Andi Yusril Cibu
Politeknik Wahdah Islamiyah Makassar, Makassar, Indonesia

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

MAC Address Randomisation and Its Implications for Service Management in Multi-Access Point Community Networks: Evolusi Sistem Terdistribusi: Perkembangan Arsitektur, Teknologi, dan Tantangan Komputasi Modern Ayu Angreini; Syahril Syam; Irhamullah Irhamullah; Iskandar Kato; Jumriati Jumriati; Andi Yusril Cibu
WITECH: Jurnal Teknologi Rekayasa Komputer dan Jaringan Vol. 2 No. 1 (2026): WITECH: Jurnal Teknologi Rekayasa Komputer dan Jaringan
Publisher : Program Studi Teknologi Rekayasa Komputer dan Jaringan, Politeknik Wahdah Islamiyah Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Community networks built on centralised hotspot control commonly rely on the MAC address as a device marker to sustain authentication continuity. That assumption of address stability is now challenged by MAC address randomisation, a privacy feature enabled by default in mobile operating systems. This study measures the level of MAC address randomisation observed in authenticated sessions, analyses its persistence characteristics, and evaluates the implications for the choice of identification basis in service management architecture. A descriptive quantitative design was applied to ex post facto operational data from a small paid community network serving 31 service identities through approximately 30 access points. A total of 46 service activation records covering 26 July to 22 August 2026 were extracted from a configuration export of the controlling device. Randomisation status was classified using the locally administered bit test. Results show that 80.4 per cent of MAC addresses were randomised, with a 95 per cent confidence interval of 69.0 to 91.9 per cent, and the proportion remained stable across observation periods. Five addresses recurred, four of them randomised yet unchanged for three to eleven days, indicating that randomisation persists per network. This characteristic causes device-based continuity to function within the same network name but fail across different network names, whereas speed class and validity policies remain consistent because they are bound to the service identity. These findings provide empirical support for placing service identity at the centre of network service management.