Internet delay forecasting for correlated and uncorrelated scenarios

Summary Most Internet end‐to‐end communications rely on the transmission control protocol, and transmission control protocol performance depends upon the efficient detection of packet losses and delay estimations. Thus, several retransmission timeout prediction algorithms, which play an important role in the quality of service assurance, have been suggested in the literature. In this paper, we propose a novel forecasting algorithm, which, in contrast to those reported previously, is suitable for both the correlated and the uncorrelated delay scenarios. The proposed algorithm is compared with other methodologies by using standardized delay traces reported in the literature, and it is shown that the proposed algorithm obtains an improved performance with smaller errors than those of current predictors. In this paper, we propose a novel forecasting algorithm, which, in contrast to those reported previously, is suitable for both the correlated and the uncorrelated delay scenarios. The proposed algorithm is compared with other methodologies by using standardized delay traces reported in the literature, and it is shown that the proposed algorithm obtains an improved performance with smaller errors than those of current predictors.