Characterization of SIP Signaling-Messages Over OpenSIPS Running On Multicore Server
Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Over the course of last decade, the demand for VoIP (Voice over Internet Protocol) applications has increased significantly among enterprises and individuals due to its low cost. This increasing demand resulted in a significant increase in users who require reliable VoIP communication systems. QoS (Quality of Service) is a major issue in VoIP implementation and is a method to impel the development of real-time multimedia services like VoIP and videoconferencing. However, there are certain challenges in achieving QoS for VoIP application, which need special attentions; like latency and packet loss.
The VoIP servers which are functioning on single core software/hardware model have high latency and packet loss issues due to their limited processing bandwidth. A multicore software/hardware model is the solution to cope up with the increasing demands of VoIP and yet an active research area in telecommunication. Using a multicore software/hardware model for VoIP has several challenges, one of the challenges is to design and implement QoS Benchmarking module for VoIP client and server on multicore.
In this thesis the focus is on latency and packet loss of SIP messages on OpenSIPS server. This is done by performing stress testing for QoS benchmarking, where delay and call drop rate is calculated for SIP (Session Initiation Protocol) signaling messages on parallel VoIP client server model. The model is built in C for multicore and is used as a simulation tool. SIP is widely deployed protocol for call establishment, maintenance and termination in VoIP.
Place, publisher, year, edition, pages
2012. , 41 p.
Delay, Latency, OpenSIPS, Packet Loss, QoS, QoS Benchmarking, SIP Signaling Messages, Videoconferencing, VoIP
Engineering and Technology
IdentifiersURN: urn:nbn:se:kth:diva-121530OAI: oai:DiVA.org:kth-121530DiVA: diva2:618766
Master of Science - Internetworking
Hemani, Ahmed, Professor