A5005735651- Advanced Adaptive Filtering Techniques
- Acoustic Wave Phenomena Research
- Speech and Audio Processing
- Vehicle Noise and Vibration Control
- Noise Effects and Management
- Neural Networks and Applications
- Blockchain Technology Applications and Security
- Hearing Loss and Rehabilitation
- Blind Source Separation Techniques
- IoT and Edge/Fog Computing
- Aerodynamics and Acoustics in Jet Flows
- Energy Load and Power Forecasting
- Digital Platforms and Economics
- Aerospace, Electronics, Mathematical Modeling
- Automotive and Human Injury Biomechanics
National Technical University of Athens
2022-2024
Intracom Telecom (Greece)
2023
6G systems are perceived to be heavily softwarized, and therefore, the collaboration among all stakeholders, such as network cloud providers, operators, application developers, service providers device equipment vendors, shall occur via software in a secured, distributed, automated manner. This could envisioned with brokerless immutable marketplace anchored distributed ledger technologies, removing need for trusted party operator while avoiding this single point of failure, increasing...
Abstract Purpose The goal of this work is to propose a local Active Noise Control system for an aircraft seat’s headrest. This should be able extend the quiet zone beyond physical microphones that can placed behind passenger’s head due space constraints. Furthermore, control algorithm perform efficiently when non-linear phenomena exist during sound propagation. Methods A functional link neural network and computationally efficient Multiple Input Output approach are used system’s...
The excessive Sound Pressure Level at the interior of aircrafts can create very annoying flight conditions for passengers, as it reach 90 dB-95 dB some aircraft types (e.g.tilt rotors).These acoustic disturbances, which have low frequency components, be successfully attenuated by Active Noise Control systems installed in seat headrests.The present work proposes a system based on mixed error FxLMS algorithm because its complexity and robustness.It is also responsible mitigation pressure...
Reservoir Computing (RC) is an alternative approach to Recurrent Neural Networks (RNN) that can avoid some shortcomings of conventional gradient decent-training are mainly related high computational complexity and the reduced ability "learn" long-range dependencies. In RC input data transformed into patterns in a high-dimensional space by RNN called "reservoir", which remains unchanged during training. The desired output signal generated as linear combination neuron's signals from...
Active noise control (ANC) is an important subject that plays a central role in many practical problems.In essence, the ANC technique based on principle can be cancelled by another with same amplitude but opposite phase.Compared passive technique, method efficient to suppress lower frequency.As adaptive controller, filtered-x least mean square (FxLMS) algorithm has gained substantial popularity owing its simplicity.Unfortunately, physical world, systems exhibit certain degrees of...