Classification of Delay-based TCP Algorithms From Passive Traffic Measurements
Chapter, Peer reviewed, Conference object
MetadataShow full item record
Original versionHagos DH, Engelstad P.E., Yazidi A: Classification of Delay-based TCP Algorithms From Passive Traffic Measurements. In: Gkoulalas-Divanis A, Marchetti, Avresky DR. 2019 IEEE 18th International Symposium on Network Computing and Applications (NCA), 2019. IEEE https://dx.doi.org/10.1109/NCA.2019.8935063
Identifying the underlying TCP variant from passive measurements is important for several reasons, e.g., exploring security ramiﬁcations, trafﬁc engineering in the Internet, etc. In this paper, we are interested in investigating the delay characteristics of widely used TCP algorithms that exploit queueing delay as a congestion signal. Hence, we present an effective TCP variant identiﬁcation methodology from trafﬁc measured passively by analyzing β, the multiplicative back-off factor to decrease the cwnd on a loss event, and the queueing delay values. We address how the β as a function of queueing delay varies and how the TCP variants of delay-based congestion control algorithms can be predicted both from passively measured trafﬁc and real measurements over the Internet. We further employ a novel non-stationary time series approach from a stochastic nonparametric perspective using a two-sided Kolmogorov–Smirnov test to classify delay-based TCP algorithms based on the α, the rate at which a TCP sender’s side cwnd grows per window of acknowledged packets, parameter. Through extensive experiments on emulated and realistic scenarios, we demonstrate that the data-driven classiﬁcation techniques based on probabilistic models and Bayesian inference for optimal identiﬁcation of the underlying delay-based TCP congestion algorithms give promising results. We show that our method can also be applied equally well to loss-based TCP variants.