A5047644703- Advanced Malware Detection Techniques
- Network Security and Intrusion Detection
- Context-Aware Activity Recognition Systems
- Human Pose and Action Recognition
- Software-Defined Networks and 5G
- Video Surveillance and Tracking Methods
- IoT and Edge/Fog Computing
- Data Stream Mining Techniques
- Air Quality Monitoring and Forecasting
- Internet Traffic Analysis and Secure E-voting
- Sustainable Supply Chain Management
- Anomaly Detection Techniques and Applications
- Modular Robots and Swarm Intelligence
- IoT Networks and Protocols
- Telecommunications and Broadcasting Technologies
- Advanced MIMO Systems Optimization
- Gait Recognition and Analysis
- Greenhouse Technology and Climate Control
- Software System Performance and Reliability
- Molecular Communication and Nanonetworks
- Smart Agriculture and AI
Toshiba (United Kingdom)
2021-2025
Toshiba (South Korea)
2023
Illinois Institute of Technology
2021
This study presents an empirical investigation into the energy consumption of discriminative and generative AI models within real-world MLOps pipelines. For models, we examine various architectures hyperparameters during training inference identify energy-efficient practices. AI, large language (LLMs) are assessed, with a focus primarily on across different model sizes varying service requests. Our employs software-based power measurements, ensuring ease replication diverse configurations,...
Data integrity becomes paramount as the number of Internet Things (ioT) sensor deployments increases. Sensor data can be altered by benign causes or malicious actions. Mechanisms that detect drifts and irregularities prevent disruptions bias in state an IoT application. This paper presents LE3D, ensemble framework drift estimators capable detecting abnormal behaviours. Working collaboratively with surrounding ioT devices, type (natural/abnormal) also identified reported to end-user. The...
Edge computing is rapidly changing the IoT-Cloud landscape. Various testbeds are now able to run multiple Docker-like containers developed and deployed by end-users on edge devices. However, this capability may allow an attacker deploy a malicious container host compromise it. This paper presents dataset based Linux Auditing System, which contains benign activity. We two scenarios, denial of service privilege escalation attack, where adversary uses device. Furthermore, we user in parallel...
A dataset of body tracking information is presented. The consists 315 captured walking sequences. Each sequence simultaneously by two Azure Kinect devices. captures are interleaved to effectively double the frame rate. Fifteen participants partook in this experiment. experiment seven actions, and having three predefined trajectories per That results 21 sequences participant. data were collected using Sensor SDK. They later processed official tools libraries provided Microsoft. For each...
Open RAN brings multi-vendor diversity and interoperability to mobile/cellular networks. It is becoming part of governmental strategies for diversifying telecoms supply chains. This paper describes the approach key achievements BEACON-5G project, jointly funded by UK government industry. The project aims at developing a competitive edge 5G contributing toward its maturity. addresses some challenges in this respect provides various innovations system integration, network slicing, marketplace...
With the ever-increasing reliance on digital networks for various aspects of modern life, ensuring their security has become a critical challenge. Intrusion Detection Systems play crucial role in network security, actively identifying and mitigating malicious behaviours. However, relentless advancement cyber-threats rendered traditional/classical approaches insufficient addressing sophistication complexity attacks. This paper proposes novel 3-stage intrusion detection system inspired by...
Generative artificial intelligence (GAI) has emerged as a pivotal technology for content generation, reasoning, and decision-making, making it promising solution on the 6G stage characterized by openness, connected intelligence, service democratization. This article explores strategies integrating monetizing GAI within future open networks, mainly from perspectives of mobile network operators (MNOs). We propose novel API-centric telecoms marketplace platform, designed to serve central hub...
With the increase of global demand for food and with adoption trends local sourcing locally grown produce, need gardening farming automation solutions ranging from industrial to home level has been on rise. In this paper, we propose an Internet Things (IoT) control system based concept Wireless Sensor Actuator Networks (WSAN) that provides ideal growing conditions user-defined crops. This is achieved by utilizing information provided a series sensors monitoring environmental (temperature,...
Data integrity becomes paramount as the number of Internet Things (IoT) sensor deployments increases. Sensor data can be altered by benign causes or malicious actions. Mechanisms that detect drifts and irregularities prevent disruptions bias in state an IoT application. This paper presents LE3D, ensemble framework drift estimators capable detecting abnormal behaviours. Working collaboratively with surrounding devices, type (natural/abnormal) also identified reported to end-user. The proposed...
Open RAN brings multi-vendor diversity and interoperability to mobile/cellular networks. It is becoming part of governmental strategies for diversifying telecoms supply chains. This paper describes the approach key achievements BEACON-5G project, jointly funded by UK government industry. The project aims at developing a competitive edge 5G contributing toward its maturity. addresses some challenges in this respect provides various innovations system integration, network slicing, marketplace...
With the ever-increasing reliance on digital networks for various aspects of modern life, ensuring their security has become a critical challenge. Intrusion Detection Systems play crucial role in network security, actively identifying and mitigating malicious behaviours. However, relentless advancement cyber-threats rendered traditional/classical approaches insufficient addressing sophistication complexity attacks. This paper proposes novel 3-stage intrusion detection system inspired by...