Home▸AI & Machine Learning▸Network Traffic CUSUM Monitor: ARL Curves and the Alarm Statistic (2D)

Network Traffic CUSUM Monitor: ARL Curves and the Alarm Statistic (2D)

The same CUSUM change-detection model in 2D: the live traffic scope with the alarm statistic, plus log ARL curves against the threshold and the detection delay against the allowance k, with the numerical checks of the shared model.

AI & Machine Learning2DModerate60 FPS📱 Mobile-adapted🟠 Blender⇄ 3D version
Network Traffic CUSUM Monitor: ARL Curves and the Alarm Statistic (2D) ↗ Open standalone

A defensive, educational intrusion monitor: a normalised traffic statistic is fed to a one-sided CUSUM that fills like a tank and raises a beacon at threshold h. Change the attack shift, the allowance k and the threshold to see the exact average run length between false alarms (ARL0) and the detection delay (ARL1) from the Brook-Evans Markov chain, compared with Siegmund's approximation. A Blender-built scope, tank and alarm beacon with numerical checks built in.

⚙ Under the hood

The same CUSUM change-detection model in 2D: the live traffic scope with the alarm statistic, plus log ARL curves against the threshold and the detection delay against the allowance k, with the numerical checks of the shared model.

Intrusion DetectionCUSUMAverage Run LengthChange DetectionDefensive Security

2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install

What did you find?

Add reproduction steps (optional)