Prediction techniques for haptic communication and their vulnerability to packet losses

Fernanda Brandi, Eckehard Steinbach

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

13 Scopus citations

Abstract

We introduce three different uses of linear regression for improving the prediction of samples in haptic communication and compare this technique to the commonly employed zero-order and first-order linear predictors. We couple the prediction techniques with (error resilient) perceptual data reduction approaches and evaluate their robustness when losses in the network are present. Experimental results show that the proposed prediction technique improves haptic data reduction while keeping lower signal distortion compared to the traditional prediction methods when facing adverse network conditions.

Original languageEnglish
Title of host publicationHAVE 2013 - 2013 IEEE International Symposium on Haptic Audio-Visual Environments and Games, Proceedings
Pages63-68
Number of pages6
DOIs
StatePublished - 2013
Event2013 12th IEEE International Symposium on Haptic Audio-Visual Environments and Games, HAVE 2013 - Istanbul, Turkey
Duration: 26 Oct 201327 Oct 2013

Publication series

NameHAVE 2013 - 2013 IEEE International Symposium on Haptic Audio-Visual Environments and Games, Proceedings

Conference

Conference2013 12th IEEE International Symposium on Haptic Audio-Visual Environments and Games, HAVE 2013
Country/TerritoryTurkey
CityIstanbul
Period26/10/1327/10/13

Fingerprint

Dive into the research topics of 'Prediction techniques for haptic communication and their vulnerability to packet losses'. Together they form a unique fingerprint.

Cite this