Markov-modulated Bernoulli-based Performance Analysis for BLUE Algorithm under Bursty and Correlated Traffics

In this study, the discrete-time performance of BLUE algorithms under bursty and correlated traffics is analyzed using two-state Markov-modulated Bernoulli arrival process (BLUE-MMBP-2). A two-dimensional discrete-time Markov chain is used to model the BLUE algorithm for two traffic classes, in which each dimension corresponds to a traffic class and the parameters of that traffic class. The MMBP is used to replace the conventional and widely-used Bernoulli process (BP) in evaluating and proposing analytical models based on the BLUE algorithm. The BP captures neither the traffic correlation nor the burstiness. The proposed approach is simulated, and the obtained results are compared with that of the BLUE-BP, which can modulate a single traffic class only. The comparison is performed in terms of mean queue length (mql), average queuing delay (D), throughput, packet loss, and dropping probability (DP). The results show that during congestion, particularly heavy congestion under bursty and correlated traffics, the BLUE-MMBP-2 algorithm provides better mql, D, and DP than the BLUE-BP.