valuasi Algoritma KNN dan SVM dalam Klasifikasi Zona Rogue Access Point Berbasis RSSI pada Data Simulasi Empat Access Point
Jurnal Informatika: Jurnal Pengembangan IT · 04.11.2026 · 11 (3) · 937-946
DOI: 10.30591/jpit.v11i3.11 ↗Kurz gefasst
Fachartikel in der Zeitschrift Jurnal Informatika: Jurnal Pengembangan IT, online erschienen am 04.11.2026. Verfasst von Rifqi Akmal Musyaffa' und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
RSSI-based detection of Rogue Access Points (RAPs) has been studied, but fully documented comparisons of KNN and SVM on multi-AP RSSI remain scarce. This study compares KNN and SVM for classifying proximity zones to a RAP (25 m radius) using RSSI from four access points on simulated data of 10,201 points (100 × 100 m area; log-distance model with uniform ±3 dB noise). The evaluation uses a 2×2 ablation (scaling and hyperparameter tuning), stratified random and spatial-block splits with 10 seeds, and McNemar and DeLong tests. On the seed-42 test set, KNN reached accuracy 98.04%, MCC 0.9198, and AUC 0.9979, while SVM reached 98.17%, 0.9257, and 0.9979; the difference was not significant (McNemar p = 0.424). Averaged over 10 seeds, MCC was 0.9143 ± 0.0055 (KNN) and 0.9096 ± 0.0094 (SVM) under random splits, and 0.8759 ± 0.0652 and 0.8897 ± 0.0485 under spatial-block splits, so neither model wins consistently. Scaling and tuning gave no significant improvement, and a linear single-feature model (rssi_ap4) already reached MCC 0.9130. All errors lay within 5 m of the zone boundary. These results are limited to simulated data with one topology and one RAP position and require validation on real networks.
Zitieren
APA
Musyaffa', R. A., Yuniarti, W. D., & Subowo, M. H. (2026). valuasi Algoritma KNN dan SVM dalam Klasifikasi Zona Rogue Access Point Berbasis RSSI pada Data Simulasi Empat Access Point. Jurnal Informatika: Jurnal Pengembangan IT, 11(3), 937-946. https://doi.org/10.30591/jpit.v11i3.11
Vancouver
Musyaffa' RA, Yuniarti WD, Subowo MH. valuasi Algoritma KNN dan SVM dalam Klasifikasi Zona Rogue Access Point Berbasis RSSI pada Data Simulasi Empat Access Point. Jurnal Informatika: Jurnal Pengembangan IT. 2026;11(3):937-946. doi:10.30591/jpit.v11i3.11
BibTeX
@article{Musyaffa2026,
author = {Musyaffa', Rifqi Akmal and Yuniarti, Wenty Dwi and Subowo, Moh Hadi},
title = {valuasi Algoritma KNN dan SVM dalam Klasifikasi Zona Rogue Access Point Berbasis RSSI pada Data Simulasi Empat Access Point},
journal = {Jurnal Informatika: Jurnal Pengembangan IT},
year = {2026},
volume = {11},
number = {3},
pages = {937--946},
doi = {10.30591/jpit.v11i3.11}
}
EUVIMED