loading page

Characterization of Probabilistic Structure of Network Traffic During COVID-19: A Study Based on MAWI Data
  • +1
  • * Karmeshu,
  • Anoushka Mittal,
  • Pranav Jain,
  • Shachi Sharma
* Karmeshu
University of Petroleum and Energy Studies

Corresponding Author:karmeshu@gmail.com

Author Profile
Anoushka Mittal
University College Dublin School of Computer Science and Informatics
Author Profile
Pranav Jain
ViaSat Inc
Author Profile
Shachi Sharma
South Asian University Department of Computer Science
Author Profile


The COVID-19 pandemic has greatly affected all aspects of human life including working of offices, businesses, industries, educational institutions etc. With more work load shifting online, changes in the network traffic are inevitable. The earlier investigations have generally focused on the qualitative aspects of network traffic data during COVID-19. In contrast, the paper presents a study based on MAWI data characterizing network traffic in terms of multimodal and unimodal probability distributions. It is found that a transition of multimodal Gaussian mixture model of byte and packet counts during normal period to that of unimodal Laplace distribution during COVID-19 period has emerged. Further it is observed that the probability distribution depicts the preponderance of small and large packets during normal period which changes to that of small sized packets during Covid-19 period. These findings are likely to be useful to the administrators to manage network during crisis periods.