Spotlight is a novel application that monitors electrical energy consumption at the individual level. Obtaining reports of energy consumption at this fine granularity allows identifying new areas for energy saving and acting upon it in real-time. Spotlight views appliances as rendering a service to a user and the energy consumption associated with the appliance as a cost for the service. Each participating appliance is specified a service range, a physical vicinity from the appliance within which the user benefits from the service. Using radio receive strength from user-wearable active RFID tags, an appliance is able to determine the users in its service range. In order to make these measurements, each appliance is instrumented with a power meter and an active RFID tag reader. The current implementation of Spotlight uses a COTS power meter and MicaZ motes as active RFID tags and readers.

The Spotlight system is deployed and tested in an experimental setup with various appliances and users. We evaluate multiple schemes of accounting energy consumption based on an individual's movement profile and the appliances' power profile. Our preliminary results show how the system could detect wasted energy and we discuss interesting individual behavior that could be interpreted for various energy optimizations.

Advances in DSP technology create important avenues of research for embedded vision. One such avenue is the investigation of tradeoffs amongst system parameters which affect the energy, accuracy, and latency of the overall system. This paper reports work on benchmarking the performance and cost of Scale Invariant Feature Transform (SIFT) for visual classification on a Blackfin DSP processor. Through measurements andmodeling of the camera sensor node, we investigate system performance (classification accuracy, latency, energy consumption) in light of image resolution, arithmetic precision, location of processing (local vs. server-side), and processor speed. A case study on counting eggs during avian nesting season is used to experimentally determine the tradeoffs of different design parameters and discuss implications to other application domains.

Advances in DSP technology create important avenues of research for embedded vision. One such avenue is the investigation of tradeoffs amongst system parameters which affect the energy, accuracy, and latency of the overall system. This paper reports work on benchmarking the performance and cost of Scale Invariant Feature Transform (SIFT) for visual classification on a Blackfin DSP processor. Through measurements and modeling of the camera sensor node, we investigate system performance (classification accuracy, latency, energy consumption) in light of image resolution, arithmetic precision, location of processing (local vs. server-side), and processor speed. A case study on counting eggs during avian nesting season is used to experimentally determine the tradeoffs of different design parameters and discuss implications to other application domains.