Skip to main content

Researchers From MIT University Have Designed An Algorithm To Read Human Emotions Through Wireless Signals


About Wi-fi, you might possibly discuss its range, multiple access points, network security or data security, etc. But what if somebody interferes the discussion and adds that Wifi can read human emotions, something you possibly might not believe? Obviously, why would you, it is like placing computer and technical science with psychology in the same division.

Anyways, we know human emotions are very complex. It is a hard task to exactly know what the other person is feeling. But what if technology lets us know about someone’s feeling?
MIT’s Computer Science and Artificial Intelligence Laboratory has come out with a device called ‘EQ Radio‘ that will help detect emotions using wireless connections.
The device uses breathing and heart rhythms calculating minute changes and is 87% accurate in detecting the emotions.
The project lead Dina Katabi says, “Our work shows that wireless signals can capture information about human behavior that is not always visible to the naked eye.” Katabi also added that the team aims to bring a result that would help medical sciences to diagnose conditions like depression and anxiety.
The EQ Radio which uses wireless signals and no on-body sensors, measures heart beats just like an ECG machine. The error margin of EQ Radio is just 0.3%.
The wireless signals emitted by the device is reflected off by the individual’s body back to the device. The reflections are broken into individual heart beats to check the small variations that would determine the levels of arousal and positive affect. If an individual’s signals show low arousal and negative affect, it implies he is sad.
If compared, EQ Radio is more accurate in detecting the emotions than Microsoft’s Emotion API which examines only facial expressions.
Much to our surprise, the system can also find if anybody’s heart skips a beat. This opens up and further proposes a lot of diagnostic possibilities of many diseases.

Comments

Follow Us

WHAT'S HOT

Google’s Open Source DIY Kit Turns Your Raspberry Pi Into An AI Assistant

The latest edition of the MagPi magazine includes a DIY kit created by Google that can be used to create a custom Google Home device powered by Raspberry Pi. A user can take advantage of the Google Assistant SDK and Google Cloud Speech API to enable voice control in their projects.

CopyCat Has Infected About 14 Million Android Smartphones And Rooted About 8 Million Of Them

The Check Point security researchers have published the details of a massive malware campaign that infected Android smartphones. Called CopyCat, this malware was able to infect about 14 million devices and root 8 million of them. The malware was spread via phishing and third-party app stores which hosted malicious versions of popular apps.

Nvidia Volta GPU Has 21 Billion Transistors And 5,120 Cores — “You Can’t Make A Chip Any Bigger”

At its annual GTC conference, Nvidia unveiled its new Volta architecture. Based on the same, Tesla V100 data center GPU comes with whopping 21 billion transistors that deliver an equivalent performance of 100 CPUs for deep learning. Compared to the current Pascal architecture, Volta delivers 5x improvements in peak teraflops. Volta chips are expected to be used by tech giants like Google, Amazon, Facebook, Baidu, etc. in their applications.