Related Concept Videos
Deconvolution
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
What are Estimates?
The estimate for the mean of a sample is denoted by ͞x, whereas the mean of the population is designated as μ. Further, parameters such...
Estimation of k and VD of Aminoglycosides
Two-Dimensional (2D) NMR: Overview
The first step is the preparation period, during which nucleus A is excited with a radiofrequency pulse....
Short-distance Transport of Resources
Power System Three-Phase Short Circuits
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
Correction: He et al. An Edge-Computing-Based Emotion-Aware Adaptive Lighting System for Intelligent Cockpits. <i>Sensors</i> 2026, <i>26</i>, 3489.
Related Experiment Video
Updated: Jan 29, 2026

Deriving the Time Course of Glutamate Clearance with a Deconvolution Analysis of Astrocytic Transporter Currents
Published on: August 7, 2013
Short-Time Homomorphic Deconvolution (STHD): A Novel 2D Feature for Robust Indoor Direction of Arrival Estimation.
1Tech Innovation Group, KT Corporation, Seoul 03155, Republic of Korea.
This study introduces a new audio feature extraction method for precise indoor positioning. The novel approach significantly improves direction of arrival estimation accuracy in challenging acoustic environments.
Area of Science:
- Acoustics and Signal Processing
- Machine Learning for Sensor Fusion
- Robotics and Navigation
Background:
- Accurate indoor positioning and navigation face significant challenges.
- Conventional audio-based localization methods struggle with computational complexity, hardware synchronization, and reverberant environments.
- Existing machine learning approaches are limited by the discriminative power of input features.
Purpose of the Study:
- To propose a novel feature extraction method for multi-channel audio signals to enhance indoor positioning and navigation.
- To develop a robust deep learning model utilizing the proposed feature for accurate sound source localization.
- To address the limitations of conventional methods in complex acoustic conditions.
Main Methods:
- Introduced Short-Time Homomorphic Deconvolution for transforming audio signals into a 2D Time × Time-of-Flight representation.
- Employed a lightweight Convolutional Neural Network with a dual-stage channel attention mechanism for Direction of Arrival (DOA) estimation.
- Trained the model on a large-scale simulated dataset and validated it with real-world data from an anechoic chamber.
Main Results:
- The proposed feature effectively captures temporal evolution and stability of time-of-flight differences.
- The system achieved a Mean Absolute Error of 1.99 degrees in Direction of Arrival estimation in real-world scenarios.
- Demonstrated remarkable consistency between simulation and physical experiments, highlighting robustness.
Conclusions:
- Short-Time Homomorphic Deconvolution provides a rich and robust input for deep learning models in audio-based localization.
- The developed deep learning system offers precise and reliable Direction of Arrival estimation for indoor navigation.
- The proposed method overcomes limitations of traditional techniques and shows strong potential for practical indoor positioning systems.

