Unlock Your Phone with Your Unique Breath: Study Finds Distinctive "Fingerprint"

Imagine your breath being the key to unlocking your smartphone or other devices. It might sound like something out of a sci-fi movie, but researchers at the Indian Institute of Technology Madras in Chennai are exploring the potential of using the turbulence in the air created by breathing as a unique method for biometric authentication.

Led by Mahesh Panchagnula, the team initially set out on a different mission. They aimed to create an artificial intelligence (AI) model capable of identifying individuals with respiratory issues for medical purposes. To do this, they recorded breathing data using an air pressure sensor. Little did they know that their experiments would lead to a surprising discovery with implications for smartphone security.

Unlike traditional methods such as fingerprint scans, the researchers found that the turbulence generated by breathing could serve as a more secure authentication method. One remarkable advantage is its morbid nature – this method cannot be fooled by a deceased person, offering a level of security that other biometric tests may lack.

The team recorded 10 breaths from each of the 94 human test subjects, collecting data at a rapid rate of 10,000 times per second using the air pressure sensor. This wealth of breathing data was then inputted into the AI model, revealing insights that surpassed their initial goal of identifying respiratory issues.

This innovative approach not only opens the door to novel authentication methods but also highlights the unexpected benefits of scientific exploration. Who would have thought that something as common as our breath could play a role in securing our digital devices? As technology continues to advance, it's exciting to see researchers uncovering new possibilities in the most unexpected places.

The researchers stumbled upon a fascinating revelation during their investigation. After analyzing breath data from a specific individual, their AI model could determine with over 97% accuracy whether a new breath belonged to that same person or not.

In a subsequent trial where the model had no prior information about the person logging in, it displayed an accuracy of more than 50% in correctly attributing the breath to one of two potential individuals.

Mahesh Panchagnula sheds light on the AI model's ability to recognize distinct patterns of turbulence linked to the unique anatomical configurations of each person's extrathoracic region. This region includes the nasal and oral passages, pharynx, and larynx.

While acknowledging the potential for improvement through refinement, Panchagnula emphasizes the initial findings confirm the feasibility of the concept. Amidst the plethora of existing biometric systems, he underscores a distinctive advantage of the breath test – its exclusive efficacy with living individuals. Notably, reports have surfaced regarding the unlocking of a deceased person's smartphone through their fingerprint, a vulnerability effectively addressed by the new system.

Panchagnula succinctly captures the essence of the breath test, stating, "You can't make a dead person pass this test because you need the person to exhale." The simple yet effective nature of this requirement adds an extra layer of security, making this innovative biometric method a promising contender in the ever-evolving realm of technology.

Fingerprint

Unlock Your Phone