An Unsupervised Learning Approach for Data Detection in the Presence of Channel Mismatch and Additive Noise

TL;DR

This paper explores an unsupervised clustering-based method for detecting encoded q-ary data over noisy channels with unknown characteristics, improving detection reliability through constrained coding.

Contribution

It introduces an unsupervised clustering approach for data detection under channel mismatch and noise, and demonstrates how constrained coding enhances detection performance.

Findings

k-means clustering performance evaluated without constrained coding

Constrained codes improve detection reliability in uncertain channels

Method applicable to channels with partially unknown characteristics

Abstract

We investigate machine learning based on clustering techniques that are suitable for the detection of encoded strings of q-ary symbols transmitted over a noisy channel with partially unknown characteristics. We consider the detection of the q-ary data as a classification problem, where objects are recognized from a corrupted vector, which is obtained by an unknown corruption process. We first evaluate the error performance of k- means clustering technique without constrained coding. Secondly, we apply constrained codes that create an environment that improves the detection reliability and it allows a wider range of channel uncertainties.