A Hybrid Modality for Human Sensing

A Hybrid Modality for Human Sensing

Networked cameras are an integral component of many infrastructures and hold great promise for providing ubiquitous services. Camera networks however cannot persistently track and identify people and camera paths are usually fragmented due to the vision correspondence problem. To bypass the vision correspondence problem, we have developed a hybrid sensing modality comprised of cameras deployed in the infrastructure and inertial sensors (accelerometers and magnetometers) found in mobile phones. Signals from all modalities are combined to establish correspondence between the identity and observed location of the person wearing the sensors. 


Continue

Formalizing Sensor Node Lifetime Prediction

Formalizing Sensor Node Lifetime Prediction

Wireless devices with limited battery or intermittent harvested resources need to consider the tradeoffs between duty cycling, and application needs to guarantee reliable operation. After experiencing the design and implementation of a few generations of sensors nodes, we have developed a detailed framework for analyzing the asymptotic lifetimes of wireless sensor nodes with respect to application needs. This considers the architecture properties, choice of components and application profiles to determine how different architectures would perform in different environments.


Continue

Extracting Spatio-Temporal Activities from GPS traces

Extracting Spatio-Temporal Activities from GPS traces

Using GPS and camera data to understand how humans spend time at different places is a key component of many automated services from reminders and triggers to safety applications, feedback on how to conserve energy in buildings and energy management of smart phones. To allow embedded devices to exploit information contained in location data, we have developed a lightweight algorithm that can automatically extract Spatio-Temporal Activities (STAs) from traces of locations and classify them according to their temporal attributes. The resulting classes of STAs provide a summary of the important places and the schedule of the user.


Continue

The Behaviorscope Architecture

The Behaviorscope Architecture

The BehaviorScope architecture brings together a heterogeneous set of sensor driven data sources into a unified hierarchical framework that converts low-level sensor data to high-level semantic forms that can be used in decision making and to provide services. This processing is done using heterogeneous set of methods including Time-Augmented Probabilistic Context Free Grammars, temporal event classification, subsequence mining and other new forms of spatio-temporal data processing.


Continue

Prospective Students

ENALAB has new openings in 2012! 1 Postdoc, 1 Software Engineer, 1 Ph.D. Student  for Fall 2012, 2 undergraduate researchers. 

[ read more + ]

Recent Publications

A. Bamis and A. Savvides, Lightweight Extraction of Frequent Spatio-Temporal Activities from GPS Traces, to appear at RTSS 2010

[ read more + ]

Demos & Data

MATSNL is a tool for analyzing the asymptotic lifetime of wireless sensor node architectures with respect to application requirements...

[ read more + ]

News & Events

Athanasios Bamis joins the UConn CS Department as an Assistant Professor, ENALAB project receives Wells Fargo gift ...

[ read more + ]

ENALAB

The Embedded Networks and Applications Lab conducts research in networked embedded systems and wireless sensor networks. The lab's research emphasis is on new forms of computing in which wireless sensors with limited power resources are able to sense and interpret their environment and use the extracted information to provide services. As part of this process we are developing a sensor network architecture that interprets low-level data collected by sensors to high-level semantic information that is later used in the decision making process and closing the loop with the proper actuation and services. Under this umbrella, we are investigating the asymptotic lifetime behavior in wireless sensor nodes and low power embedded systems, human behavior interpretation using wireless sensors and human sensing. Our research derives its context from three main application domains:

  • Elder Care Technologies - we investigate how sensing technologies can be used to monitor elders at home by supervising safety and extracting spatio-temporal behavior patterns with the ultimate goal of predicting/preventing precipitous events and providing automated services. As part of this effort we investigate how to sense humans with new sensing modalities and how to build up higher precision macroscopic sensing from smaller, simpler sensors.
  • Security - using sensor systems to identify coordinating groups and interesting patterns in large sequences of data. The goal is to find important patterns that are embedded in longer traces of data without prior knowledge of the pattern structure.
  • Smart Buildings - we investigate the correspondence between building usage patterns and building power consumption patterns to identify energy waste, quantify consumption at the individual level and understand consumption breakdown within the building. For more details visit the Intelligent Buildings Website and the Spring 2011 Seminar.

 

To assist our research, our group is developing prototypes of real systems, an example demonstration of one of our lab testbeds for people sensing and tracking is shown below: