their profiles and configure settings when using GAR to visit a gym. Then, GAR can help users locate appropriate gym equipment based on their workout programs or favorite exercise specified in their profiles. GAR provides instructions on how to properly use the gym equipment and also makes it possible for gym users to socialize with each other, which may motivate them to go to the gym regularly. GAR is compared with other related AR systems. The comparison shows that GAR is superior to others by virtue of its use of ILARF; specifically, it provides more information, such as user location and direction, and has more desirable properties, such as secure communication and a 3D graphical user interface.

Deep Learning Anomaly Classification Using Multi-Attention Residual Blocks for Industrial Control Systems.

This paper proposes a novel method monitoring network packets to classify anomalies in industrial control systems (ICSs). The proposed method combines different mechanisms. It is flow-based as it obtains new features through aggregating packets of the same flow. It then builds a deep neural network (DNN) with multi-attention blocks for spotting core features, and with residual blocks for avoiding the gradient vanishing problem. The DNN is trained with the Ranger (RAdam + Lookahead) optimizer to prevent the training from being stuck in local minima, and with the focal loss to address the data imbalance problem. The Electra Modbus dataset is used to evaluate the performance impacts of different mechanisms on the proposed method. The proposed method is compared with related methods in terms of the precision, recall, and F1-score to show its superiority.

Fingerprint Feature Extraction for Indoor Localization.

This paper proposes a fingerprint-based indoor localization method, named FPFE (fingerprint feature extraction), to locate a target device (TD) whose location is unknown. Bluetooth low energy (BLE) beacon nodes (BNs) are deployed in the localization area to emit beacon packets periodically. The received signal strength indication (RSSI) values of beacon packets sent by various BNs are measured at different reference points (RPs) and saved as RPs' fingerprints in a database. For the purpose of localization, the TD also obtains its fingerprint by measuring the beacon packet RSSI values for various BNs. FPFE then applies either the autoencoder (AE) or principal component analysis (PCA) to extract fingerprint features. It then measures the similarity between the features of PRs and the TD with the Minkowski distance. Afterwards, k RPs associated with the k smallest Minkowski distances are selected to estimate the TD's location. Experiments are conducted to evaluate the localization error of FPFE. The experimental results show that FPFE achieves an average error of 0.68 m, which is better than those of other related BLE fingerprint-based indoor localization methods.

Time Series Multiple Channel Convolutional Neural Network with Attention-Based Long Short-Term Memory for Predicting Bearing Remaining Useful Life.

This paper proposes two deep learning methods for remaining useful life (RUL) prediction of bearings. The methods have the advantageous end-to-end property that they take raw data as input and generate the predicted RUL directly. They are TSMC-CNN, which stands for the time series multiple channel convolutional neural network, and TSMC-CNN-ALSTM, which stands for the TSMC-CNN integrated with the attention-based long short-term memory (ALSTM) network. The proposed methods divide a time series into multiple channels and take advantage of the convolutional neural network (CNN), the long short-term memory (LSTM) network, and the attention-based mechanism for boosting performance. The CNN performs well for extracting features from data with multiple channels; dividing a time series into multiple channels helps the CNN extract relationship among far-apart data points. The LSTM network is excellent for processing temporal data; the attention-based mechanism allows the LSTM network to focus on different features at different time steps for better prediction accuracy. PRONOSTIA bearing operation datasets are applied to the proposed methods for the purpose of performance evaluation and comparison. The comparison results show that the proposed methods outperform the others in terms of the mean absolute error (MAE) and the root mean squared error (RMSE) of RUL prediction.

Evolutionary Beamforming Optimization for Radio Frequency Charging in Wireless Rechargeable Sensor Networks.

This paper investigates how to efficiently charge sensor nodes in a wireless rechargeable sensor network (WRSN) with radio frequency (RF) chargers to make the network sustainable. An RF charger is assumed to be equipped with a uniform circular array (UCA) of 12 antennas with the radius λ, where λ is the RF wavelength. The UCA can steer most RF energy in a target direction to charge a specific WRSN node by the beamforming technology. Two evolutionary algorithms (EAs) using the evolution strategy (ES), namely the Evolutionary Beamforming Optimization (EBO) algorithm and the Evolutionary Beamforming Optimization Reseeding (EBO-R) algorithm, are proposed to nearly optimize the power ratio of the UCA beamforming peak side lobe (PSL) and the main lobe (ML) aimed at the given target direction. The proposed algorithms are simulated for performance evaluation and are compared with a related algorithm, called Particle Swarm Optimization Gravitational Search Algorithm-Explore (PSOGSA-Explore), to show their superiority.

Pricing resources in LTE networks through multiobjective optimization.

The LTE technology offers versatile mobile services that use different numbers of resources. This enables operators to provide subscribers or users with differential quality of service (QoS) to boost their satisfaction. On one hand, LTE operators need to price the resources high for maximizing their profits. On the other hand, pricing also needs to consider user satisfaction with allocated resources and prices to avoid "user churn," which means subscribers will unsubscribe services due to dissatisfaction with allocated resources or prices. In this paper, we study the pricing resources with profits and satisfaction optimization (PRPSO) problem in the LTE networks, considering the operator profit and subscribers' satisfaction at the same time. The problem is modelled as nonlinear multiobjective optimization with two optimal objectives: (1) maximizing operator profit and (2) maximizing user satisfaction. We propose to solve the problem based on the framework of the NSGA-II. Simulations are conducted for evaluating the proposed solution.