Comparative Analysis of Congestion Window Size and Throughput for BBR versus Elastic TCP

Abstract

In the current trend of technology evolution, Transmission Control Protocol (TCP) has gained wide popularity. In order to provide efficient data transfer, we need to develop and evaluate an efficient TCP mechanism for longdistance networks. The importance of the TCP congestion control mechanism has increased significantly to assure the Internet’s guaranteed quality of service (QoS) in the presence of wire and wireless networks. In our study, we have successfully implemented Elastic-TCP and Bottleneck Bandwidth and Round-trip propagation time (BBR) using NS-3. This innovative congestion control algorithm operates based on network delay, and it is independent of round-trip time (RTT). BBR is another recent TCP algorithm which is based on the maximum volume of network links that is determined by bandwidth-delay product (BDP). The primary goal of Elastic TCP is to enhance bandwidth utilization in high BandwidthDelay Product (BDP) networks. It outperforms TCP Vegas and Veno in terms of average throughput; however, it gives comparatively lower throughput than TCP BBR and CUBIC.