Computer Science
A Real-Time Intrusion Detection System for Enhancing Cybersecurity in Robotic Systems
The increasing integration of robotic systems across various sectors has highlighted the critical need for robust cybersecurity measures to safeguard these systems against cyber threats.
This research presents a novel Real-Time Intrusion Detection System (IDS) framework specifically designed to enhance the cybersecurity of robotic systems.
The proposed IDS framework monitors network traffic and continuously identifies potential threats in real time. A testbed is set up using an AlphaBot robotic device and a server machine to perform experiments under both normal and attack conditions. Network traffic data is captured in real-time using tools like Wireshark generating raw datasets from actual data exchanges between the robotic device and the server. The dataset undergoes preprocessing including feature extraction data cleaning and normalization. This processed dataset is then used to train machine learning algorithms such as Decision Trees K-Nearest Neighbors and Random Forest designed to identify patterns distinguishing between normal and malicious activities.
The IDS framework is tested on the AlphaBot robotic device and server machine demonstrating effective results in real-world conditions. The system achieved an accuracy rate of 96.61% in distinguishing between normal and attack traffic highlighting its robustness and practicality.
The proposed real-time IDS framework shows promise in enhancing the cybersecurity of robotic systems by effectively identifying potential threats in real time.
Patent Selections
Hybrid Model for Spectral Analysis of Multilayered Structures with Fractal Boundaries: A Combination of RCWA and FDTD
Multilayer structures are an important element of modern optical electronic and nanotechnological devices. Their spectral characteristics determine the efficiency of optical coatings photonic sensors and nanostructures. However traditional spectral analysis methods often do not take into account the influence of fractal irregularities local inhomogeneities and correlations between layers which limits the accuracy of predicting optical properties.
The aim of this research is to develop a hybrid numerical model for accurate analysis of spectral characteristics of multilayer structures by taking into account realistic irregularities and inhomogeneities. The software implementation of the modeling algorithm is carried out in the Python environment. As a result of numerical experiments the model configuration is optimized which ensures the precision and efficiency of spectroscopic studies.
The proposed model is based on a combination of Rigorous Coupled-Wave Analysis (RCWA) and Finite-Difference Time-Domain (FDTD) methods taking into account wave effects interference phenomena and local variations of the material. The novelty of the research lies in the development of a hybrid model of spectral analysis which combines RCWA and FDTD methods with adaptive discretization and description of fractal boundaries. The proposed methodology takes into account local inhomogeneities and correlations between layers which is critically important for high-precision spectral measurements. To increase the accuracy adaptive discretization is implemented which increases the resolution in areas with high gradients. Experimental verification is carried out on synthetic test structures reference data and real multilayer systems obtained by the laser-induced evaporation method.
The developed model demonstrates high accuracy in predicting the spectral characteristics of multilayer structures. The results of the study indicate that taking into account fractal irregularities and correlations between layers allows for achieving a more accurate match between the simulated and experimental spectra. The proposed hybrid numerical approach reduces computational costs by 30-50% while increasing the accuracy of spectral analysis by 15-20% compared to traditional methods. The study of a multilayer structure consisting of SiO2 TiO2 and polycrystalline silicon confirmed the significant influence of fractal irregularities in TiO2 on wave localization and light absorption. Analysis of the spectral characteristics of films created by the method of laser-induced evaporation of copper sulfate demonstrated the ability of the developed model to accurately reproduce key spectral features in particular the exponential decrease in transmission and oscillations in reflection. In addition a formalized model for simulating electromagnetic and thermal processes in lithium-ion batteries is proposed which opens up prospects for its application for analyzing internal processes in multilayer electrode structures and predicting their degradation.
For the first time an improved model for the spectral analysis of multilayer structures has been proposed and implemented incorporating adaptive algorithms and hybrid numerical methods to achieve higher accuracy compared to classical approaches. The obtained results confirm the effectiveness of the proposed methodology for calculating transmission spectra which aligns with experimental data and surpasses existing literature models in accuracy. Modeling of fractal irregularities confirmed that the Hurst parameter plays a key role in shaping the spectral characteristics of multilayer structures determining the level of smoothness or chaos of the boundaries between layers. Taking into account the correlation between layers showed that the interdependence of irregularities at the boundaries significantly affects the light transmission and creates additional diffraction peaks in the reflection spectrum. Optimization of numerical algorithms showed that the combination of RCWA and FDTD methods in a hybrid format provides a balance between accuracy and speed of calculations reducing the error to ±2% compared to experimental data. The use of adaptive discretization contributed to a reduction in computational costs by 30-40% while maintaining high accuracy of calculations which is especially important for complex multilayer systems. The results obtained demonstrate the versatility of the proposed model and its applicability for the development of high-precision spectral analyzers optical coatings and photonic sensors.
A Deep Learning Framework with Learning without Forgetting for Intelligent Surveillance in IoT-enabled Home Environments in Smart Cities
Internet of Things (IoT) technology in smart urban homes has revolutionised sophisticated monitoring. This progress uses interconnected devices and systems to improve security resource management and resident safety. Smart cities use technology to improve efficiency sustainability and quality. Internet of Things-enabled intelligent monitoring technologies are key to this goal.
Intelligent monitoring in IoT-enabled homes in smart cities improves security convenience and quality of life from advanced technologies. Using live monitoring and risk identification tools to quickly discover and resolve security breaches and suspicious activity to protect citizens. Intelligent devices allow homeowners to remotely control lighting security locks and surveillance cameras. Using advanced technologies to regulate heating cooling and lighting based on occupancy and usage.
This study introduces a deep learning architecture that uses LwF (Learning without Forgetting) to keep patterns while absorbing new data. The authors use IoT devices to collect and analyse data in real-time for monitoring and surveillance. They use sophisticated data pre-processing to handle IoT devices' massive data. The authors train the deep learning model with historical and real-time data and cross-validation to ensure resilience.
The proposed model has been validated on two different Robloflow datasets of 7382 images. The proposed model gains an accuracy level of 98.27%. The proposed Yolo-LwF model outperforms both the original Yolo and LwF models in terms of detection speed and adaptive learning.
By raising the bar for intelligent monitoring solutions in smart cities the suggested system is ideal for real-time adaptive surveillance in IoT-enabled households. By embracing adaptability and knowledge retention authors envision heightened security and safety levels in urban settings.
Machine Learning based Cancer Detection and Classification: A Critical Review of Approaches and Performance
Cancer is known as a deadly disease which includes several types of cancer. Cancer cannot be cured without proper treatment. Also it is crucial to detect cancer at an early stage. The objective of this study is to examine assess classify and explore recent advancements in the detection of different human body cancer types such as breast brain lung liver and skin cancer.
This study explores several tools and methods in machine learning either supervised or unsupervised and deep learning involved in treatment procedures. It also highlights current issues and provides directions for future research projects. In this review study different advanced machine learning deep learning and artificial intelligence algorithms are used for the detection and classification of different types of cancers including breast skin lung cancer and brain tumor.
This paper reviews advanced techniques standard dataset comparison and analysis of identification of skin breast lung cancer and brain tumors. It also evaluates these techniques from the perspectives of F-measure sensitivity specificity accuracy and precision.
This review article focuses on detecting cancer using machine learning techniques. Successive improvements and detection of cancer over the past decades are reviewed covering various types of cancer-like breast brain lung liver skin and others. This paper focuses on the usage of machine learning in the diagnosis treatment and improvement of cancer.
A Study on Learning Resources Recommendation based on Multi-Domain Fusion Network
Considering the singularity of collaborative filtering algorithms in recommending learning resources and the problem that existing knowledge graph convolutional networks cannot deeply mine the neighbourhood information of learning resource nodes in application scenarios with less neighbourhood information a multi-domain fusion convolutional network learning resource recommendation model based on knowledge graphs is proposed here.
This study aimed to improve the accuracy and personalization of recommendations of filtering algorithms.
First the model mapped learner nodes learning resources and their neighbour nodes into low-dimensional dense vectors. Second the multi-domain fusion layer and the multi-domain aggregator were used to obtain the fused multi-domain learning resource vector. Finally the learner vector and the multi-domain learning resource vector were fed into the prediction layer to calculate the interaction probability.
To verify the effectiveness of the algorithm we conducted a comparative experiment using the publicly available datasets MOOPer and MOOCCubeX. The experimental results showed that the proposed model outperformed baseline models such as CKE MKR KGCN DEKGCN and KGIN in terms of evaluation metrics such as AUC ACC and F1. At the same time when the neighborhood information was limited the AUC ACC and F1 values of the proposed model still maintained the optimal value.
Compared to the optimal baseline model the effectiveness of the proposed model has been proven.
An Adaptive Enhanced Residual Attention Method for Small Object Detection
In the field of object detection small object detection has been a challenging problem. Existing CenterNet mainly focuses on deep features while ignoring shallow features and there is also strong similarity object interference which leads to insufficient detection ability for small objects. To solve these problems this paper proposes a small object detection method AR-CenterNet based on the adaptive enhanced context model and residual attention mechanism.
Firstly to enhance the feature representation capability an Adaptive Enhanced Context Model (AEC) is designed which balances the contextual information of shallow features at different scales. context information of different scales of shallow features and fuses them with deeper features by different scales of convolutional expansion. In addition to reduce the influence of strong interference objects the RAM (Residual Attention Mechanism) module is proposed which reduces the interference of surrounding features by introducing the residual attention mechanism recognizes the channel and spatial attributes of the small object features by using the coordinate attention mechanism and preserves the original feature information through the jump connection.
Experimental results show that AR-CenterNet achieves excellent small object detection performance on PASCAL VOC RSOD and KITTI datasets.
The method has important application value in the field of intelligent transportation.