Dependable IoT: circuit boards on a table

Dependable IoT

graphic showing dependable data

The proliferation of sensors and the reliance on telemetry/data has never been greater and continues to show a propensity for exponential growth. Billions of IoT devices are currently being deployed from agricultural farms, factory floors, to smart buildings/cities to sense and monitor the environment. This growth is fueled by technological innovation that has enabled manufacturers to produce large numbers of sensors at a fraction of the cost. While the manufacturing scale has led to affordable sensors, it also has the unintended consequence of reduced durability. Former CTO of GE Digital, called out “About 40 percent of all data from the edges of IoT networks is spurious (opens in new tab)”. KPMG had a report indicating “84% of CEOs are concerned about the quality of the data they’re basing decisions on (opens in new tab).” And the effect of bad data can lead to cascading ill-effects upstream, leading to ineffective and at times catastrophic decisions.

Data quality plays a vital role in the increasing adoption of IoT devices, where organizations rely on the dependability of the data for all the decision support and management functions. Our group has progressed on a research agenda, with fundamental IPs, to ensure that IoT devices deployed in any of the numerous domains are built to be highly dependable.

Dependability of an IoT device is directly linked to the trustworthiness and reliability of the data received from the IoT device. 

The reliability of the data depends on whether the device collecting the data is non-faulty, calibrated properly, and whether the reported data is acceptable. We have state of the art techniques to remotely monitor the health of the sensor (working, faulty, drifted) using lightweight Software APIs.

Our core value proposition is to provide a simple and easy way to (i) remotely measure and observe the health of a sensor, and (ii) empower users to specify their acceptable data quality threshold driven by the application requirements. The core technology behind this value proposition is the ability to automatically generate a working sensor fingerprint. These sensor fingerprints capture the electrical properties of the sensor such as voltage and current, which are unique to each sensor circuit design. This can be measured alongside the data being captured on the IoT device in real-time.