Performance Analysis of BBR Congestion Control Protocol Based on NS3

With the increasing complexity of the network communication environment. Research on TCP congestion control algorithm is one of the effective ways to alleviate network congestion. Google published the bottleneck bandwidth and round-trip time (BBR) congestion control algorithm in 2016. Unlike many currently loss-based congestion control algorithms (like BIC), BBR uses an estimate of the available bottleneck link bandwidth and RTT to govern its pacing rate. BBR has a good performance on Google's internal network and gains widespread attention. In this paper, we use NS3 to establish different network communication models. By analyzing the simulation data, we find that BBR outperforms BIC on high-latency, high-bandwidth network; BBR could periodically reduce the pacing rate to alleviate the bufferfloat problem; and there is a fairness problem between BBR and other versions of TCP protocols. In addition, we confirm that BBR flows have RTT fairness problem: a BBR flow with longer RTT dominates a competing flow with shorter RTT. Finally, we analyze the adaptability between BBR and AQM algorithms.