Zero-Waste Thinking in Digital Signal Processing: Reducing Computational, Data, and Energy Waste

Amjat Al Baihaqi (Al Azhar University of Indonesian)
Ibrohim Hasan (Al Azhar University of Indonesia)
Damar Royyan Saputra (Al Azhar University Of Indonesia)
Sofian Hamid (Institute for High Frequency Technology, RWTH Aachen University)
Dwi Astharini (Universitas Al Azhar Indonesia)

Exhibition and Seminar on Science and Creative Technology – Al Azhar Proceeding
Vol 3, No 1 (2025)

In the modern digital era, the concept of waste extends beyond the physical realm and into the digital domain, manifesting as redundant data, excessive computation, and continuous processing of unnecessary signals. This study introduces the application of zero-waste thinking to Digital Signal Processing (DSP), with a focus on minimizing computational, data, and energy waste in always-on audio systems. A lightweight, energy-aware Voice Activity Detection (VAD) method is proposed, utilizing signal energy and zero-crossing rate (ZCR) features to intelligently activate or suppress processing based on speech presence. MATLAB-based simulations were conducted to evaluate system performance under various noise conditions, measuring computational load, activation efficiency, and detection accuracy. The results show that the proposed approach significantly reduces unnecessary processing while maintaining reliable speech detection. This work offers a practical framework for sustainable and efficient DSP, contributing to the emerging paradigm of digital zero-waste systems.