Your browser is not supported. Please update it.

18 July 2019

Creepy Celeno WiFi Doppler can personalize TV and detect intruders

Not to be left out of last week’s WiFi parade, Celeno has made one of the more unconventional technology announcements of the year so far – unveiling a product which uses WiFi for the purpose of imaging as a replacement for cameras and sensors. The technology sounds ahead of its time which is exactly why consumers are unlikely to trust it and why operators are unlikely to make any money from it.

Celeno’s new WiFi Doppler software, named after the Doppler effect which describes fluctuations in frequency or wavelength, aims to disrupt the hardware market by tracking object location without requiring additional devices. It does this by capturing the Doppler and Micro Doppler signatures of objects moving by applying advanced signal processing and machine learning algorithms to classify human postures, gestures and even breathing. Creepy, right?

WiFi Doppler is described as employing standard WiFi packets in combination with the Doppler effect to accurately characterize and represent complex motions and movements.

Security is WiFi Doppler’s primary purpose, operating in the 5 GHz band (and future 6 GHz), but the technology can also be applied to analytics in front of the TV for personalized content and targeted advertising. Naturally, Celeno is rather cagy regarding this personalization element which sounds intrusive beyond reckoning. Seemingly WiFi Doppler is able to detect viewers’ emotions and reactions to TV and cater content recommendations and ads accordingly. As we said, consumer trust is a huge issue here.

We contacted Celeno for an explanation on how WiFi Doppler helps deliver targeted ads and personalized content experiences. “It would be able to label the viewer as child or parent, as well as to analyze if there are no viewers, one viewer, two viewers or more. It would help to personalize as well as help advertisers measure their exposure. Doppler imaging is not detecting body contours and faces, but rather analyzing the movements – hence it is less intrusive to privacy vs. cameras and other technologies,” explained VP of Marketing and Business Development Lior Weiss.

The idea is to link WiFi Doppler up to a cloud-based platform with security and personalization functionalities, presumably of an operator’s choice, which could ping alerts directly to a smartphone app or alert the authorities, for example, although Weiss could not confirm any platform infrastructure information, nor could he provide an example of any field tests.

Such is the advanced nature of this technology, specifically citing it can identify human movements as opposed to general stuff falling over, that Weiss confirmed to Faultline Online Reporter that WiFi Doppler can tell the difference between a human intruder and the household pet cat causing mayhem.

Celeno claims that most consumers and businesses are reluctant to welcome intrusive additional equipment into their properties. It’s true that kitting a home out with a load of cameras and sensors comes at a cost and may look unsightly, but we would argue that such untested technology as WiFi Doppler would be viewed with more apprehension than a standard smart home camera.

We recall Ericsson discussing Doppler in the past relating to CP-OFDM (Cyclic Prefix Orthogonal frequency-division multiplexing), a longstanding and prominent digital encoding technique used widely in telecommunications and WiFi. Research from the Swedish vendor has said CP-OFDM technology is more robust to oscillator phase noise and Doppler than other multicarrier waveforms which is crucial for operation at high carrier frequencies like mmWave band.

With Celeno’s installed base of millions of units across Europe within the Liberty Global footprint, the company certainly isn’t short of potential data points from which its machine learning capabilities can slurp up metrics.

“The network is becoming the sensor,” reads the tag line, although the WiFi chipset maker has plenty of adversaries to convince.