A Computational Model for Bats Echolocation Based on Mammalian Auditory System
Bats can easily pick up an individual prey among multiple targets by using the echoes of the emission that overlapped in time.So far,the process produced by the batsecholocation system is not completely understood.A computational Model for batsecholocation based on mammalian auditory system is proposed and analyzed in this work.This model is a description for the signal processing mechanism that may play a role in bats’ echolocation system.In this model,a cochlear filter bank is used to divide an FM emission and the echoes into several bands.Then a half-wave rectification and a low-pass filter were added to each path just like what have been done in the mammalian inner hair cell (IHC).The smoothed envelope at the output of the low-pass filter is used to trigger event by pick-detection mechanism.Then an integration processing is taken out among band-pass filter channels to improve the precision of the echolocation.At the end of the model a cross-correlation function is presented to play the coincidence-detection cell’s role in the mammalian auditory system.Bats can detect distances of the targets from the output of the cross-correlation module.Simulation results imply that this model can distinguish two glints with highly overlapped echoes and get accurate ranges for each target in noisy environments.
Binbin Cheng Hai Zhang
国际会议
北京
英文
2008-10-10（万方平台首次上网日期，不代表论文的发表时间）