TY - JOUR

T1 - An average queue-length-difference–based congestion detection algorithm in TCP/AQM network

AU - Zhu, Jin

AU - Luo, Tong

AU - Yang, Lin

AU - Xie, Wanqing

AU - Dullerud, Geir E.

N1 - Funding Information:
This work was supported in part by the National Natural Science Foundation of China under grants 61374073 and 61503356 and by the Anhui Provincial Natural Science Foundation under grant 1608085QF153.

PY - 2018/5

Y1 - 2018/5

N2 - Congestion detection in transmission control protocol/active queue management networks remains a challenging problem in which the choosing of congestion signal is one of the most important factors. Exponentially weighted moving average of queue length, the most widely used congestion signal, is facing difficulties in detecting incipient congestion and quantifying the optimal forgetting factor. Aiming at these 2 disadvantages, we propose an average queue-length-difference–based congestion detection algorithm where exponentially weighted moving average of queue-length difference is chosen as the congestion signal with the theoretical optimal forgetting factor deduced. First, by defining the queue-length difference as the state, the corresponding state equation is derived from the fluid model. Second, we prove that the inflow traffic in state equation is a discrete-time martingale, which can be transformed to a Wiener process according to the martingale representation theorem. Noticing that the observation of state will be coupled with noise because of the unstable transmission, the state estimation is then derived with the application of recursive least squares filter. The filter gain of state estimation, which is a function of the noise-signal ratio, corresponds to the optimal forgetting factor in average queue-length-difference–based congestion detection algorithm. Simulation results in NS-3 and MATLAB illustrate the effectiveness of the proposed algorithm.

AB - Congestion detection in transmission control protocol/active queue management networks remains a challenging problem in which the choosing of congestion signal is one of the most important factors. Exponentially weighted moving average of queue length, the most widely used congestion signal, is facing difficulties in detecting incipient congestion and quantifying the optimal forgetting factor. Aiming at these 2 disadvantages, we propose an average queue-length-difference–based congestion detection algorithm where exponentially weighted moving average of queue-length difference is chosen as the congestion signal with the theoretical optimal forgetting factor deduced. First, by defining the queue-length difference as the state, the corresponding state equation is derived from the fluid model. Second, we prove that the inflow traffic in state equation is a discrete-time martingale, which can be transformed to a Wiener process according to the martingale representation theorem. Noticing that the observation of state will be coupled with noise because of the unstable transmission, the state estimation is then derived with the application of recursive least squares filter. The filter gain of state estimation, which is a function of the noise-signal ratio, corresponds to the optimal forgetting factor in average queue-length-difference–based congestion detection algorithm. Simulation results in NS-3 and MATLAB illustrate the effectiveness of the proposed algorithm.

KW - average queue-length difference

KW - congestion detection

KW - discrete-time martingale

KW - optimal forgetting factor

KW - recursive least squares

UR - http://www.scopus.com/inward/record.url?scp=85042773008&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85042773008&partnerID=8YFLogxK

U2 - 10.1002/acs.2863

DO - 10.1002/acs.2863

M3 - Article

AN - SCOPUS:85042773008

VL - 32

SP - 742

EP - 752

JO - International Journal of Adaptive Control and Signal Processing

JF - International Journal of Adaptive Control and Signal Processing

SN - 0890-6327

IS - 5

ER -