Smart Home Energy Management System using Intelligent Edge Computing

Home Smart Home Energy Management System using Intelligent Edge Computing

Smart Home Energy Management System using Intelligent Edge Computing, PI: Murad Khan, Funder: KFAS, Award Amount: 5500 KWD, Start: Feb 2024, End: March 2025

Funded by KFAS

Award Amount: 
5500 KWD
PI: Dr. Murad Khan

Start date: Feb 2024
End date: March 2025
Status: Ongoing

Theme: Internet of Things (IoT), Sensors, and Networking 


Impact (SDG): Affordable and Clean Energy (SDG 7), Industry, Innovation, and Infrastructure (SDG 9), Sustainable Cities and Communities (SDG 11), Responsible Consumption and Production (SDG12)

Figure 1 Graphical Abstract - Smart Home Energy Management System using Intelligent Edge Computing

Description:
Kuwait’s energy system is unsustainable due to the energy consumption rate of 15,590 kWh per capita. Most of the energy in the state of Kuwait is consumed in the residential sectors such as homes, buildings, offices, etc. To control the wide use and wastage of energy in these sectors, Internet of Things (IoT) applications could be used as a solution. However, developing an IoT-based solution will require the processing of energy data using deep learning models in a home environment. Further, implementing deep learning techniques in the home environment requires extensive processing and memory management, therefore, it is essential to transfer the energy data from home appliances to the edge network consisting of high-end processing devices in real time. The working of the proposed project is twofold, firstly, we will research the IoT-based energy management systems to identify the required data from the home appliances. In this phase, we will design data filtering and sampling techniques to convert the raw data suitable for various operations. The sampling and filtration techniques will be designed based on statistical techniques and graph neural networks. Secondly, we will develop a software model which will run over the edge using the deep learning models to extract the information from the filtered data suitable for automatic actions in smart homes. This information will be fed into the smart home to schedule the operational time of the appliances within the home environment. Resulting in controlling the unnecessary usage of the appliances. Finally, we aim to develop a sustainable solution that will support the scalability of the homes by adding and removing new appliances, automatic configuration, etc. Further, we aim to develop high-end results which will meet the objectives of the proposed project. These results will be published in well-known international journals and conferences.
  1. Khan, M., Hossni, Y. A comparative analysis of LSTM models aided with attention and squeeze and excitation blocks for activity recognition. Sci Rep 15, 3858 (2025). https://doi.org/10.1038/s41598-025-88378-6
  2. Khan, M. (2024). Safeguarding IoT networks against DDoS attacks using deep learning based zero trust network access. Electronics Letters, 60(21), e70075
    https://doi.org/10.1049/ell2.70075