network-reliability

C++20 research system for adaptive FEC, cross-layer control, and resilient heterogeneous networking under energy and link constraints.

Stoer-Wagner algorithm finding the global minimum cut in an undirected weighted graph without designating source and sink nodes in O(V^3).

Stoer-Wagner algorithm finding the global minimum cut in an undirected weighted graph without designating source and sink nodes in O(V^3).

🌐 Cross-platform network monitoring tool with GUI for 24/7 internet connectivity tracking. Real-time dashboard, automated diagnostics, ISP issue detection with MTR logs. Works on Raspberry Pi, home servers, and homelabs. Python 3.8+, Tkinter GUI, optional standalone .exe/.app builds available.

Lightweight, reliable Dynamic DNS updater for Cloudflare. Automatically detects your public IP and updates DNS records with verification. Built in Python, container-ready (Docker), minimal footprint. Perfect for home servers, VPNs, self-hosting, and edge deployments.

Continuous network reliability monitor that watches when your ISP and property manager won't. Pings, DNS, STUN, and Wi-Fi — every outage timestamped for the conversation that follows.

Open-source AI network reliability diagnostics for ChatGPT, Codex, Claude Code and AI agents.

Distributed network reliability simulator with graph analysis, failure modeling, and resilience evaluation.

Reproducible candidate-link sentinel for safe OSPF reconfiguration under network-digital-twin drift, with Containerlab/FRRouting experiments and holdout validation.

C++20 packet/slot simulation of deadline scheduling under burst losses, clock holdover, retries and replication.

Model to detect missing and spurious interactions in complex networks

Reproducible disaster-response digital twin for Türkiye using AFAD/OSM data, graph optimization, coupled Monte Carlo simulation and bootstrap reliability-boundary inference. DOI: 10.5281/zenodo.21903850

Generalized reliability polynomials for quality-weighted network analysis. Every reliability library assumes binary survival — this one doesn't. Zero dependencies.

Network reliability lab for BGP/OSPF fault injection, traffic-impact measurement, and deterministic recovery verification.