본문 바로가기
  • Home

Performance Evaluation of the Error Recovery based Object Relation in SCTP

  • Journal of Knowledge Information Technology and Systems
  • Abbr : JKITS
  • 2018, 13(1), pp.139-146
  • DOI : 10.34163/jkits.2018.13.1.014
  • Publisher : Korea Knowledge Information Technology Society
  • Research Area : Interdisciplinary Studies > Interdisciplinary Research
  • Published : February 28, 2018

CHOI WON KEUN 1

1인하공업전문대학

Accredited

ABSTRACT

The main issue of designing multimedia services is the support and integration of a wide variety of applications. QoS(Quality of Service) provides a unifying theme on which the facilities and functions of the system standards can be constructed. In distributed system, QoS guarantees are fundamentally concerned with the end-to-end communication between applications. The reliability of the transmission is one of the important QoS parameters. Including SCTP(Stream Control Transport Protocol), conventional error detection and recovery methods, however, do not consider the integrated viewpoint of multimedia object data. In the first reference paper, we have designed and proposed the multimedia-oriented error control method based on multimedia object information called EROR(Error Recovery based Object Relation). In this paper, we have investigated its performance evaluation by transmission delay. The results show that the EROR method manifests superior performance as the error occurs more frequently. When the frame transmission succeeds, it can be found that the EROR method reveals shorter average frame delay than SRER(Selective Repeat Error Recovery). On the other hand, the performance evaluation on the both methods show closer results as the error rate decreases. In the case of the extremely low error rate, furthermore, the both methods show the almost same performance results. Therefore, it makes SCTP have the high performance and flexibility. It should have the reducing effect of the network resources as well as satisfy the user’s requirements.

Citation status

* References for papers published after 2023 are currently being built.