Modeling TCP performance in reordering and lossy networks

Abstract

Modeling TCP performance is an important issue in computer networks and plays an important rule to improve network performance. There are two types of packet transmission errors in network, packet loss and delay. Different errors will induce to different influence on network performance. Most of the recent researches only consider modeling TCP in packet lost networks. This paper first studies the problem of modeling TCP throughput in packet delay networks. Firstly, we give a sufficient and necessary condition to identify whether a data packet flow is slightly out of order or in order. Based on this condition, the probability that a data packet flow is slightly out of order or in order is analyzed. Finally, TCP throughput from the obtained probability of data packet flow is slightly out of order or in order is evaluated. Numerical results indicate that the difference between the probability of data packet flow is slightly out of order or in order of our model and which in real networks is tiny, and compared with those existing models which do not consider packet delay, our model are more accurate in maximum congestion window prediction and TCP throughput prediction.