Indoor crowd estimation scheme using the number of Wi-Fi probe requests under MAC address randomization

Yuki Furuya, Hiromu Asahina, Masashi Yoshida, Iwao Sasase

Research output: Contribution to journalArticlepeer-review

5 Citations (Scopus)


As smartphones have become widespread in the past decade, Wi-Fi signal-based crowd estimation schemes are receiving increased attention. These estimation schemes count the number of unique MAC addresses in Wi-Fi signals, hereafter called probe requests (PRs), instead of counting the number of people. However, these estimation schemes have low accuracy of crowd estimation under MAC address randomization that replaces a unique MAC address with various dummy MAC addresses. To solve this problem, in this paper, we propose an indoor crowd estimation scheme using the number of PRs under MAC address randomization. The main idea of the proposed scheme is to leverage the fact that the number of PRs per a unit of time changes in proportion to the number of smartphones. Since a smartphone tends to send a constant number of PRs per a unit of time, the proposed scheme can estimate the accurate number of smartphones. Various experiment results show that the proposed scheme reduces estimation error by at most 75% compared to the conventional Wi-Fi signal-based crowd estimation scheme in an indoor environment.

Original languageEnglish
Pages (from-to)1420-1426
Number of pages7
JournalIEICE Transactions on Information and Systems
Issue number9
Publication statusPublished - 2021


  • Automatic people counting
  • Building automation
  • Crowd estimation
  • MAC address randomization
  • Wi-Fi probe request

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering
  • Artificial Intelligence


Dive into the research topics of 'Indoor crowd estimation scheme using the number of Wi-Fi probe requests under MAC address randomization'. Together they form a unique fingerprint.

Cite this