Summary
For an improved neuro-spike representation of auditory signals within cochlea models, a new digital ARMA-type low-pass filter structure is proposed. It is compared to more conventional AR-type counterpart ou a classification of biosonar echoes, in which echoes from various tree species insonified with a bat-like chirp call are converted to biologically plausible feature vectors. Next, parametric and non-parametric models of the class-conditional densities are built from the echo feature vectors. The models are deployed in single-shot and sequential-decision classification algorithms. The results indicate that the proposed ARMA filter structure offers an improved single-echo classification performance, which leads to faster sequential-decision making than its AR-type counterpart.
See the full content of this document
Extract
A New Digital Cochlea Model Neuro-Spire Representation of Auditory Signals and Its Application to Classification of Bat-Like Biosonar Echoes
(ProQuest: ... denotes formulae omitted.)
1. IntroductionGeneral audio signal processing for simulating the model of mammalian auditory system consists of a two-stage operation: cochlear filtering (preprocessing) followed by coding, [I]. In the first stage, the input waveform is passed through a bank of band-pass filters (BPF), in which, mostly, the center frequencies /c of BPFs are varied systematically, while retaining the 3dB quality factor(s), Q-sdB, constant, which is commonly adopted by gammatone filter bank models, [2]. The output of each channel BPF is then passed to an envelope extraction in the form of half-wave rectification followed by low-pass filtering to emit spike trains.In contrast to well-understood preprocessing, the actual mechanism underlying the coding and processing coded spikes at upper level is not known thoroughly. In order to introduce a biologically-plausible spatio-temporal representation of auditory signals, Ho dgkin- Huxley (HH) equations, which yield a single-pole low-pass filter (LPF) structure, can be considered to account for action potential dynamics as an auditory nerve cell membrane model, [3], [4]. However, for most biologicallyinspired applications, [5-8], HH equations have highly nonlinear interdependence between model parameters due to such complications as inactivation of ion levels, which acts as a prohibiting factor in implementation.As a simple yet more practica...See the full content of this document
Sponsored links
