SIGMA is a self-configuring intelligent building management system in which building state and conditions are observed over time using a wide variety of heterogeneous sensors. The two main goals behind the system are to increase the efficiency of building functionalities as well as to provide comfortable living conditions for the inhabitants. In order to achieve these two goals, buildings are equipped with multiple sensors and actuators. SIGMA provides the required means for collecting raw sensor data from the building, processing this data, and issuing the actuation commands required to control or adjust certain building functionalities.

SIGMA provides a wide variety of information (context) processors which are responsible for processing the collected raw sensor data and generating higher level context information which then can be used for issuing certain actuation commands. An example of such information processors is the device classification functionality (see our Smart Metering Toolkit for further details). Energy consumption traces of household appliances are collected over time and used by SIGMA to recognize the type of the running device and any abnormal behavior of household appliances.

SIGMA is based on complex event processing (CEP). An event bus is responsible for collecting all the events generated by sensors and context processors. A user-friendly web interface is provided for controlling all SIGMA functionalities and presenting the building state in an elegant and understandable way.

Multiple SIGMA systems are able to interact with each other and to provide their processing functionalities in terms of context processing services. The interaction is based on the Service Oriented Architecture (SOA) paradigm. SIGMA systems register their processing functionalities to a marketplace which enables them to be remotely found and invoked by other systems.