Smart home sensor systems that infer residents' activities enable a number of exciting medical monitoring and energy conservation applications. Existing home activity recognition systems are invasive, since they require significant manual effort from the end user in installing or training the system, inconvenience the resident by requiring them to constantly wear tags, or require invasive cameras or expensive sensors. Our main hypothesis is that by effectively using data fusion techniques, leveraging the existing smart meter infrastructure in homes, and using only weak biometric sensing, we can build convenient, accurate home activity recognition solutions for the end user.The key challenges for home activity recognition systems addressed in this dissertation include reducing configuration effort from the end user, reducing sensor installation effort, and identifying residents in multi-person homes without using invasive sensors. To reduce user configuration effort, we develop an unsupervised activity recognition algorithm called AUTOLABEL that leverages data fusion and cross-home activity models to accurately recognize resident activities without user training. To eliminate many direct sensors in the home for activity recognition, we develop effective Bayesian data fusion techniques, which combine the existing smart meter infrastructure in homes, with one low cost, non-invasive sensor per room. In our WaterSense approach, we combine a single smart water meter per home, with an occupancy sensor per room to eliminate direct sensors on individual water fixtures. In our LightSense approach, we combine a single smart energy meter per home with a light sensor per room, to eliminate direct sensors on individual light fixtures. Finally, we propose the use of resident height, which is a weak biometric, in an effective data fusion approach, to identify residents for activity recognition in multi-person homes.We evaluate our proposed activity recognition solutions through short term prototype sensor deployments in home environments lasting from 7 to 10 days each. We show that our low cost, ii convenient solutions satisfy the activity recognition needs of numerous smart home applications, such as remote medical monitoring for elderly residents, and fine-grained resource consumption monitoring of light and water fixtures in the home. Finally, we observe that our unsupervised activity recognition algorithm can be used in a wireless snoop attack on smart homes, to infer the residents' daily activities with high accuracy in spite of encrypted wireless transmissions. We propose and evaluate a suite of privacy solutions to mitigate the inference accuracy of such an attack without affecting the performance or functionality of the home activity recognition system. iv