FB pixel

NIST provides guidance for using mobile device biometrics to authenticate first responders

DHS S&T develops hands-free prototype

draft guidance on using mobile device biometrics

The U.S. National Institute of Standards and Technology (NIST) has issued draft guidance on using mobile device biometrics to authenticate first responders to help public safety organizations adopt them effectively.

The guidance was composed by team members from NIST, Mitre, and Scarfone Security. It covers the critical role of biometrics in authentication and biometric components in systems, including FIDO authenticators, as well as challenges to mobile biometrics’ effectiveness. Sections are also dedicated to biometrics on shared mobile devices and future biometric developments. The latter section delves into wearable sensors, the quality of behavioral biometric data, and biometric fusion.

“Using biometrics with mobile devices could potentially help make authentication faster and easier, but there are challenges with mobile device biometrics in general and also specifically for first responders,” the draft notes.

Understanding just how effective the biometric systems of mobile devices are can be challenging, the guidance notes, due to the proprietary details of those technologies, and the corresponding lack of independent verification of claims.

NIST is seeking public comments on the draft until July 19, 2021.

DHS developing noise-tolerant voice and speech recognition

The Department of Homeland Security (DHS) Science and Technology Directorate (S&T) and partners, meanwhile, have developed an automated speech recognition (ASR) technology to enable first responders to communicate hands-free and perform their emergency duties at the same time, according to a department release.

In a partnership with Johns Hopkins University Applied Physics Laboratory (APL) and subcontractor Think-A-Move, DHS S&T developed the Direct Artificial Intelligence System Interface, or DAISI, to provide voice-activated capabilities in noisy environments.

S&T identified the reduced effectiveness of speech recognition systems in the presence of background noise as a challenge to situational awareness common in first responder situations. What really differentiates the system from existing technologies, according to the report, is its machine learning biometric capability.

DAISI can provide voice control over a mobile data terminal for initial communications with central dispatch, as well as information like hydrant location and friction loss calculation (a water pressure measurement) to save time. The WISER (Wireless Information System for Emergency Responders) system can also be accessed through DAISI to identify hazardous materials and guide precautionary action.

A prototype of DAISI has been tested by Maryland’s Howard County Department of Fire & Rescue Services over the past three years, and validated to meet more than 80 different requirements to respond to a wide range of potential scenarios.

Work continues to keep the word error rate below the 15 percent industry standard, and to make DAISI more robust, with efficient processing and strong connectivity.

DAISI could reach commercial availability within the next couple of years, DHS says. One of the final hurdles is integrating the microphone into the self-contained breathing apparatus firefighters are equipped with. It will also be adapted to other first responders, and the article notes the technology could lead to the development of future innovations like wireless biometric sensors.

“This is a cornerstone technology,” said Ruth Vogel of APL. “If you don’t have the voice recognition capability, then a lot of the other next generation solutions are not going to work well.”

Related Posts

Article Topics

 |   |   |   |   |   |   |   |   | 

Latest Biometrics News


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Most Read From This Week

Featured Company

Biometrics Insight, Opinion

Biometrics White Papers

Biometrics Events

Explaining Biometrics