Frequency-Masked Audio Watermarking for Imperceptible Tracking

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing media monitoring systems face challenges in efficiently and imperceptibly embedding and extracting audio watermarks to track media consumption and control device behavior, particularly in environments where audio signals are reproduced.

Innovation Solution

The system employs an encoder to insert inaudible audio watermarks into media signals using frequency masking techniques, and a decoder to recover these watermarks for tracking and control purposes, utilizing methods like discrete Fourier transformation and error correction to ensure robust extraction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If audio watermarks are embedded into media signals using frequency masking techniques, then the tracking and control capability is improved, but the audio signal may become distorted or the watermark may become audible

Engineering Contradiction:
Improvetracking capabilityVSAvoidaudio distortion
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The system dynamically adjusts watermark embedding parameters including frequency selection, amplitude modulation depth, and temporal distribution based on the host audio signal characteristics. The encoder analyzes the audio signal in real-time and modifies watermark parameters to ensure imperceptibility while maintaining detectability, resolving the contradiction between reliable tracking and audio quality preservation

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The watermark embedding process is made adaptive and dynamic rather than static. The system continuously monitors audio signal properties and adjusts watermark strength, frequency position, and temporal placement accordingly. This dynamic adaptation allows the system to maintain reliable tracking capability while preventing audible artifacts and distortion across varying audio content

Inventive Principle:
Principle #15Dynamics

2Reliability

If complex error correction and synchronization techniques are applied to watermark extraction, then the extraction reliability is improved, but the device complexity increases

Engineering Contradiction:
Improvewatermark extraction reliabilityVSAvoiddecoder complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

Error correction codes and synchronization patterns are pre-encoded into the watermark signal during the embedding process. Synchronization words and error correction data are inserted in advance at known positions, allowing the decoder to quickly acquire and verify watermark presence without complex real-time analysis, thus improving extraction reliability while limiting complexity growth

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system uses redundant copying of synchronization patterns and error correction data throughout the watermark sequence. Multiple copies of critical information are embedded at different time positions and frequency locations, enabling the decoder to recover the watermark even if some portions are lost or corrupted, thereby improving reliability without requiring overly complex decoding algorithms

Inventive Principle:
Principle #26Copying

3Measurement precision

If audio watermarks are embedded at high amplitude to ensure detectability, then the detection precision is improved, but the watermark becomes audible and distorts the original audio

Engineering Contradiction:
Improvewatermark detection precisionVSAvoidaudible watermark
Core Design Contradiction:
Measurement precisionVSObject-generated harmful factors

Solution Approach 1:

The watermark energy is distributed non-uniformly across different frequency bands and time intervals based on the local characteristics of the host audio signal. The system identifies frequency regions and time periods where the audio signal provides natural masking, and concentrates watermark energy in those regions. This local optimization allows detectable watermark levels without creating audible artifacts in critical frequency ranges

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system exploits the natural masking properties of the host audio signal by embedding watermark energy in frequency and temporal regions where the audio content naturally masks the watermark. What would normally be harmful (watermark energy that could be audible) is converted into a benefit by strategically placing it where the audio signal itself provides camouflage, thereby achieving detectable watermarks without audible distortion

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enables effective tracking of media consumption and control of devices through imperceptible audio watermarks, enhancing data collection and device responsiveness in media monitoring and control systems.

Implementation Method 1

a code may be inserted into the audio or video of media, wherein the code is later detected at monitoring sites

Methodology Applied
Scientific EffectFrequency masking:

Implementation Method 2

utilizing methods like discrete Fourier transformation and error correction to ensure robust extraction

Methodology Applied
Scientific EffectDiscrete Fourier transformation:

Data Source

PatentUS12437769B2Methods and apparatus to perform audio watermarking and watermark detection and extraction
Publication Date: 2025.10.07 THE NIELSEN CO (US) LLC
  • US12437769B2 patent drawing
  • US12437769B2 patent drawing
  • US12437769B2 patent drawing

AI summary

Methods and apparatus to perform audio watermarking and watermark detection and extraction are disclosed. Example apparatus disclosed herein are to select frequency components to be used to represent a code, different sets of frequency components to represent respectively different information, respective ones of the frequency components in the sets of frequency components located in respective code bands, there being multiple code bands and spacing between adjacent code bands being equal to or less than the spacing between adjacent frequency components in the code bands. Disclosed example apparatus are also to synthesize the frequency components to be used to represent the code, combine the synthesized frequency components with an audio block of an audio signal, and output the audio signal and a video signal associated with the audio signal.