As latency-sensitive IoT applications proliferate, edge computing becomes critical to sustaining real-time performance. Yet, limited edge infrastructure and reliance on costly static task profiling constrain its potential. This paper presents fReeLoaders, an IoT ecosystem addressing both challenges through opportunistic offloading and adaptive scheduling. fReeLoaders uses nearby idle smart devices to augment edge availability and uses a deadline-driven reinforcement learning scheduler to learn task behavior on the fly, eliminating expensive a priori profiling. Evaluation on real hardware shows it improves quality of service by 11.4% over a state-of-the-art profiling scheduler and adapts to dynamic workloads.

Edge computing enables scalability and privacy improvements for Internet of Things (IoT) systems, by shifting applications from the cloud to edge servers closer to IoT devices. Conceptually, IoT devices communicate directly with the edge, but in real-world IoT deployments often IoT gateways are needed to bridge devices and edge servers. Design decisions at this gateway layer directly contribute to the responsiveness of edge applications and scalability of the platform, yet these gateways are often overlooked and under-explored. IoT gateways have a compelling mix of features, including reasonable compute capabilities, low cost, direct contact with devices, and spatial distribution in deployments. We hypothesize that a new management layer that organizes already existing gateways can replace expensive edge servers while enabling the privacy, reliability, and performance benefits of executing IoT applications on the edge. We utilize a decentralized architecture that creates a nexus among disjoint gateways using out-of-band discovery, low-overhead abstraction layers, and runtime application scheduling. This platform supports heterogeneous devices, minimizes configuration overhead, executes applications, and provides resiliency to failure. We develop a prototype of the architecture, NexusEdge, and deploy it across several gateways and hundreds of low-power and energy-harvesting devices. When compared to Amazon's AWS IoT Greengrass, NexusEdge shows a 10x improvement in application latency, and a 2.5x reduction in network traffic, indicating better scalability and responsiveness. We demonstrate how NexusEdge supports applications without cloud support, and envision future extensions of this platform.

Commercial Internet of Things (IoT) deployments are mostly closed-source systems that offer little to no flexibility to modify the hardware and software of the end devices. Once deployed, retrofitting such systems to an upgraded functionality requires replacing all the devices, which can be extremely time and cost prohibitive. End users cannot generally leverage deployed infrastructure to add their own sensors or custom data. However, we observe that IoT systems sometimes report battery voltage information to the cloud, and batteries are often user-serviceable. This indicates that perturbing the battery voltage to encode customized information could be a minimally invasive method to retrofit existing IoT devices.

Occupant identification is paramount for many building applications. Regardless, several practical concerns limit existing solutions to be ubiquitously deployed. Current systems are either intrusive, privacy-invasive, or require obtrusive, maintenance-heavy, and special-purpose infrastructure. As an alternative, the shadow pattern of a person reflected in the output voltage of a photovoltaic harvester power supply in many energy-harvesting devices can be used as a unique person identifying feature. In this paper, we present the first dataset containing the time-series open circuit output voltage traces of indoor photovoltaic cell corresponding to occupant door crossing events to perform occupant identification in smart homes. We collect shadow patterns of five participants from two different doors in two rooms of a building. The dataset consists of a total of 900 door entry and exit events during different hours of the day. We sample the voltage at 50 hz and provide the raw timestamped data. We also pre-process the data to filter the event of interest and label the data with associated occupant id and type of door events. Moreover, we provide insights into future research directions using the dataset. The dataset is available at https://doi.org/10.5281/zenodo.7195748

While relying on energy harvesting to power Internet of Things (IoT) devices eliminates the maintenance burden of battery replacement, energy generation fluctuation constitutes a major source of uncertainty to design reliable self-powered IoT devices. To characterize spatial-temporal variability of energy harvesting, data acquisition campaigns are needed across the range of potential harvesting sources. In this work we present a dataset to characterize thermal energy sources in residential settings by measuring thermoelectric generator (TEG) operating conditions over 16 deployment locations for periods ranging from 19 to 53 days. We present our easy-to-use thermal energy measurement platform built from off-the-shelf component modules and a custom TEG interface circuit. We demonstrate how the collected measurements can inform the design of energy harvesting IoT devices by deriving the TEG's maximum power output and estimating the available energy at each harvesting location.