Morphological Detector for Multilevel Signals in epsilon-Noise

TL;DR

This paper introduces a morphological detector for multilevel signal detection in impulsive non-Gaussian noise channels, leveraging 3D image processing techniques to improve demodulation performance.

Contribution

It presents a novel method that applies morphological nonlinear image filtration principles to two-dimensional signals for enhanced detection in challenging noise environments.

Findings

Significant performance improvement demonstrated

Successful application of 3D image processing algorithms

Encourages further research in image-based signal processing

Abstract

The novel approach was developed for multilevel signal detection in channels with impulsive non-Gaussian noise. This approach consists of using morphological nonlinear image filtration principles for two dimensional signals. It is a new method of signal demodulation, using three - dimensional image processing algorithms. Successful results of this morphological detector encourage more investigation towards using image processing theory and algorithms for two dimensional signal processing. As can be seen in the example in section IV, this new approach of reusing well developed and extensively developing image processing has significantly improved performance.