Please use this identifier to cite or link to this item: https://repository.cihe.edu.hk/jspui/handle/cihe/996
Title: Packet-loss-resilient perception-based haptic data reduction and transmission using ACK packets
Author(s): Pang, Raymond Wai Man 
Author(s): Qin, J.
Choi, K.-S.
Xu, R.
Heng, P.-A.
Issue Date: 2012
Publisher: IEEE
Related Publication(s): Proceedings of the 2012 IEEE 11th International Conference on Signal Processing (ICSP), Volume 2
Start page: 1165
End page: 1170
Abstract: 
Numerous studies have shown that haptic interaction plays a key role in enriching the sense of immersion and copresence of distributed users in collaborative virtual environments (CVEs). However, to ensure high-fidelity haptic interaction in CVEs, a high packet rate is required, resulting in considerable increase in overall traffic over the network. While perceptual deadband approach for haptic signals can successfully reduce high packet rates, this method is vulnerable to packet loss because the losing of single perceivable update packet results in a succession of wrong predictions. In this paper, we improve the perceptual deadband model by incorporating a packet loss resilient scheme using acknowledgement (ACK) packets. In case that an ACK packet is not returned to the sender within a trigger time, the sender will send an additional haptic data packet to the receiver to offset the effect caused by packet loss. By carefully selecting trigger time, buffer length and acknowledge time, the proposed scheme can be applied in a network environment with variable packet loss rates. The ACK-packets-based scheme is experimented by using different packet loss rates and lengthes of burst loss. Experimental results show that the proposed scheme can maintain stable and robust haptic interaction in terms of both objective and subjective measurements in a lossy network environment, and meanwhile perform well in haptic data reduction.
URI: https://repository.cihe.edu.hk/jspui/handle/cihe/996
DOI: 10.1109/ICoSP.2012.6491784
CIHE Affiliated Publication: Yes
Appears in Collections:CIS Publication

SFX Query Show full item record

Google ScholarTM

Check

Altmetric

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.