NC State researchers, Muhammad Shahzad, assistant professor in the department of computer science, and Computer Science Ph.D. student Raghav Venkatnarayan, wrote about a system they developed to help identify location and measure distance using Wi-Fi. Their approach, called Wi-Fi-assisted Inertial Odometry (WIO), uses Wi-Fi as a velocity sensor to accurately track how far something has moved.
“Essentially, the system’s movement method has been designed for devices that have Wi-Fi communication capabilities in them, so like a smartphone, tablet, drones, maybe robots and so on,” Shahzad said. “So if you want to measure how much distance that object has moved, then the system comes into play and helps out there.”
According to Shahzad, WIO technology can help with fitness tracking, virtual reality gaming, and can even be used to track customers and their shopping habits. Additionally, Venkatnarayan said the technology can be used to measure the dimensions of an object, quantify the size of a space and even help individuals locate their phone.
“I got my inspiration one day when I was running on a treadmill with my Fitbit, and just one day forgot to wear that heart rate monitor and I was thinking, ‘Okay we have so many sensors around us, but still the measurements that they take are not really high resolution,’” Venkatnarayan said. “My Fitbit told me the distance in steps, and wasn’t able to tell me the actual distance that I walked that day, and so that was that moment when I said ‘how can I make this work in a way that doesn’t require any infrastructure?’”
Venkatnarayan said initially the researchers tried to solve these issues with environmental features that weren’t always on hand, such as outdoor sound waves or correctly positioned microphones. However, Venkatnarayan said Wi-Fi signals are all around us, and because these signals measure velocity, they were integral in the creation of WIO.
“I saw that smartphones, they use these devices called accelerometers, but they are not good for tracking the distance over long periods, and then I found out that this is a very prevalent research problem that a lot of researchers have tried to solve by various means, but the means that they have employed are not always available,” Venkatnarayan said.
Following the realization that Wi-Fi could seriously contribute to location technology in smart devices, Shahzad said the first question they sought to answer was, “How do you use Wi-Fi to measure distance?” According to Shahzad, there is a well-established property that a Wi-Fi signal changes in proportion to the distance someone using it has covered, meaning the quality or reach of a Wi-Fi signal depends on how far or close someone is.
“We used this property basically to find out how to correct the error in the IMU,” said Shahzad.
IMUs, which stands for inertial measurement units, produce a small noise according to Shahzad, which disrupts the signal and makes it difficult to accurately measure distance. Shahzad and Venkatnarayan used Wi-Fi to correct errors introduced due to the noise.
That system, WIO, uses Wi-Fi signals to correct the large drift errors produced by accelerometers, which is how distance is traditionally measured indoors. This technology can be used practically by everyday people to track exercise or measure the dimensions of certain furniture accurately, and could also contribute to advances in less mundane ways such as robotics.
Shahzad and Venkatnarayan received interest from Sony, and are currently working with them to improve the accuracy of their product.
Venkatnarayan said that developing application programming interfaces (APIs) would allow their system to be implemented right away as software in the form of an app.
Students who are interested in learning more about this research can view their full research.
