Cognitive Radio Network (CRN) Simulation with Jamming & Spectrum Sensing

Project Overview

This project is a high-fidelity Python simulation of a Cognitive Radio Network (CRN). It models the physical (PHY) and network layers to demonstrate core CRN concepts: Spectrum Sensing (Listen Before Talk), Primary User Priority, and Resilience against Jamming Attacks.

The system simulates a real-world wireless environment where "Secondary Users" (unlicensed) must opportunistically access the spectrum without interfering with "Primary Users" (licensed). It also includes a malicious "Jammer" node to test the network's robustness and security protocols.

Key Features

• Dynamic Spectrum Access: Secondary users employ Markov Chain-based Spectrum Sensing to detect channel occupancy before transmitting. If the noise floor exceeds the threshold, the node backs off.
• Primary User Protection: The Base Station enforces strict priority. If a Primary User (e.g., JAZZ, UFONE) transmits, active Secondary users are preempted (queued or disconnected) to ensure licensed quality of service.
• Physical Layer Simulation:
  • Modulation: Adaptive modulation schemes including QAM-16, QAM-64, and QAM-256 combined with OFDM (Orthogonal Frequency-Division Multiplexing) for realistic signal modeling.
  • Error Correction: Reed-Solomon (RS) coding (RSCodec-40) to detect and correct transmission errors.
  • Channel Modeling: Simulates AWGN (Additive White Gaussian Noise) and variable interference levels.
• Cybersecurity:
  • Encryption: AES-GCM (Galois/Counter Mode) provides authenticated encryption, ensuring both confidentiality and data integrity.
  • Key Derivation: PBKDF2 (Password-Based Key Derivation Function 2) with salting is used to derive strong cryptographic keys from passphrases.
• Jamming Simulation: A specialized "Jammer" node that floods the Base Station with high-power noise. This triggers a Bit-Flipping Attack on legitimate packets, simulating physical layer corruption and Denial of Service (DoS).
• Rich Visualization: Generates detailed PDF reports containing Constellation Diagrams (visualizing signal quality), OFDM I/Q Signals, and Power Sensing plots.

System Architecture

1. The Sender (Primary/Secondary)

The sender initiates communication by performing Spectrum Sensing.

• Sensing: It checks the environmental noise floor. If below the threshold, it proceeds.
• Processing: The message is encoded (Reed-Solomon), modulated (QAM/OFDM), and encrypted (AES-GCM).
• Transmission: The secure frame is sent to the Base Station.

2. The Base Station (Controller)

Acts as the central relay and policy enforcer.

• Jamming Check: Continually monitors for high-power interference from registered Jammers. If detected, it applies a Bit-Flipping corruption algorithm to passing frames.
• Priority Check: Validates if the source is a Primary User. If yes, it updates the "Protection Window". If a Secondary User attempts to transmit during this window, the packet is queued or dropped.
• Routing: Forwards valid, allowed frames to the destination Receiver.

3. The Receiver (Target)

The final destination for data.

• Preemption: Checks incoming connection requests. If a Primary User connects while a Secondary User is active, the Receiver automatically terminates the Secondary session.
• Decryption & Visualization: Attempts to decrypt the payload. If successful, it logs the message. If decryption fails (due to jamming), it logs the corruption. Finally, it generates visual plots of the signal constellation.

File Structure

File	Role	Description
base_station.py	The Core / Air Interface	Acts as the central hub. Enforces priority logic, manages routing, and simulates physical layer corruption when jamming is active.
Dynamic_Sender.py	Transmitter Node	Can act as a Primary or Secondary user. Implements "Listen Before Talk" sensing logic, adaptive modulation, and encryption.
Dynamic_Receiver.py	Target Node	Receives frames, handles decryption, and generates the PDF visualizations. Includes logic to auto-disconnect Secondary users if a Primary user connects.
Dynamic_Jammer.py	Attacker Node	Broadcasts high-power noise to the Base Station to disrupt communication and trigger packet corruption.

Installation & Requirements

Ensure you have Python 3.8+ installed. Install the required dependencies:

pip install numpy matplotlib pycryptodome reedsolo
## Usage Guide
To run the full simulation, open four separate terminal windows and execute the scripts in the following order:

1. Start the Base Station
This initializes the network coverage area and starts listening on the default port (50050).

Bash
python base_station.py

2. Start the Receiver
Register a receiver to listen for incoming messages.

Bash
python Dynamic_Receiver.py
# When prompted, Enter Receiver ID: R1

3. Start the Sender (Normal Operation)
You can login as a Secondary User (S1-S60) or a Primary User (JAZZ, UFONE, etc.).

Bash
python Dynamic_Sender.py
# Enter Sender ID: S1 (for Secondary) OR JAZZ (for Primary)
# Enter Recipient ID: R1
# Type a message to verify connection.

4. Start the Jammer (Attack Simulation)
To test network resilience, launch the jammer in a new terminal. This node will broadcast noise bursts.

Bash
python Dynamic_Jammer.py
# The script will automatically start flooding the Base Station.

---

## Test Scenarios

Scenario A: Spectrum Sensing (Listen Before Talk)
Start Dynamic_Sender.py as Secondary User (S1).
The system simulates background noise using a Markov Chain model.
Observation: If the simulated channel energy is high (BUSY), S1 will print [SENSING] Busy. Backing off... and refuse to transmit. This proves the node is respecting the spectrum availability.

Scenario B: Primary User Preemption
Establish a connection between Secondary User (S1) and R1. Send a few messages.
Start a second sender instance as Primary User (JAZZ) and connect to R1.
Observation:
The Receiver (R1) will detect the high-priority request.
It will print [PRIORITY] Primary User 'JAZZ' is preempting Secondary User 'S1'!.
S1 will be disconnected immediately, and JAZZ will take over the channel.

Scenario C: Jamming Attack & Physical Layer Corruption
Establish a stable connection between any sender and receiver.
Run Dynamic_Jammer.py.
Attempt to send a message from the Sender.
Observation:
Base Station: Detects interference and prints ! JAMMING ACTIVE ! Corrupting packet....
Receiver: Fails to verify the AES-GCM tag and prints [!] PACKET CORRUPTED/JAMMED.
Visualization: The generated PDF report will show the corrupted packet as a Red, Scattered Constellation Diagram, visually proving the destruction of the signal integrity.

---

## Results & Visualization
The system automatically generates PDF reports in the node_logs directory (e.g., R1_plots_TIMESTAMP.pdf). These reports are crucial for analyzing the PHY layer performance.

1. Clean Signal (Normal Operation)
When the channel is clear, the Constellation Diagram shows distinct, well-separated blue dots (symbols).

Visual: A clean grid (4x4 for QAM-16, 8x8 for QAM-64).
Meaning: High Signal-to-Noise Ratio (SNR) and successful demodulation.

2. Spectrum Sensing Plot
This plot visualizes the instantaneous power of the received signal against the noise threshold.

Visual: A blue waveform representing signal energy vs. a red dashed line representing the threshold.
Meaning: Peaks above the red line indicate active transmission or high noise.

3. Jammed Signal (Under Attack)
During a jamming attack, the Base Station injects random bit errors. The Receiver plots these corrupted packets in Red.

Visual: The constellation diagram becomes a chaotic "cloud" of red dots with no grid structure.
Label: The plot text box explicitly states Msg: "[JAMMED] CORRUPTED DATA...".
Meaning: The physical integrity of the signal was destroyed by the Jammer, preventing successful decoding.