Towards a Simulation Framework for Smart Indoor Spaces.

Area of Science:

• Computer Science
• Ubiquitous Computing
• Human-Computer Interaction

Background:

• Sensor-based applications in smart homes and buildings rely heavily on effective sensor deployment configurations.
• Real-world testing of these configurations is resource-intensive, hindering development and optimization.

Purpose of the Study:

• To introduce a simulation-based methodology for evaluating sensor deployment configurations in smart indoor environments.
• To develop a toolkit, SIMsis, capable of modeling indoor spaces, occupant activities, and sensor behaviors.

Main Methods:

• Developed the SIMsis toolkit to simulate indoor environments, occupant activities, and diverse sensor behaviors.
• Focused on generating realistic, time-series event streams for individual sensors and aggregated sensor event compositions.
• Validated the simulator within the Smart-Condo ambient-assisted living platform for analyzing activities of daily living (ADLs).

Main Results:

• SIMsis demonstrated the capability to produce realistic agent traces and sensor readings.
• The simulation approach effectively models complex interactions within smart indoor spaces.
• The toolkit's output aligns with the requirements for evaluating sensor deployment strategies.

Conclusions:

• The SIMsis toolkit provides a viable, resource-efficient alternative to real-world testing for sensor deployment evaluation.
• This simulation methodology supports the development and deployment of more effective sensor-based applications for smart homes and buildings.
• The toolkit shows significant potential for advancing ambient-assisted living platforms and similar smart environments.