A computational model of the bottlenose dolphin sonar: Feature-extracting method

TL;DR

This paper presents a computational model of bottlenose dolphin sonar that mimics echo-image formation and demonstrates comparable echo-discrimination capabilities to real dolphins through formalized data and testing.

Contribution

The study introduces a novel computational model based on formalized echo-image data, achieving dolphin-level echo-discrimination performance.

Findings

Model matches dolphin echo-discrimination capabilities

Formalized data effectively simulate dolphin sonar perception

Computational approach offers a new tool for sonar research

Abstract

The data describing a process of echo-image formation in bottlenose dolphin sonar perception were accumulated in our experimental explorations. These data were formalized mathematically and used in the computational model, comparative testing of which in echo-discrimination tasks revealed no less capabilities then those of bottlenose dolphins.