Skip to main content
QUICK REVIEW

[Paper Review] ChirpCast: Data Transmission via Audio

Francis Iannacci, Yanping Huang|arXiv (Cornell University)|Aug 28, 2015
Indoor and Outdoor Localization Technologies3 references3 citations
TL;DR

ChirpCast is a system that transmits network access keys via inaudible ultrasonic audio using commodity laptop speakers and microphones. It achieves robust, room-localized data transmission at 200 bits/second using differential phase-shift keying (DPSK), with over 90% bit transmission accuracy even at 2-meter distance under typical background noise conditions.

ABSTRACT

In this paper we present ChirpCast, a system for broadcasting network access keys to laptops ultrasonically. This work explores several modulation techniques for sending and receiving data using sound waves through commodity speakers and built-in laptop microphones. Requiring only that laptop users run a small application, the system successfully provides robust room-specific broadcasting at data rates of 200 bits/second.

Motivation & Objective

  • To address the challenge of providing selective, location-based access to public wireless networks without relying on traditional passkeys or radio-based solutions.
  • To design a system that enables nearly automatic, user-friendly key distribution while ensuring access is restricted to physical proximity.
  • To explore modulation techniques that maximize data rate and noise immunity using inaudible sound over standard laptop audio hardware.
  • To evaluate the robustness of audio-based data transmission under real-world environmental noise conditions such as music, conversation, laughter, and key jangling.

Proposed method

  • Uses ultrasonic frequencies between 18 kHz and 19.5 kHz to ensure inaudibility to humans while maintaining signal detectability.
  • Employs differential phase-shift keying (DPSK) modulation to improve noise immunity and enable reliable data recovery despite background interference.
  • Utilizes a dual-channel transmission scheme with left and right speaker channels to increase data rate and signal integrity.
  • Applies amplitude shaping at signal transitions to reduce spectral leakage and maintain signal power without affecting modulation or demodulation.
  • Employs adaptive kernel filtering and 96 kHz sampling rate to enhance signal processing and improve SNR at the receiver.
  • Relies on built-in laptop microphones and standard sound cards for reception, enabling deployment without specialized hardware.

Experimental results

Research questions

  • RQ1Can ultrasonic audio be used to reliably transmit network access keys with high accuracy in real-world environments?
  • RQ2How does the choice of modulation technique (e.g., FSK vs. DPSK) affect data rate and robustness under background noise?
  • RQ3What is the maximum achievable data rate for audio-based transmission using commodity laptop hardware while maintaining inaudibility and reliability?
  • RQ4How does background noise—such as music, conversation, laughter, or key jangling—affect bit transmission success rate?
  • RQ5To what extent can signal amplitude be increased to improve SNR without violating inaudibility constraints?

Key findings

  • ChirpCast achieves a maximum data transmission rate of 200 bits per second using DPSK modulation with over 90% bit transmission accuracy at a 2-meter distance.
  • The system maintains robust performance under various background noise conditions, with key jangling posing the highest risk due to its flat spectral power distribution.
  • Signal-to-noise ratio (SNR) has a strong exponential relationship with bit transmission success rate, indicating high sensitivity to SNR improvements.
  • At 1 meter distance, the system achieves over 90% accuracy at 200 bps, with bit error rate (BER) estimated at 1/(BTSR × 800) per trial, where BTSR is the success rate.
  • The use of DPSK over multiple orthogonal frequencies could theoretically increase the data rate by a factor of 2^k for k frequencies.
  • Amplitude scaling of the ultrasonic signal can be increased to improve SNR without affecting human audibility, offering a practical path to higher reliability.

Better researchstarts right now

From reading papers to final review, dramatically reduce your research time.

No credit card · Free plan available

This review was created by AI and reviewed by human editors.