Abstract

Internet only has best effort service characteristic, does not provide a QoS (Quality of Service) mechanism and there are no traffic classification. Real-time applications such as videoconferencing, is the most prioritized service to get a QoS guaranteed in the network. So it is important to provide QoS guarantees for applications that are sensitive to delay and jitter. In this study, writer try to evaluate in detail the readiness of existing IP networks to support desktop videoconferencing services H.323 point-to-point calls for a session capacity that can be supported based on bandwidth and delay bounds that occur in the network for voice and video services. As well as evaluating several methods commonly used queuing disciplines, such as first-in-first-out (FIFO), priority queuing (PQ), weighted fair queuing (WFQ), as well as WFQ queuing algorithm based differentiated service code point (DSCP) with random early detection (RED) to get the best QoS performance values using OPNET IT Guru. Simulation results show the number of videoconferencing sessions that can be supported more determined by the bandwidth instead of end-to-end delay, which can support 63 calls. But when the final accuracy check performed by generating 63 calls simultaneously, there was a considerable mismatch between video and voice packets sent and received. Simulation success by generating 45 calls simultaneously. The generation of 45 calls simultaneously indicate good health network where there is no packet loss, and end-to-end delay around 5.8 ms and adequate utilization for routers, switches, and links. WFQ_DSCP method provide the best value, especially in the number of packets dropped, and ratio of packet sent and received against time among the other methods.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call