Accurate, real-time measurement of energy expended during everyday activities would enable development of novel health monitoring and wellness technologies. A technique using three miniature wearable accelerometers is presented that improves upon state-of-the-art energy expenditure (EE) estimation. On a dataset acquired from 24 subjects performing gym and household activities, we demonstrate how knowledge of activity type, which can be automatically inferred from the accelerometer data, can improve EE estimates by more than 15% when compared to the best estimates from other methods.

Individuals with Autism Spectrum Disorders (ASD) frequently engage in stereotyped and repetitive motor movements. Automatically detecting these movements in real-time using comfortable, miniature wireless sensors could advance autistic research and enable new intervention tools for the classroom that help children and their caregivers monitor and cope with this potentially problematic class of behavior. We present activity recognition results for stereotypical hand flapping and body rocking using data collected from six children with ASD repeatedly observed in both laboratory and classroom settings. In the classroom, an overall recognition accuracy of 88.6% (TP: 0.85; FP: 0.08) was achieved using three sensors. Challenges encountered when applying machine learning to this domain, as well as implications for the development of real-time classroom interventions and research tools, are discussed.

A major challenge in pervasive computing is to develop systems that can reliably recognize human activity patterns, such as bathing from sensor data. Typical sensor deployments generate sparse datasets with thousands of sensor readings and few instances of activities. The imbalance between the number of features (i.e. sensors firing) and the classification targets (i.e. activities) complicates the learning process. In this paper, we propose a novel framework for discovering relationships between sensor signals and observed human activities from sparse datasets. The framework builds on the use of Bayesian networks for modeling activities by representing statistical dependencies between sensors. This allows us to solve two key problems: firstly, how to automatically determine an effective structure for a Bayesian network that recognizes a particular activity without human intervention; and, secondly, we address the pragmatic problem of sparse training data, where the data available to train the activity recognizers is limited. In our approach, we `learn' the structure of the Bayesian networks automatically from the sensor data. We optimize this process in 3 ways: firstly, we perform multicollinearity analysis to focus on orthogonal sensor data with minimal redundancy. Secondly, we propose Efron's bootstrapping to generate large training sets that capture important features of an activity. Finally, we find the best Bayesian network that explains our data using a heuristic search that is unbiased to the ordering between consecutive variables. We evaluate our approach using a data set gathered from MIT's PlaceLab. The inferred networks correctly identify activities for 85% of the time

This paper describes methods for service selection and service access for mobile, sensor-enhanced web clients such as wireless cameras or wireless PDAs with sensor devices attached. The clients announce their data-creating capabilities in "Produce" headers sent to servers; servers respond with forms that match these capabilities. Clients fill in these forms with sensor data as well as text or file data. The resultant system enables clients to access dynamically discovered services spontaneously, as their users engage in everyday nomadic activities.