Group synchronization method with fast response time for VoD services

We propose a group synchronization method for VoD services based on syncing user PC internal clocks. A group synchronization video watching service enables users to simultaneously watch video along with communication functions. This service enables users to communicate with other group members onlin...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Moriyasu, S., Tajima, K., Ohshima, K., Terada, M.
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We propose a group synchronization method for VoD services based on syncing user PC internal clocks. A group synchronization video watching service enables users to simultaneously watch video along with communication functions. This service enables users to communicate with other group members online as if the group members were watching the video in the same room. A number of criteria must be satisfied in order to realize this service in VoD: 1) video synchronization accuracy (>; 120 ms), 2) fast response time of play control (≥ 100 ms), and 3) a group synchronization technique suitable for a VoD service (play control function). In order to satisfy these criteria, group members synchronize their internal clocks, and play is controlled by the synchronized internal clocks. Play control is considered for the case of an unstable Internet connection (e.g., delay or jitter) and low-processing-power PCs. We implement a prototype system in order to evaluate video synchronization accuracy and client response time. As a result, we realized the synchronization accuracy goal, except for the seek command. The seek command synchronization accuracy was not good. Thus, we proposed another seek command process to solve the problem of low-processing-power PCs. The results demonstrate that the response time of the low-latency client was good, which means that client classification according to latency was effective for user response time.
ISSN:1550-445X
2332-5658
DOI:10.1109/ICOIN.2012.6164374