Skip to content

LeahMachaka/gridwatch-csoc

Repository files navigation

GridWatch CSOC

IoT Cyber Security Operations and Incident Intelligence Platform

Author: Leah Machaka Programme: National Diploma: ICT, 3rd Year Location: Gauteng, South Africa GitHub: github.com/LeahMachaka Status: In Development — May 2025 Live Demo: gritwatch-csoc-2.onrender.com/


What Is GridWatch CSOC?

GridWatch CSOC is a student-built IoT Cyber Security Operations and Incident Intelligence Platform. It simulates a Dubai government campus network to demonstrate real-world IoT security monitoring, network segmentation, and an original security innovation called Passive Behavioural Anomaly Detection (PBAD).

The platform monitors 15 IoT and operational technology devices across three network security zones. When any device behaves differently from its established baseline, GridWatch CSOC raises a CSOC alert and generates a formal Incident Intelligence Report — the same type of document a security analyst would deliver to a CISO after a real incident.


The Problem This Solves

Most IoT security systems verify a device identity when it connects to the network. Once authenticated, the device is trusted. The problem is that threat actors commonly compromise a legitimate IoT device so it passes authentication, then use that device to do something it should never do — exfiltrate data, communicate with external servers, or move laterally across the network.

Zero Trust answers: who are you? PBAD answers: are you still behaving as expected after we trusted you?

This is the gap. GridWatch CSOC builds the next layer.


Innovation: Passive Behavioural Anomaly Detection (PBAD)

PBAD profiles the normal communication behaviour of every IoT device after it joins the network. Each device gets a baseline covering approved destination IPs, active hours, data transfer volumes, protocols used, and connection direction.

The CSOC dashboard calculates a real-time deviation score for each device from 0 to 100. When the score crosses 80, a CRITICAL alert fires. When it crosses 60, a HIGH alert fires. The dashboard logs the event and a formal Incident Intelligence Report is generated.

This directly extends the Dubai Electronic Security Centre 2025 Zero Trust Assessment Tool launched at GISEC Global, which secures device identity at authentication. PBAD secures device behaviour after authentication — the next unsolved problem.


Simulated Environment

Name: Al Noor Government Services Campus (Fictional, Dubai, UAE) Tool: Cisco Packet Tracer 9.0.0 Scenario: A fictional Dubai government entity campus with three security zones, monitored continuously by GridWatch CSOC.

Zone VLAN Subnet Router Interface Contents
Zone 1 10 192.168.10.0/24 GigabitEthernet0/0 IoT cameras, motion sensors, entry panel, environmental monitor
Zone 2 20 192.168.20.0/24 GigabitEthernet0/1 HVAC, power meter, lighting controller, fire suppression, UPS
Zone 3 30 192.168.30.0/24 GigabitEthernet0/2 CSOC workstation, app server, SIEM server, admin workstation
Core 99 192.168.99.0/24 Management only Switch management interfaces

Device Register

Hostname IP Address Zone Function
IoT-MS-01 192.168.10.11 Zone 1 Motion Sensor A
IoT-MS-02 192.168.10.12 Zone 1 Motion Sensor B
IoT-CAM-01 192.168.10.21 Zone 1 IP Camera 01
IoT-CAM-02 192.168.10.22 Zone 1 IP Camera 02
IoT-EP-01 192.168.10.31 Zone 1 Smart Entry Panel
IoT-ENV-01 192.168.10.41 Zone 1 Environmental Monitor
OT-HVAC-01 192.168.20.11 Zone 2 HVAC Controller
OT-PWR-01 192.168.20.21 Zone 2 Smart Power Meter
OT-LGT-01 192.168.20.31 Zone 2 Lighting Controller
OT-FSS-01 192.168.20.41 Zone 2 Fire Suppression Sensor
OT-UPS-01 192.168.20.51 Zone 2 UPS Monitor
SRV-CSOC-01 192.168.30.10 Zone 3 CSOC Monitoring Workstation
SRV-APP-01 192.168.30.20 Zone 3 Government Application Server
SRV-SIEM-01 192.168.30.30 Zone 3 SIEM and Syslog Server
WS-ADM-01 192.168.30.40 Zone 3 Admin Workstation

Network Security Design

The GridWatch CSOC topology enforces strict traffic rules using two Extended Access Control Lists applied inbound on the Zone 1 and Zone 2 router interfaces.

ACL Rule Action Reason
ZONE1_IN permit icmp echo-reply to Zone 3 PERMIT Allows ping replies from IoT devices back to CSOC
ZONE1_IN permit Zone 1 to 192.168.30.30 PERMIT IoT devices may send logs to SIEM server only
ZONE1_IN deny Zone 1 to Zone 3 DENY Blocks all other Zone 1 to Zone 3 traffic
ZONE1_IN deny Zone 1 to Zone 2 DENY Prevents IoT devices reaching OT layer
ZONE2_IN permit icmp echo-reply to Zone 3 PERMIT Allows ping replies from OT devices back to CSOC
ZONE2_IN permit Zone 2 to 192.168.30.30 PERMIT OT devices may send logs to SIEM server only
ZONE2_IN deny Zone 2 to Zone 3 DENY Blocks all other Zone 2 to Zone 3 traffic
ZONE2_IN deny Zone 2 to Zone 1 DENY Prevents OT devices reaching IoT layer

Zone 3 (CSOC and admin) has full access to all zones for monitoring purposes.


Verification Results

All tests performed from SRV-CSOC-01 in Cisco Packet Tracer 9.0.0. ACL match counts confirmed via show access-lists.

Test Destination Result
Zone 3 internal SRV-APP-01 (192.168.30.20) PASS
Zone 3 internal SRV-SIEM-01 (192.168.30.30) PASS
Zone 3 internal WS-ADM-01 (192.168.30.40) PASS
Zone 3 to Zone 1 IoT-MS-01 (192.168.10.11) PASS
Zone 3 to Zone 1 IoT-CAM-01 (192.168.10.21) PASS
Zone 3 to Zone 2 OT-HVAC-01 (192.168.20.11) PASS
Zone 3 to Zone 2 OT-PWR-01 (192.168.20.21) PASS
Zone 3 to Zone 2 OT-FSS-01 (192.168.20.41) PASS
Zone 3 to Zone 2 OT-UPS-01 (192.168.20.51) PASS

PBAD Device Profiles

Device Normal Behaviour Anomaly Trigger Severity
IP Camera 01/02 Streams to SIEM only, no external IPs Traffic to any external IP CRITICAL
HVAC Controller Internal communication only Any external IP contact CRITICAL
Fire Sensor Silent unless triggered, one heartbeat per minute Multiple triggers within 5 minutes CRITICAL
SIEM Server Receives syslog only, never initiates traffic Any outbound connection initiated CRITICAL
Smart Entry Panel Max 100 events per day, 07:00 to 20:00 Outside hours or above volume HIGH
Power Meter Within 20% of 7-day rolling average Spike above 20% without cause HIGH
UPS Monitor Battery above 80%, charge cycle every 72 hours Battery drops below 50% unexpectedly HIGH
CSOC Workstation Authenticated sessions only, 08:00 to 18:00 Login outside hours or 3 failed attempts HIGH
Motion Sensor A/B Max 30 triggers per hour, 06:00 to 22:00 Outside hours or above trigger rate MEDIUM
Lighting Controller Active 06:00 to 23:00, scheduled cycles only Commands outside defined schedule MEDIUM
Env. Monitor Temp 18 to 26 degrees, humidity 30 to 70% Reading outside range or 15-minute silence LOW

Sample Incident: IIR-2025-001

The docs folder contains a full Incident Intelligence Report for the following simulated scenario.

At 02:17:43 UTC on 15 May 2025, PBAD detected that IoT-CAM-01 (192.168.10.21) initiated an outbound HTTPS connection to an external IP (185.220.101.47) and transferred 47.3 MB of data in 26 seconds during scheduled inactivity. PBAD deviation score: 94 out of 100. Severity: CRITICAL. ACL rules contained the incident to Zone 1. No lateral movement to Zone 2 or Zone 3 was possible.


Repository Structure

gridwatch-csoc/
├── README.md
├── packet-tracer/
│   └── GridWatch_CSOC_Topology_LeahMachaka.pkt
├── dashboard/
│   ├── index.html
│   ├── style.css
│   └── dashboard.js
└── docs/
    ├── GridWatch_PacketTracer_BuildGuide.docx
    └── GridWatch_Incident_Intelligence_Report.docx

Technology Stack

Layer Technology
Network Simulation Cisco Packet Tracer 9.0.0
Routing Static inter-VLAN routing, one interface per zone
Security Extended ACLs (ZONE1_IN, ZONE2_IN) on Cisco 2911
Segmentation VLANs 10, 20, 30, 99 on Cisco 2960 switches
Dashboard HTML, CSS, JavaScript
Backend (planned) Python, Django
Innovation PBAD (Passive Behavioural Anomaly Detection)

DESC Compliance Alignment

This project was designed with reference to the Dubai Electronic Security Centre published frameworks.

  • DESC IoT Security Standard — mandatory controls for IoT device security in Dubai government entities
  • DESC Zero Trust Assessment Tool (GISEC 2025) — PBAD extends this into the post-authentication behavioural layer
  • DESC ISR Version 3 — incident response structure follows ISR protocol
  • DESC CSOC Monitoring Standard — continuous PBAD monitoring with real-time alerting
  • DESC ICS and OT Security Standard — Zone 2 OT devices profiled separately with appropriate baselines

Certifications

Certificate Provider Status
Introduction to IoT Cisco NetAcad Completed — May 2025
Networking Basics Cisco NetAcad In Progress
Introduction to Cybersecurity Cisco NetAcad In Progress

Contact

Leah Machaka National Diploma: ICT, 3rd Year Gauteng, South Africa GitHub: github.com/LeahMachaka


GridWatch CSOC — Securing What Zero Trust Cannot See PBAD is an original innovation developed by Leah Machaka as part of this portfolio project.

About

IoT Cyber Security Operations Platform with Passive Behavioural Anomaly Detection (PBAD). Built with Django. DESC aligned.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors