Sensor Signal Prediction at Unreported Frequencies

Case ID:

The Internet of Things (IoT) systems are increasingly complex as people continually prefer more information and automation. This trend has been growing in manufacturing, home safety, and building automation. One of the major realities faced by IoT systems is an overwhelming amount of data coming from sensors. Not only that, but the cost of having all sensors constantly monitoring on all frequencies is very high. This technology is a step toward a cost-effective solution.

The general idea in this invention is a framework for predicting unreported signals by estimating the correlation of those signals with historic and concurrent reported frequencies. Furthermore, the method uses sensors’ historical data in different scenarios coupled with a correlative factor and a gaussian distribution to predict other unreported frequencies. This enhances sensors’ adaptability into differing environmental conditions.

The Internet of Things (IoT) sensors and the data collected from there has increased dramatically over the past several years. And, while the number of sensors increase, the accuracy and versatility of sensor readings become more important. One of the problems associated with data collection accuracy is that processing large amounts of data is costly. And basic sensors may find it difficult to report wide ranges of frequencies which may therefore distort the reported data.

To more accurately assess the frequency data given from these sensors, this technology offers a way of predicting a more wholistic understanding of the data reported from IoT sensors from either unreported low or higher frequencies. This provides a way for those sensors to be more versatile and accurate.


  • All types of IoT sensors
  • Software analysis of the sensor data


  • Versatile
  • Widely applicable
  • Increasingly accurate

Status: issued U.S. patent #11,423,051

Patent Information:
Contact For More Information:
Tariq Ahmed
Sr Licensing Manager, College of Engineering
The University of Arizona
Lead Inventor(s):
Mohammed Shafae
Zhang Bing
Asthana Shubhi
Aly Megahed
Alaa Elwany