Time-frequency composition of mosquito flight tones obtained using Hilbert spectral analysis

摘要: Techniques for estimating temporal variation in the frequency content of acoustic tones based on short-time fast Fourier transforms are fundamentally limited by an inherent time-frequency trade-off. This paper presents alternative methodology, Hilbert spectral analysis, which is not affected this weakness, and applies it to accurate estimation mosquito wing beat frequencies. Mosquitoes known communicate with one another via sounds generated their flapping wings. Active modulation between pairs mosquitoes thought take place as a precursor courtship. Studying acoustically-based interactions therefore relies representation flight time-evolving property, yet conventional spectrograms unable capture rapid modulations that exhibit. The algorithms introduced able automatically detect extract fully temporally resolved information from audio recordings. Application technique experimental recordings single tethered reveals corroboration previous reported findings. advantages method animal communication studies discussed, particular attention given its potential utility studying pairwise interactions.