Building a controlled delay assured forwarding class in differentiated services networks
Proceedings of the 2006 SIGCOMM workshop on Internet network management
Congestion control in differentiated services networks by means of fuzzy logic
FSKD'05 Proceedings of the Second international conference on Fuzzy Systems and Knowledge Discovery - Volume Part II
| Hi-index | 0.00 |
This paper presents a new active queue managementscheme, Fuzzy Explicit Marking (FEM), implementedwithin the differentiated services (Diff-Serv) framework toprovide congestion control using a fuzzy logic controlapproach. Network congestion control remains a criticaland high priority issue. The rapid growth of the Internetand increased demand to use the Internet for time-sensitivevoice and video applications necessitate thedesign and utilization of effective congestion controlalgorithms, especially for new architectures, such as Diff-Serv.As a result, a number of researchers are nowlooking at alternative schemes to TCP congestion control.RED (Random Early Detection) and its variants are oneof these alternatives to provide quality of service (QoS) inTCP/IP Diff-Serv networks. The proposed fuzzy logicapproach for congestion control allows the use oflinguistic knowledge to capture the dynamics of nonlinearprobability marking functions and offer effectiveimplementation, use multiple inputs to capture the(dynamic) state of the network more accurately, enablefiner tuning for packet marking behaviors (eitherdropping a packet or setting its ECN -ExplicitCongestion Notification- bit) for aggregated flows, andthus provide better QoS to different types of datastreams, such as TCP/FTP traffic or TCP/Web-liketraffic, whilst maintaining high utilization.