Smart Buildings Can Contribute to Sustainable Cities and Society

Strategic Insights -

Smart buildings play a critical role in developing sustainable cities and societies, contributing to human safety, health, comfort, and building energy efficiency based on indoor human behaviors. This dynamic interaction between the built environment and its users enables the creation of smarter, more efficient, and more comfortable spaces. However, the adoption of technology for human behavior recognition in smart buildings has been limited due to high costs, privacy invasion, low resolution, and the need for intrusive wearables. Researchers from the Beijing University of Technology have explored Wi-Fi-based technology as a potential solution to overcome these limitations, demonstrating its potential in recognizing human behaviors in smart buildings.

Key Takeaways:

  • The researchers developed a Wi-Fi-based technology for non-intrusive human behavior recognition in smart buildings, which can overcome the limitations of high costs, privacy invasion, low resolution, and the need for intrusive wearables.
  • The technology was tested using a public dataset with an accuracy of 87.5%, indicating its potential for real-world applications.
  • The Bidirectional Long Short-Term Memory (Bi-LSTM) network achieved the highest accuracy of 98.1% among the seven classification algorithms evaluated, demonstrating its effectiveness in recognizing human behaviors.
  • The study highlights the importance of developing affordable and non-intrusive technology for human behavior recognition in smart buildings to support sustainable development and improve human safety, health, and comfort.
  • The researchers recruited 10 participants and captured a total of 2500 channel state information (CSI) data for recognition on five body movements, demonstrating the potential of Wi-Fi-based technology in capturing human behaviors.
  • The study was peer-reviewed and received financial support from the National Natural Science Foundation of China (NSFC).
  • The research has significant implications for the development of smart buildings and sustainable cities, which can contribute to the creation of healthier, more comfortable, and more energy-efficient spaces.

Statistics:

  • 98.1% accuracy was achieved by the Bidirectional Long Short-Term Memory (Bi-LSTM) network among the seven classification algorithms evaluated.
  • 87.5% accuracy was achieved using a public dataset to validate the model.
  • 10 participants were recruited to test the technology.
  • 2500 channel state information (CSI) data were captured for recognition on five body movements.
  • 2025 is the year the research was conducted.

Sources:

  • A Bi-lstm-based Wi-fi Csi Approach for Non-intrusive Human Behavior Recognition In Smart Buildings. Energy and Buildings, 2025;345.
  • National Natural Science Foundation of China (NSFC).
  • Beijing University of Technology.
  • NewsRx. New Energy and Buildings Data Have Been Reported by Researchers at Beijing University of Technology (A Bi-lstm-based Wi-fi Csi Approach for Non-intrusive Human Behavior Recognition In Smart Buildings). Energy Weekly News. October 24, 2025; p 322.