Wednesday, January 22, 2014

Talk Li Li

Neural Wireless Sensor & Actuator Network for Smart Home – Chair for Computer Aided Architectural Design

Date: January 22 2013, 12:00-13:00
Location: ETH Zürich Hoenggerberg Campus (Science City)
Building HIL Room E 67.1 (Bloody Hell)

There are many smart home demonstrators in research and industry. However, the adoption of smart homes in the market has remained slow over the last years. There are several reasons for that. First, most of the smart home systems in market work like a time or event triggered machine. In another word, they are not ‘smart’ enough. Secondly, they are not so easy to be installed into the existing houses. Thirdly, The high up‐front investment costs for the smart home system are also prohibitive.

To overcome these problems, we try to introduce Artificial Neural Network (ANN) and Wireless Sensor and Actuator Network (WSAN) into the smart home system. ANN is a Machine learning technique, which is widely used in pattern classification, prediction, control and optimization. WSAN refers to a group of sensors and actuators linked by wireless medium to perform distributed sensing and acting tasks. ANN can give the smart home the ability to learn from the residents and make decisions and predictions based on the habits of the residents. WSAN can make the whole system more lightweight and adaptive to the house environment.