Digital FIR Filter Design Using the Adaptive Window Function for Avian Acoustic Spotlighting

Biodiversity management is essential for preserving Earth's array of life, with birds playing a crucial role as indicators of environmental well-being. Avian acoustics has emerged as a valuable tool for assessing ecosystem health, forest degradation, and climate change. However, traditional bird survey methods are often invasive and time-intensive. Existing non-invasive techniques such as the deployment of passive acoustic and autonomous recorders encounter challenges such as high background noise, which can lead to misinterpretations. Digital filters are vital for processing acoustic signals particularly Finite Impulse Response (FIR) filters by reducing noise and limiting bandwidth. Yet, conventional FIR filter designs have limitations, especially regarding fixed window functions. This study proposed an innovative approach by employing an adaptive window function in FIR filter design using the Enhanced Cuckoo Search Optimization (ECSO) algorithm. This method dynamically adjusts the window function to balance main lobe width and sidelobe attenuation, thereby enhancing filter performance. The results of the simulation revealed that the proposed filter model achieved better filter characteristics and performance in terms of SNR, THD, and MSE. Likewise, the experimental setup, which involved testing the prototype in a natural habitat, showed that the avian acoustic spotlighting device successfully detected the intended bird vocalizations compared to other recording devices. Furthermore, its adaptable methodologies hold promise for addressing various signal processing needs beyond bioacoustics, thereby extending its impact and significantly contributing to ecological research and conservation efforts.