Preventing Handheld Phone Distraction for Drivers by Sensing the Gripping Hand

Wang, Ruxin; Huang, Long; Wang, Chen · 2021 · arXiv

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the critical safety issue of handheld phone distraction while driving, which significantly increases crash risk. While existing methods focus on identifying whether a user is a driver or passenger, they fail to detect the precise moment a driver picks up their phone, preventing immediate intervention. The authors propose a system that monitors the phone’s status to distinguish between handheld and handsfree conditions, enabling real-time safety measures such as disabling non-essential apps or alerting nearby vehicles. The proposed system utilizes active acoustic sensing via the smartphone’s built-in hardware. It emits periodic ultrasonic pulses (18–22 kHz) that interact with the object contacting the phone. The resulting signals, captured by the phone’s microphones, are uniquely damped, reflected, and refracted depending on whether the phone is held by a hand or placed on surfaces like a seat, cup holder, or pocket. The system derives Short-Time Fourier Transforms (STFT) from these audio recordings to create time-frequency images. A Convolutional Neural Network (CNN) binary classifier processes these images to determine the phone’s status. To handle classification noise and transient states during grabbing or dropping, an adaptive window-based filter corrects errors and identifies the start, end, and duration of each distraction instance. Experiments were conducted with 14 participants, three smartphone models, and two car types across various driving conditions (city, highway, engine on/off). The system achieved a 99% accuracy in recognizing handheld phone-use instances and a median error of 0.76 seconds in estimating the start time of distraction. The CNN classifier demonstrated a True Positive Rate of 98.4% and a False Positive Rate of 0.5% when integrating data from both microphones. The system maintained high detection rates across diverse scenarios, including reading, texting, scrolling, and calling, as well as various handsfree placements. It proved robust against in-vehicle noises, such as engine sounds and car audio, and performed consistently across different device and car models. The significance of this work lies in its ability to provide precise, real-time detection of distracted driving without requiring additional hardware. By accurately capturing the timing of handheld phone use, the system enables immediate mitigation strategies, such as shutting down distracting apps while preserving emergency call functionality. This approach offers a practical solution to reduce traffic accidents caused by phone distraction, leveraging existing smartphone capabilities to enhance driver safety and support automated vehicle systems.

Key finding

Active ultrasonic acoustic sensing using a stock smartphone's existing speaker and microphones can discriminate handheld phone use from handsfree placements with 99% accuracy and sub-second onset timing, enabling real-time intervention against handheld-phone distraction without dedicated hardware.

Methodology

on_road

Sample size: 14

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via discover_arxiv on 2026-05-04 (4 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• distraction detection algorithms
• mobile phones
• distraction laws
• auditory
• manual
• visual manual

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: observational prevalence