TY - JOUR

T1 - Machine learning for measurement-based bandwidth estimation

AU - Khangura, Sukhpreet Kaur

AU - Fidler, Markus

AU - Rosenhahn, Bodo

PY - 2019/8/15

Y1 - 2019/8/15

N2 - The dispersion that arises when packets traverse a network carries information that can reveal relevant network characteristics. Using a fluid-flow model of a bottleneck link with first-in first-out multiplexing, accepted probing tools measure the packet dispersion to estimate the available bandwidth, i.e., the residual capacity that is left over by other traffic. Difficulties arise, however, if the dispersion is distorted compared to the model, e.g., by non-fluid traffic, multiple bottlenecks, clustering of packets due to interrupt coalescing, and inaccurate time-stamping in general. It is recognized that modeling these effects is cumbersome if not intractable. This motivates us to explore the use of machine learning in bandwidth estimation. We train a neural network using vectors of the packet dispersion that is characteristic of the available bandwidth. Our testing results reveal that even a shallow neural network identifies the available bandwidth with high precision. We also apply the neural network under a variety of notoriously difficult conditions that have not been included in the training, such as randomly generated networks with the multiple bottleneck links and heavy cross traffic burstiness. Compared to two state-of-the-art model-based techniques as well as a recent machine learning-based technique (Yin et al., 2016), our neural network approach shows improved performance. Further, our neural network can effectively control the estimation procedure in an iterative implementation. We also evaluate our method with other supervised machine learning techniques.

AB - The dispersion that arises when packets traverse a network carries information that can reveal relevant network characteristics. Using a fluid-flow model of a bottleneck link with first-in first-out multiplexing, accepted probing tools measure the packet dispersion to estimate the available bandwidth, i.e., the residual capacity that is left over by other traffic. Difficulties arise, however, if the dispersion is distorted compared to the model, e.g., by non-fluid traffic, multiple bottlenecks, clustering of packets due to interrupt coalescing, and inaccurate time-stamping in general. It is recognized that modeling these effects is cumbersome if not intractable. This motivates us to explore the use of machine learning in bandwidth estimation. We train a neural network using vectors of the packet dispersion that is characteristic of the available bandwidth. Our testing results reveal that even a shallow neural network identifies the available bandwidth with high precision. We also apply the neural network under a variety of notoriously difficult conditions that have not been included in the training, such as randomly generated networks with the multiple bottleneck links and heavy cross traffic burstiness. Compared to two state-of-the-art model-based techniques as well as a recent machine learning-based technique (Yin et al., 2016), our neural network approach shows improved performance. Further, our neural network can effectively control the estimation procedure in an iterative implementation. We also evaluate our method with other supervised machine learning techniques.

KW - Available bandwidth estimation

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=85065878754&partnerID=8YFLogxK

U2 - 10.1016/j.comcom.2019.05.005

DO - 10.1016/j.comcom.2019.05.005

M3 - Article

AN - SCOPUS:85065878754

VL - 144

SP - 18

EP - 30

JO - Computer communications

JF - Computer communications

SN - 0140-3664

ER -