χ2-BidLSTM: A Feature Driven Intrusion Detection System Based on χ2 Statistical Model and Bidirectional LSTM.

In a network architecture, an intrusion detection system (IDS) is one of the most commonly used approaches to secure the integrity and availability of critical assets in protected systems. Many existing network intrusion detection systems (NIDS) utilize stand-alone classifier models to classify network traffic as an attack or as normal. Due to the vast data volume, these stand-alone models struggle to reach higher intrusion detection rates with low false alarm rates( FAR). Additionally, irrelevant features in datasets can also increase the running time required to develop a model. However, data can be reduced effectively to an optimal feature set without information loss by employing a dimensionality reduction method, which a classification model then uses for accurate predictions of the various network intrusions. In this study, we propose a novel feature-driven intrusion detection system, namely χ2-BidLSTM, that integrates a χ2 statistical model and bidirectional long short-term memory (BidLSTM). The NSL-KDD dataset is used to train and evaluate the proposed approach. In the first phase, the χ2-BidLSTM system uses a χ2 model to rank all the features, then searches an optimal subset using a forward best search algorithm. In next phase, the optimal set is fed to the BidLSTM model for classification purposes. The experimental results indicate that our proposed χ2-BidLSTM approach achieves a detection accuracy of 95.62% and an F-score of 95.65%, with a low FAR of 2.11% on NSL-KDDTest+. Furthermore, our model obtains an accuracy of 89.55%, an F-score of 89.77%, and an FAR of 2.71% on NSL-KDDTest-21, indicating the superiority of the proposed approach over the standard LSTM method and other existing feature-selection-based NIDS methods.

A deep learning-based medication behavior monitoring system.

The internet of things (IoT) and deep learning are emerging technologies in diverse research fields, including the provision of IT services in medical domains. In the COVID-19 era, intelligent medication behavior monitoring systems for stable patient monitoring are further required, because many patients cannot easily visit hospitals. Several previous studies made use of wearable devices to detect medication behaviors of patients. However, the wearable devices cause inconvenience while equipping the devices. In addition, they suffer from inconsistency problems due to errors of measured values. We devise a medication behavior monitoring system that uses the IoT and deep learning to avoid sensing errors and improve user experiences by effectively detecting various activities of patients. Based on the real-time operation of our proposed IoT device, the proposed solution processes captured images of patents via OpenPose to check medication situations. The proposed system identifies medication status on time by using a human activity recognition scheme and provides various notifications to patients' mobile devices. To support reliable communication between our system and doctors, we employ MQTT protocol with periodic data transmissions. Thus, the measured information of patient's medication status is transmitted to the doctors so that they can periodically perform remote treatments. Experimental results show that all medication behaviors are accurately detected and notified to the doctor efficiently, improving the accuracy of monitoring the patient's medication behavior.

Precision Interface Engineering of an Atomic Layer in Bulk Bi2Te3 Alloys for High Thermoelectric Performance.

Grafting nanotechnology on thermoelectric materials leads to significant advances in their performance. Creation of structural defects including nano-inclusion and interfaces via nanostructuring achieves higher thermoelectric efficiencies. However, it is still challenging to optimize the nanostructure via conventional fabrication techniques. The thermal instability of nanostructures remains an issue in the reproducibility of fabrication processes and long-term stability during operation. This work presents a versatile strategy to create numerous interfaces in a thermoelectric material via an atomic-layer deposition (ALD) technique. An extremely thin ZnO layer was conformally formed via ALD over the Bi0.4Sb1.6Te3 powders, and numerous heterogeneous interfaces were generated from the formation of Bi0.4Sb1.6Te3-ZnO core-shell structures even after high-temperature sintering. The incorporation of ALD-grown ZnO into the Bi0.4Sb1.6Te3 matrix blocks phonon propagation and also provides tunability in electronic carrier density via impurity doping at the heterogeneous grain boundaries. The exquisite control in the ALD cycles provides a high thermoelectric performance of zT = 1.50 ± 0.15 (at 329-360 K). Specifically, ALD is an industry compatible technique that allows uniform and conformal coating over large quantities of powders. The study is promising in terms of the mass production of nanostructured thermoelectric materials with considerable improvements in performance via an industry compatible and reproducible route.