Physio-Key-Gen: ECG-Based Key Agreement Protocol

Project Description

Physio-Key-Gen is a Proof-of-Concept (PoC) implementation of a secure key agreement protocol for Wireless Body Area Networks (WBAN). It utilizes the Inter-Pulse Interval (IPI) derived from Electrocardiogram (ECG) signals as a shared source of randomness to generate cryptographic keys between two sensors.

The system employs Bloom Filters to efficiently reconcile the set of biometric features (IPIs) between a sender (Sensor S) and a receiver (Sensor R) without exposing the raw biometric data, ensuring both security and privacy.

Architecture

The following diagram illustrates the key agreement flow between Sensor S and Sensor R:

sequenceDiagram
    participant S as Sensor S (Sender)
    participant R as Sensor R (Receiver)
    
    Note over S,R: Both sensors capture ECG from the same host
    
    S->>S: Generate IPI Features (Fs)
    R->>R: Generate IPI Features (Fr)
    
    S->>S: Create Bloom Filter (BF) from Fs
    S->>R: Send (ID_S, BF, Random_Nonce)
    
    R->>R: Check Fr against BF
    R->>R: Identify Common Features
    R->>R: Generate Key (Kr) = SHA1(Common Features)
    R->>R: Generate HMAC1 = HMAC(Kr, Common Features + Nonce)
    
    R->>S: Send (ID_R, HMAC1, Indices_Common)
    
    S->>S: Reconstruct Common Features from Indices
    S->>S: Generate Key (Ks) = SHA1(Common Features)
    S->>S: Verify HMAC1 using Ks
    
    alt HMAC Verified
        S->>S: Key Agreement Successful (Ks == Kr)
        S->>R: Send Confirmation HMAC
    else HMAC Failed
        S->>S: Abort Protocol
    end

Loading

Features

• Biometric Feature Extraction: Extracts Inter-Pulse Intervals (IPI) from raw ECG data using wfdb.
• Bloom Filter Implementation: Custom Python implementation of a Bloom Filter for efficient set membership testing.
• Secure Key Generation: Uses SHA-1 hashing and HMAC for secure key derivation and verification.
• Simulation: Simulates the interaction between two sensors using the MIT-BIH Arrhythmia Database.

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/physio-key-gen.git
   cd physio-key-gen

2. Install the required dependencies:

   pip install numpy matplotlib wfdb bitarray

   Note: You may need to download the MIT-BIH Arrhythmia Database files if they are not included or use the wfdb library to fetch them.

Usage

To run the key agreement simulation:

python KeyAgreement.py

This script will:

1. Load an ECG record.
2. Generate features for Sensor S and Sensor R (with a simulated offset/noise).
3. Execute the key agreement protocol.
4. Print the status of the key generation and verification steps to the console.

Files Structure

• KeyAgreement.py: Main entry point. Orchestrates the protocol between S and R.
• BloomFilter.py: Class implementation of the Bloom Filter data structure.
• InterPulseInterval.py: Helper functions for ECG signal processing and IPI calculation.
• LICENSE: MIT License file.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.