A Modified Approach for Measuring TCP Performance in Wireless Adhoc Network

Abstract

The Transmission Control Protocol (TCP) was designed to provide reliable end-to-end delivery of data over unreliable networks. In practice, most TCP deployments have been carefully designed in the context of wired networks. Ignoring the properties of wireless Ad Hoc Networks can lead to TCP implementations with poor performance. In this paper, an overview of this issue and a detailed discussion of the major factors involved have been presented. In particular, it has been shown how TCP can be affected by mobility and lower layers protocols. To make routing protocols aware of lost data packets and ACKs and help reduce TCP timeouts for mobility induced losses, there are two mechanisms: early packet loss notification (EPLN) and best effort ACK delivery (BEAD). EPLN seeks to notify TCP senders about lost data packets. A new technique for improving TCP performance in an ad hoc network that uses a table driven type of routing protocol paying attention to short-duration link failure has been described. The effect of the collision of a data packet and an ACK packet is suppressed by Delayed ACK and resending the ACK packet preferentially has also been evaluated. Through simulation, it has been shown that the combination of these improvements can increase TCP throughput about 20%. TCP performance over a mobile wireless link in different realistic scenarios has been explored in the paper. A simple method to combine a physical layer modeling and a network simulation using the network simulator ns-2 is being used. The results of the propagation channel simulation and the results of the corresponding network simulation using TCP protocol have been presented. The proposed approach demonstrates that in order to improve TCP performance over wireless links, one need not only to tune the parameters of TCP but also to modify the TCP semantics.