Interact with Intel® product support specialists on client concerns and recommendations
22 Discussions

Wi-Fi Sensing: Adding Sensing Capability To Intel Wireless Platforms

1 1 6,902

Rahul Shah is a research scientist at Intel Labs, where he focuses on technologies and applications related to mobile and ambient sensing for personal, health, and industrial usages.


Wi-Fi is one of the most widely-used technologies today, with dozens of Wi-Fi-equipped devices in a typical household. While innovations over the last couple of decades have brought us faster data rates, seamless connectivity, and better security, we are now on the cusp of another capability built over ubiquitous wireless technology: ambient sensing. This emerging ability is the basis for the respiration detection demo that Intel Labs will present at the upcoming Innovation Event on September 27-28 in San Jose, CA. Beyond current works, the technology shows great promise for additional features and applications in the future.

The basic idea is relatively simple; wireless signals bounce off everything in the environment – walls, furniture, people, etc. This means that any motion or change in the environment, such as a person walking or even a simple hand gesture, results in a variation in the Wi-Fi signals. Wi-Fi sensing uses the information embedded in this time-varying multi-path signal to understand the cause of the motion, hence sensing ambient motion in the environment. Figures 1 and 2 below demonstrate multiple channel state information (CSI) curves with the CSI amplitude plotted on the Y-axis and the Wi-Fi Orthogonal Frequency Division Multiplexing (OFDM) subcarriers on the X-axis. Each curve is the CSI response for a single Wi-Fi transmission, so the overlapping curves in Fig 1 show that the CSI is not changing much over time. On the other hand, the CSI response changes significantly in Fig 2, which shows the case of a person walking in the room. It is also important to note that any Wi-Fi communication transmits simultaneously over multiple frequencies.


Researchers at Intel Labs built on this idea to demonstrate the use of Wi-Fi for sensing concurrently during communication. Since the development of the IEEE 802.11n Wi-Fi standard in the late 2000s, devices measure the CSI during transmission for the purposes of optimizing the communications to maximize the data rate and reliability. Wi-Fi sensing uses this measurement of the multi-path channel via CSI information and applies signal processing and machine learning techniques to extract meaning from the varying CSI. This can be used in multiple ways: to detect human proximity, room occupancy, gestures, chest movement due to respiration, physical activity monitoring, and more.

Yet Another Sensor?

Wi-Fi sensing has some clear advantages compared to other sensing technologies. One which has been alluded to is the reuse of hardware already present on most platforms to perform the new function of sensing. Additionally, with multiple Wi-Fi devices in most environments, we can get excellent ambient sensing coverage, plus multi-path effects give us 360 degree sensing around the device. Another big advantage is it is much more privacy-friendly. Unlike a camera, it only captures motion in a given environment. This is much less information than a camera gathers, yet in most cases that is perfectly adequate for what is needed. A good example of this is elderly care or in-home scenarios where there is a need to detect falls or physical activity levels, but cameras are not acceptable for privacy reasons. Finally, it can sense micro-motions (e.g. chest movement) that are not visible to cameras or other ambient sensors.

Wi-Fi Sensing Program at Intel

Building on the initial proof of concept, a group of researchers and engineers across Intel Labs, Wireless Communications Solution group, Next Generation Standards group, Client Computing group and Internet of Things group delivered a TechEd (now called TSD – Technology Strategic Discussion) in 2020 that led to the formation of a Wi-Fi sensing program across the company. The program enabled CSI on all Intel Wi-Fi platforms and developed novel algorithms for sensing along with other new features for Intel PCs.

This has resulted in the first commercial Wi-Fi sensing feature of Wake-on-approach and Walk-away lock on Intel’s Raptor Lake platform. This feature detects when a person approaches the PC so as to bring it out of sleep and get it ready for user input by the time the person reaches the device. Similarly, the PC can detect when a person walks away, automatically locking the device.

Further features based on this technology are already in development. With an initial focus on respiration detection, we hope to extend the technology to detect other physical activities as well. Intel Labs will demonstrate an early prototype of breathing detection at the Intel Innovation Event on September 27th and 28th. The solution detects the rhythmic change in CSI due to chest movement during breathing. It then uses that information to detect the presence of a person near a device, even if the person is sitting silently without moving. The respiration rates gathered by this technology could play an important role in stress detection and other wellness applications.

Finally, Intel is also deeply involved with the IEEE 802.11bf working group which is developing standards related to Wi-Fi sensing. Wi-Fi sensing is becoming a core part of our Wi-Fi offerings and will add to the suite of sensing and context-awareness capabilities on our platforms. As Wi-Fi gets more and more entrenched in our devices, the sky is the limit for novel usages based on this new capability.

Tags (1)
1 Comment


Can you provide me information regarding what I need to analyse. I am more interested to know more about it. Any information regarding about sensing.

Thanks and regards,




About the Author
Rahul C. Shah received a B.Tech. degree (Hons.) in electronics and electrical communications engineering from the Indian Institute of Technology, Kharagpur, India, as well as M.S. and Ph.D. degrees in electrical engineering from the University of California at Berkeley, Berkeley, CA, USA. He is a Research Scientist and a Principal Engineer with Intel Labs and focuses on technologies and applications related to mobile and ambient sensing for personal, health and industrial usages. He has over 40 patents and 30 publications in peer-reviewed conferences and journals. His research interests are in wireless sensing, communications, wireless networking, location sensing, and applied machine learning.