Protecting your daily in-home activity information from a wireless snooping attack:
In this work, we expose a new privacy leak in residential wireless ubiquitous computing systems, and also propose guidelines for designing future systems to prevent this problem. We show that we can observe private activities in the home such as cooking, showering, toileting, and sleeping by eavesdropping on the wireless transmissions of sensors in a home, even when all of the transmissions are encrypted. We call this the Fingerprint and Timing-based Snooping (FATS) attack. This attack can already be carried out on millions of homes today, and may become more important as ubiquitous computing environments such as smart homes and assisted living facilities become more prevalent. We demonstrate and evaluate the FATS attack on eight different homes containing wireless sensors. We also propose and evaluate a set of privacy preserving design guidelines for future wireless ubiquitous systems and show how these guidelines can be used in a hybrid fashion to prevent against the FATS attack with low implementation costs.
Vijay Srinivasan, John A. Stankovic and Kamin Whitehouse. Protecting Your Daily In-home Activity Information from a Wireless Snooping Attack. In Ubicomp 2008. (19% acceptance rate)

Unsupervised activity inference in smart homes:
An important challenge in activity inference for smart homes is to reduce user effort in training and configuring the sensor system. To address this challenge, we design, implement and evaluate a multi-tier inference algorithm that leverages cross-home activity models to automatically infer resident activities and locations given very little information about the types of sensors and their installation locations. Our approach was found to have a promising inference accuracy of 80-90% based on real world sensor deployments in eight diverse homes for one week each.
Vijay Srinivasan, John A. Stankovic and Kamin Whitehouse. Protecting Your Daily In-home Activity Information from a Wireless Snooping Attack. In Ubicomp 2008. (19% acceptance rate)

Using Height Sensors for Biometric Identification in Multi-Resident Homes:
A significant drawback of indoor tracking and identification systems in smart homes is that they require residents to constantly wear tags or allow the installation of invasive camera systems. To address this challenge, we design and implement a height based resident identification and tracking system using ultrasonic rangers deployed above the doorways in a home. We found that this system is cheap, easy to install, and minimally invasive for the residents. Height is typically only a weak biometric, but we show that it is well suited for identifying among a few residents in the home, and can potentially be improved by using the history of height measurements at multiple doorways in a tracking approach. We evaluate this approach using 20 people in a controlled laboratory environment and by installing in 3 natural, home environments. We combine these results with public anthropometric data sets that contain the heights of residents in 2077 elderly multi-resident homes to conclude that height sensors could potentially achieve at least 95% identification accuracy in 95% of elderly homes in the US.
Vijay Srinivasan, John Stankovic and Kamin Whitehouse. Using Height Sensors for Biometric Identification in Multi-resident Homes. In Pervasive 2010. (17% acceptance rate)

WaterSense: Unsupervised Water Flow Disaggregation using Motion Sensors:
Smart water meters will soon provide real-time access to instantaneous water usage in many homes, and disaggregation is the problem of deciding how much of that usage is due to individual fixtures in the home. Existing disaggregation techniques require additional water sensing infrastructure and/or a manual characterization of each water fixture, which can be expensive and time consuming. In this work, we design, deploy and evaluate a novel system called WaterSense to perform fixture-level disaggregation using only a handful of inexpensive motion sensors. We propose a novel Bayesian clustering approach to automatically infer how many fixtures are in each room, and how much water each fixture uses. We evaluate the system using a 7-day in-situ evaluation in 2 diverse multi-resident homes with a total of 10 different water fixtures and 467 fixture events and observe that our approach achieves 86% classification accuracy in identifying individual water fixture events and 80-90% accuracy in determining the water consumption of individual water fixtures. We are currently scaling up the deployment of WaterSense and the ground truth sensing system to perform a longitudinal evaluation.
Vijay Srinivasan, John Stankovic and Kamin Whitehouse. WaterSense: Water Flow Disaggregation using Motion Sensors. In Buildsys 2011, co-located with Sensys 2011.

Providing Fine-Grained, Unsupervised Energy Feedback on Home Lighting: Light fixtures account for 12% of the annual household electricity bill, and providing fine-grained energy feedback on home lighting will enable residents to make informed decisions to save energy. Existing approaches to detect light fixture usage require expensive sensing hardware or deployments, are limited to certain lighting categories, or require manual training. In this work, we design, deploy and evaluate the unsupervised LightSense system, which combines two easily available noisy sensor sources, namely whole house power consumption data from utility smart meters with 16% precision, and one simple light sensor per room with 25% precision, to automatically detect the number of light fixtures, their usage times and overall energy costs with high accuracy. To address false positives from the light sensor and power meter data streams, we design algorithms for adaptive edge detection and data fusion, and a novel Bayesian matching algorithm that leverages long term correlation between the light and power data to accurately assign energy costs to ON-OFF events. Our approach is evaluated using in-situ sensor deployments in 4 homes for 10 days each, with light fixture events from 41 light fixtures across a total of 32 rooms. We observe 81% recall and 86% precision in detecting individual light fixture events, and 91.1% average accuracy in determining the energy cost of individual light fixtures consuming 90% of the home's lighting energy.
Vijay Srinivasan, John Stankovic, Kamin Whitehouse. LightSense: Using Light Sensors and Smart Meters to provide fine-grained Energy Feedback on Home Lighting. Under review - draft available on request.

Smart Thermostat:
Heating, ventilation and cooling (HVAC) is the largest source of residential energy consumption. In this work, we demonstrate how to use cheap and simple sensing technology to automatically sense occupancy and sleep patterns in a home, and how to use these patterns to save energy by automatically turning off the home's HVAC system. We call this approach the smart thermostat. We evaluate this approach by deploying sensors in 8 homes and comparing the expected energy usage of our algorithm against existing approaches. We demonstrate that our approach will achieve a 28% energy saving on average, at a cost of approximately $25 in sensors. In comparison, a commercially-available baseline approach that uses similar sensors saves only 6.8% energy on average, and actually increases energy consumption in 4 of the 8 households. My contribution to this project was to design, deploy and evaluate the underlying sensing and inference system using occupancy sensors and a HMM approach to infer the occupancy and sleep states of residents for use by the smart thermostat.
Jiakang Lu, Tamim Sookoor, Vijay Srinivasan, Gao Ge, Brian Holben, John Stankovic, Eric Field, Kamin Whitehouse. The Smart Thermostat: Using Occupancy Sensors to Save Energy in Homes. In Sensys 2010. (17% acceptance rate)

The Hitchhiker's Guide to Successful Residential Sensing Deployments:
Homes are rich with information about people's energy consumption, medical health, and personal or family functions. In this guide, we present our experiences deploying large-scale residential sensing systems in over 20 homes. Deploying small-scale systems in homes can be deceptively easy, but in our deployments we encountered a phase transition in which deployment effort increases dramatically as residential deployments scale up in terms of 1) the number of nodes, 2) the length of time, and 3) the number of houses. In this guide, we distill our experiences down to a set of guidelines and design principles to help future deployments avoid the potential pitfalls of large-scale sensing in homes.
Timothy Hnat, Vijay Srinivasan, Jiakang Lu, Tamim Sookoor, Raymond Dawson, John Stankovic, Kamin Whitehouse. The Hitchhiker's Guide to Successful Residential Sensing Deployments. In Sensys 2011. (19.5% acceptance rate)