As a matter of convenience, many current consumer devices switch to reduced energy consumption modes when not active in use, to be instantly ready for use upon explicit user invocation. Therefore the “standby” energy dissipation of electric appliances and devices have remarkably raised energy consumption in homes, offices and buildings. To overcome this problem we propose an implicit switching to reduced energy modes and even total “off” states based on a context aware activity tracking mechanism. As opposed to traditional activity tracking approaches, the PowerSaver is a single sensor, wireless, light weight, pocketworn and detachable at any time of use. This is a significant improvement over related activity tracking solutions. 1 Context Aware Energy Management In the smart home domain there is significant potential for saving energy simply by switching off devices (in the sense of physically disconnecting them from the power supply) instead of sending them to a standby state. Table 1 gives an excerpt about the generated energy consumption of devices in standby mode. According to this in table 2 the shares of some electronic appliances in an averTV laser printer satellite receiver multimedia center espresso machine 83.20 kWh 105.60 kWh 138.70 kWh 96.40 kWh 185.00 kWh Table 1. Annual energy consumption of typical household devices generated in standby mode[1]. age three person household are given: Although devices become more and more energy efficient, the growing number of electronic devices leads to an increase of energy consumption as high as 7.4% from 2003 to 2010 [2]. However, most of these devices support more than one energy state, thus the context aware energy management task is to set the devices in energy states that are best suitable for the current context.
[1]
Paul Lukowicz,et al.
Continuous recognition of arm activities with body-worn inertial sensors
,
2004,
Eighth International Symposium on Wearable Computers.
[2]
Joshua R. Smith,et al.
RFID-based techniques for human-activity detection
,
2005,
Commun. ACM.
[3]
Ling Bao,et al.
Activity Recognition from User-Annotated Acceleration Data
,
2004,
Pervasive.
[4]
Paul Lukowicz,et al.
Combining Motion Sensors and Ultrasonic Hands Tracking for Continuous Activity Recognition in a Maintenance Scenario
,
2006,
2006 10th IEEE International Symposium on Wearable Computers.
[5]
Albrecht Schmidt,et al.
Multi-sensor context aware clothing
,
2002,
Proceedings. Sixth International Symposium on Wearable Computers,.
[6]
Christopher G. Atkeson,et al.
Simultaneous Tracking and Activity Recognition (STAR) Using Many Anonymous, Binary Sensors
,
2005,
Pervasive.
[7]
Albrecht Schmidt,et al.
Multi-sensor Activity Context Detection for Wearable Computing
,
2003,
EUSAI.