Hybrid Post-quantum Secure Wireless Healthcare Network Using Ml-kem, Ml-dsa, Blockchain Audit Logging and Ai-based Intrusion Detection

<title>Abstract</title> <p>Wireless Body Area Networks (WBANs) used for remote patient monitoring carry sensitive physiological data such as heart rate, SpO₂, blood pressure and body temperature over open wireless channels, which exposes the traffic to eavesdropping, replay, man-in-the-middle tampering, packet injection and flooding. A second, longer-term problem is equally serious. The public-key algorithms in almost all present WBAN systems, RSA and ECC, are known to be breakable by Shor's algorithm on a sufficiently large quantum computer, so encrypted records captured today could be decrypted once such hardware exists. In this paper a hybrid security framework is developed in which the classical public-key layer is fully replaced with the NIST post-quantum standards: ML-KEM (CRYSTALS-Kyber, FIPS 203) for key encapsulation and ML-DSA (CRYSTALS-Dilithium, FIPS 204) for digital signatures, together with AES-256-GCM for payload encryption and HMAC-SHA256 for per-packet integrity. A hash-chained blockchain ledger is maintained for tamper-evident audit logging, and a Random Forest based intrusion detection module examines the traffic in parallel for attacks passing the cryptographic checks. The framework was implemented with a Python TCP gateway, SQLite storage under role-based access control, a Streamlit clinician dashboard and a React/Node.js web client, and was tested against eight adversarial scenarios. The measured end-to-end latency was 147 ms at a throughput of 342 packets per second, the IDS obtained 87% accuracy over five attack classes, and blockchain chain integrity was verified at 100%. In addition, the same Random Forest pipeline was applied to the BLEWBAN dataset, where it classified twelve on-body sensor positions with 76.9% accuracy.</p>