科研速览 · Science Skim继续刷下去 · Keep skimming →
◆ Frontiers in Computational Neuroscience2026-04-08· Spectrogram

Simulated target search by bats using biomimetic SCAT biosonar model

James A. Simmons, Prithvi Thakur, Ashok Ragavendran, Chen Ming, Andrea Megela Simmons

原始摘要(英文原文)· Original abstract
Echolocating big brown bats broadcast short, wideband ultrasonic FM pulses for foraging and navigation. These broadcasts contain frequencies from 100 to 20 kHz (wavelengths 0.34-1.7 cm). Bats perceive target distance by measuring the time delay between the outgoing pulse and the returning echo. Acuity of this delay perception depends on the frequency content of echoes and the associated microsecond-level coherence between neural representations of the 1st and 2nd harmonic frequencies. Bats perceive target shape by estimating differences in the delay of mini-echoes from different reflecting points, or glints, within the target. A matched-filter receiver would register glints as prominent peaks in the pulse-echo cross-correlation output, but in bats the overlapping glint reflections mix together to create echo interference patterns that are transposed back into delay estimates. The process is modeled as spectrogram correlation and transformation (SCAT). The first, nearest glint is registered by echo delay itself, but subsequent glints are extracted from the nulls in the interference spectrum. Here, the SCAT receiver was evaluated for its ability to locate targets with a specific glint spacing in the 2D range/cross-range plane while rejecting other targets with larger or smaller spacings.
读原文 · Read the paper ↗

AI 追问PRO

登录后使用 AI 追问

讨论区

登录后参与讨论

相关论文 · Related

Simulated target search by bats using biomimetic SCAT biosonar model — 科研速览 Science Skim