Network Simulation and Traffic Data Generation
Budget: $30 – $250 USD
I am looking for a freelancer to design and build a controlled virtual lab environment where network traffic can be generated and captured.
The main goal is to:
* Build a virtual lab with multiple machines
* Generate both normal and simulated network traffic
* Capture all traffic using an Ubuntu server
* Export the captured traffic as PCAP files
* Convert the captured data into a structured CSV dataset
The Ubuntu server will act as the monitoring point, where all generated traffic will be recorded. The final output required from this project is a clean and properly captured PCAP dataset along with a corresponding CSV dataset.
---
### 1. Introduction & Objectives
Provide a clear introduction to:
* AI in cybersecurity
* Importance of network traffic datasets
Define specific project objectives, such as:
* Generating realistic network traffic
* Creating labeled attack vs normal data
* Producing a clean dataset suitable for ML training
---
### 2. Lab Design & Network Topology
Design a fully isolated virtual lab environment using tools such as:
* VirtualBox / VMware / Proxmox
Required Virtual Machines:
* Kali Linux for attack simulation
* Windows Server as the target machine
* Windows Client for normal user traffic
* Ubuntu Server for traffic capture and monitoring
Requirements:
* Clear network topology diagram
* IP addressing scheme
* Justification of design choices
* Proper network isolation (host-only or internal network)
---
### 3. Network Traffic Generation
You must generate two types of traffic:
Normal Traffic:
* Web browsing simulation
* File transfers (FTP/SMB)
* SSH or standard user behavior
Attack Traffic:
Simulate realistic attacks such as:
* SYN Flood (e.g., hping3)
* UDP Flood
* HTTP Flood (e.g., GoldenEye or similar tools)
---
### 4. Traffic Capture & PCAP Generation
Use tools such as:
* Wireshark
* tcpdump
Capture all traffic from the Ubuntu monitoring server and export the captured data as .pcap files.
Requirements:
* Traffic must include both normal behavior and attack scenarios
* Data must be clean, structured, and free from unnecessary noise where possible
---
### 5. Data Labeling & Documentation
Clearly label all traffic as:
* Normal
* Attack (with specific type)
Provide documentation explaining:
* When attacks occur
* Which machine generated them
* Duration of each activity
---
### 6. Dataset Preparation (Final Stage of This Project)
Convert the PCAP data into a structured CSV dataset and organize it to ensure:
* Clarity
* Consistency
* Proper labeling
The dataset must contain at least 100,000 rows to ensure sufficient volume and diversity of traffic.
---
### Note:
This dataset will later be used for machine learning purposes, so quality is critical.
---
### Deliverables
Project Report (PDF/Word) including:
* Introduction and objectives
* Lab design and topology
* Traffic generation methodology
* Tools used
* Data labeling explanation
PCAP Files:
* Captured traffic including both normal and attack scenarios
CSV Dataset:
* Extracted from the PCAP file
* Properly structured with rows and columns
* Clean and ready for further processing
* Minimum size of 100,000 rows
Dataset Documentation:
* Clear explanation of dataset structure
* Label definitions
* Explanation of how the CSV dataset was generated
The main goal is to:
* Build a virtual lab with multiple machines
* Generate both normal and simulated network traffic
* Capture all traffic using an Ubuntu server
* Export the captured traffic as PCAP files
* Convert the captured data into a structured CSV dataset
The Ubuntu server will act as the monitoring point, where all generated traffic will be recorded. The final output required from this project is a clean and properly captured PCAP dataset along with a corresponding CSV dataset.
---
### 1. Introduction & Objectives
Provide a clear introduction to:
* AI in cybersecurity
* Importance of network traffic datasets
Define specific project objectives, such as:
* Generating realistic network traffic
* Creating labeled attack vs normal data
* Producing a clean dataset suitable for ML training
---
### 2. Lab Design & Network Topology
Design a fully isolated virtual lab environment using tools such as:
* VirtualBox / VMware / Proxmox
Required Virtual Machines:
* Kali Linux for attack simulation
* Windows Server as the target machine
* Windows Client for normal user traffic
* Ubuntu Server for traffic capture and monitoring
Requirements:
* Clear network topology diagram
* IP addressing scheme
* Justification of design choices
* Proper network isolation (host-only or internal network)
---
### 3. Network Traffic Generation
You must generate two types of traffic:
Normal Traffic:
* Web browsing simulation
* File transfers (FTP/SMB)
* SSH or standard user behavior
Attack Traffic:
Simulate realistic attacks such as:
* SYN Flood (e.g., hping3)
* UDP Flood
* HTTP Flood (e.g., GoldenEye or similar tools)
---
### 4. Traffic Capture & PCAP Generation
Use tools such as:
* Wireshark
* tcpdump
Capture all traffic from the Ubuntu monitoring server and export the captured data as .pcap files.
Requirements:
* Traffic must include both normal behavior and attack scenarios
* Data must be clean, structured, and free from unnecessary noise where possible
---
### 5. Data Labeling & Documentation
Clearly label all traffic as:
* Normal
* Attack (with specific type)
Provide documentation explaining:
* When attacks occur
* Which machine generated them
* Duration of each activity
---
### 6. Dataset Preparation (Final Stage of This Project)
Convert the PCAP data into a structured CSV dataset and organize it to ensure:
* Clarity
* Consistency
* Proper labeling
The dataset must contain at least 100,000 rows to ensure sufficient volume and diversity of traffic.
---
### Note:
This dataset will later be used for machine learning purposes, so quality is critical.
---
### Deliverables
Project Report (PDF/Word) including:
* Introduction and objectives
* Lab design and topology
* Traffic generation methodology
* Tools used
* Data labeling explanation
PCAP Files:
* Captured traffic including both normal and attack scenarios
CSV Dataset:
* Extracted from the PCAP file
* Properly structured with rows and columns
* Clean and ready for further processing
* Minimum size of 100,000 rows
Dataset Documentation:
* Clear explanation of dataset structure
* Label definitions
* Explanation of how the CSV dataset was generated