Compressive Sensing Based Compression Algorithm for the Audio Signal

Title

Compressive Sensing Based Compression Algorithm for the Audio Signal

Creator

Electa Alice Jayarani, A.; Joy, Helen K.; Thenmozhi, S.

Description

Nyquist sampling is used in the conventional digital converter to convert the analog audio signal to digital audio data. Traditional approaches, such as Nyquist sampling, require high sampling rates, leading to large datasets. After sampling, the digital audio data is often compressed using algorithms like Moving Picture Experts Group Layer-3 Audio Coding (MP3), Advanced Audio Coding (AAC), or other codecs to re-duce the file size for storage or transmission. Recently, Compressive Sensing (CS) has been used instead of Nyquist Sampling. Instead of sampling at the Nyquist rate, Com-pressive Sensing samples the signal at a much lower rate. The recovery of the original signal from these fewer samples is possible through minimization techniques. Hence, in the proposed algorithm, we use Compressive Sensing for audio compression. The basis matrix for Compressive Sensing is generated by exploring different transform matrices. The proposed algorithm leverages the principles of Compressive Sensing to enhance audio compression by reducing the number of samples needed and using efficient recovery techniques, resulting in a high compression rate suitable for modern audio applications. 2025.

Source

Journal of Information Hiding and Multimedia Signal Processing;Volume;16;Issue;1;pp.453-465

Date

01-01-2025

Publisher

Taiwan Ubiquitous Information CO LTD

Subject

Audio Compression; Audio signal; Compressive sensing; l1 minimization

Coverage

Electa Alice Jayarani A., Department of Electronics and Communication, K.S. Institute of Technology, Bengaluru, India; Joy H.K., Department of Computer Science, Christ University, Bengaluru, India; Thenmozhi S., Department of Electronics and Communication, Dayananda Sagar College of Engineering, Bengaluru, India

Rights

Restricted Access; Hardcopy may be available in the library

Relation

ISSN: 20734212;

Format

online

Language

English

Type

Article

Identifier

https://www.scopus.com/pages/publications/105000076339?origin=resultslist