Congestion avoidance and control
SIGCOMM '88 Symposium proceedings on Communications architectures and protocols
MSWIM '01 Proceedings of the 4th ACM international workshop on Modeling, analysis and simulation of wireless and mobile systems
Effectiveness of Loss Labeling in Improving TCP Performance in Wired/Wireless Networks
ICNP '02 Proceedings of the 10th IEEE International Conference on Network Protocols
TCP Westwood and Easy RED to Improve Fairness in High-Speed Networks
PIHSN '02 Proceedings of the 7th IFIP/IEEE International Workshop on Protocols for High Speed Networks
Discriminating Congestion Losses from Wireless Losses using Inter-Arrival Times at the Receiver
ASSET '99 Proceedings of the 1999 IEEE Symposium on Application - Specific Systems and Software Engineering and Technology
Efficiency/Friendliness Tradeoffs in TCP Westwood
ISCC '02 Proceedings of the Seventh International Symposium on Computers and Communications (ISCC'02)
Proceedings of the 2006 international conference on Wireless communications and mobile computing
Cross-Layer loss differentiation algorithms to improve TCP performance in WLANs
PWC'06 Proceedings of the 11th IFIP TC6 international conference on Personal Wireless Communications
TCP performance issues over wireless links
IEEE Communications Magazine
Hi-index | 0.00 |
The problem of TCP and all its existing variations within MANETs resides in its inability to distinguish between different data packet loss causes. Thus, TCP has not always the optimum behaviour in front of packet losses which might cause network performance degradation and resources waste. Multiple Loss Differentiation Algorithms (LDAs) have been designed to improve TCP performances. They had been optimized for data networks where only the last link is a wireless link. In these LDAs the most common packet losses that are handled are those due to wireless channel errors or congestion. We show in this paper that a third common packet loss cause, link failure, has to be handled in multi-hop wireless networks such as mobile ad hoc networks (MANETs). In order to handle these three packet loss causes, we propose TCP-WELCOME. This latter follows a two-step process: (i) first, distinguish between most common packet loss causes, and (ii) then, triggers the most appropriate packet loss recovery according to the identified loss cause. The performance evaluation shows that TCP-WELCOME optimizes both energy consumption and throughput. Also, TCP-WELCOME does not change the standard as its loss differentiation and recovery algorithms can operate with the already existing TCP variants.