Research from Queen Mary University of London, which was published in the journal PLOS ONE, examines how an AI strategy based on wireless signals could result in new emotion detection methods.

The research reveals radio waves' use to calculate heart rate and breathing signals, to predict how someone feels in the absence of other visual signals, including facial expressions.

“AI strategy based on wireless signals could result in new emotion detection methods.“

Achintha Avin Ihalage, a PhD student at QMUL, highlighted that the deep learning method is a novel neural architecture that can simultaneously process wireless time-dependent signal and frequency domain wavelet transformation images while retaining temporal and spatial relationships.

During the study, participants were asked to watch a chosen video to evoke sadness, anger, pleasure or joy. Researchers released harmless radio signals towards the individual, whilst the individual watched the video and measured the signals that bounced off them. Researchers stated that they could reveal undisclosed information about the individual's breathing and heart rates by analysing the signals' changes.

Ahsan Noor Khan, PhD student and first author, stated: “Being able to detect emotions using wireless systems is a topic of increasing interest for researchers as it offers an alternative to bulky sensors and could be directly applicable in future smart home and building environments. In this study, we’ve built on existing work using radio waves to detect emotions and show that the use of deep learning techniques can improve the accuracy of our results.”

Avin Ihalage stated: “Based on our results, this technology can detect emotions at an accuracy of seventy-one per cent. The experiment was conducted on fifteen participants and no female participants were involved. Ideally, a widespread study involving more participants should be performed to evaluate the generalisability of this method. We note that increasing the number of emotions considered could be useful for future applications, again, this requires more data.”