From sensor networks to power grids

Energy forecasting and smart homes

Deep learning for electricity prices, wind power and household demand, and edge computing and data fusion to make homes more flexible on the grid.

• Day-ahead zonal electricity price forecasting using 1D-LSTM with neighbouring zones data · IEEE PowerTech, 2023
• Temporal-spatial graph neural network for wind power forecasting considering the blockage effects · ICAPAI, 2023
• A novel spatial-temporal deep neural network for electricity price forecasting · ICAPAI, 2023
• Dynamic time-series clustering for improving appliances forecasting in smart homes · IEEE ISGT Europe, 2024
• Edge computing for improving energy management in smart homes · CIRED, 2023
• Cognitive data fusion for improving flexibility in smart homes · CIRED, 2023

LLMs for reliability engineering

Large language models with retrieval-augmented generation to automate failure modes and effects analysis (FMEA).

• A framework for automating FMEA using large language models and retrieval-augmented generation · International Journal of System Assurance Engineering and Management, 2026
• Application of a large language model (LLM) for FMEA · 2024

Health AI

Explainable deep learning for sleep-stage identification from EEG signals.

• SleepXAI: An explainable deep learning approach for multi-class sleep stage identification · Applied Intelligence, 2023
• Sleep stage identification based on single-channel EEG signals using 1-D convolutional autoencoders · IEEE HealthCom, 2022

Federated and edge learning

Training models across many small devices: which devices take part, and how to share scarce bandwidth so learning stays fast.

• Optimal device selection in federated learning for resource-constrained edge networks · IEEE Internet of Things Journal, 2023
• Low latency federated learning over wireless edge networks via efficient bandwidth allocation · IEEE WF-IoT, 2022
• A deep learning-based channel aware single step signal detection in downlink multi-user NOMA · IEEE VTC-Fall, 2022

Energy harvesting and UAVs

Charging IoT devices wirelessly: where to place energy transmitters and drones, and how to schedule them with reinforcement learning.

• Network-aware RF-energy harvesting for designing energy efficient IoT networks · Internet of Things, 2023
• Mobile energy transmitter scheduling in energy harvesting IoT networks using deep reinforcement learning · IEEE WF-IoT, 2022
• Energy-efficient UAV trajectory planning in rechargeable IoT networks · IEEE SPCOM, 2022
• UAV altitude optimization for efficient energy harvesting in IoT networks · NCC, 2022
• Context-aware RF-energy harvesting for IoT networks · IEEE WF-IoT, 2021

Static wireless sensor and IoT networks

In static IoT networks, sensor nodes are deployed randomly with limited power, which calls for energy-efficient data transmission. Network clustering, adaptive wake-up scheduling, mobile sinks and adaptive sampling all extend battery life. Cooperative data transmission models can overcome the limits of these conventional methods.

• Network lifetime improvement using landmark-assisted mobile sink scheduling · Elsevier Ad Hoc Networks, 2019
• A cluster based sensor-selection scheme for energy-efficient agriculture sensor networks · IEEE CCNC, 2021
• Energy-efficient wake-up radio protocol using optimal sensor-selection for IoT · IEEE WCNC, 2018
• Multi-sensor data fusion for cluster-based data aggregation in IoT applications · IEEE ANTS, 2019
• Low complexity landmark-node tracing in WSNs using multi-agent random walks · ACM ICDCN, 2019
• Energy-efficient landmark tracing in WSNs using random walks on network graphs · COMSNETS, 2017

Time-varying IoT networks

When nodes move, as in vehicular networks, the network changes from one moment to the next and needs a different approach to data transmission. Cooperative models with low latency and high reliability become essential.

• Optimal relay node selection for robust data forwarding over time-varying IoT networks · IEEE Transactions on Vehicular Technology, 2019
• Optimal relay node selection in time-varying IoT networks using apriori contact pattern information · Elsevier Ad Hoc Networks, 2020
• Joint data latency and packet loss optimization for relay-node selection in time-varying IoT networks · ACM MobiCom, 2018

Reinforcement learning for sensor networks

Q-learning to schedule mobile sinks, aggregate data and manage buffer overflow in cooperative wireless sensor networks, where fixed rules break down as the network changes.

• Cooperative network model for joint mobile sink scheduling and dynamic buffer management using Q-learning · IEEE Transactions on Network and Service Management, 2020
• Joint mobile sink scheduling and data aggregation in asynchronous wireless sensor networks using Q-learning · IEEE ICASSP, 2018

Beamforming for wireless networks

Clustering and 3D beamforming to raise throughput and manage power in mmWave and resource-constrained networks.

• Clustering-assisted 3D beamforming for throughput maximization in mmWave networks · IEEE ICC Workshops, 2021
• Cluster-based 3D-beam power management in resource-constrained wireless networks · COMSNETS, 2022