PUF-Based IoT Security: Building Lightweight Authentication for WiFi Networks

IoT Security Research
2022 9 Min Read

Undergraduate Minor Project: This research was conducted during my final year of undergraduate studies. The project focused on designing a lightweight authentication protocol using Physically Unclonable Functions (PUFs) to secure IoT devices connecting to WiFi networks—addressing the critical gap between security requirements and resource constraints.

The Internet of Things is exploding—over 12 billion devices connected today, with projections reaching 30+ billion by 2030. But here's the problem: most IoT devices are severely resource-constrained, making traditional cryptographic authentication impractical.

8P + 4H
Computational Cost
6n bits
Storage Required
3 CRPs
Challenge-Response Pairs

The IoT Security Problem

IoT devices face a fundamental security paradox: they need robust protection but lack the computational power, memory, and energy budgets for traditional security measures. This creates multiple attack vectors:

⚠️ MAC Address Spoofing

Attackers impersonate legitimate devices by mimicking MAC addresses, bypassing security filters

⚠️ Rogue Access Points

Malicious APs mimic legitimate networks to intercept and manipulate communications

⚠️ Physical Tampering

Attackers extract stored credentials from device memory through physical manipulation

⚠️ Replay Attacks

Captured authentication packets are replayed to gain unauthorized network access

Traditional WiFi security (WEP, WPA2) relies on shared passwords and doesn't adequately protect the physical layer. Enter Physically Unclonable Functions (PUFs)—a hardware-based security primitive that leverages microscopic manufacturing variations to create unique, unclonable device identities.

What is a PUF? Think of it as a hardware fingerprint. Just as no two human fingerprints are identical, no two PUFs produce the same response to a given challenge—even if manufactured identically. This makes them ideal for device authentication without storing sensitive keys.

Protocol Design: Three-Phase Authentication

Our protocol consists of three distinct phases, each serving a specific security function while minimizing computational overhead.

1

One-Time Enrollment Phase

Performed once per device-router pair in a secure environment.

  1. Generate Challenges: Create three random numbers C₁, C₂, C₃ (bit-stream length = n)
  2. Acquire Responses: Query PUF: R₁ = P(C₁), R₂ = P(C₂), R₃ = P(C₃)
  3. Store Securely: Save (C₁,R₁), (C₂,R₂), (C₃,R₃) in router database, indexed by device MAC address
// Enrollment Phase (Simplified)
for i = 1 to 3:
    C[i] = Random(n_bits)
    R[i] = PUF(C[i])
    Store(MAC_device, C[i], R[i])
2

Initialization Phase

Device initiates connection request to router.

  1. Connection Request: Device (Λ) sends MAC address (M∧) to router (π)
  2. Database Lookup: Router checks for matching MAC entry
  3. Proceed or Terminate: If match found, continue to authentication; otherwise, reject
3

Authentication Phase

Mutual authentication using challenge-response transformations.

Router → Device

  • • Generate random nonce
  • • Transform challenges: C′ = nonce ⊕ C
  • • Transform responses: R′ = nonce ⊕ R
  • • Compute hash: H∧ = Hash(C‖R‖nonce)
  • • Send: C′₁, C′₂, C₃, R′₁, R′₃, H∧

Device → Router

  • • Recover nonce: nonce = R₃ ⊕ R′₃
  • • Recover original C, R values
  • • Verify hash matches H∧
  • • Generate new nonce′ = R₂ ⊕ nonce
  • • Send: R′₁, R′₂, R′₃, Hπ
// Authentication - Device Side (Simplified)
nonce = R3 XOR R_prime_3
C1 = nonce XOR C_prime_1
C2 = nonce XOR C_prime_2
R1 = nonce XOR R_prime_1

// Verify router authenticity
H_verify = Hash(C1 || C2 || R1 || R3 || nonce)
if H_verify == H_received:
    router_authenticated = true
    
// Generate device authentication
nonce_prime = R2 XOR nonce
R_new_1 = PUF(C1 XOR nonce_prime)
R_prime_1 = R_new_1 XOR nonce_prime
H_pi = Hash(R_new_1 || R_new_2 || R_new_3 || nonce_prime)

Security Analysis: Why This Works

The protocol's security rests on several cryptographic principles, each addressing specific attack vectors identified in IoT WiFi networks.

XOR Operation Security

XOR is used extensively for challenge-response transformation. For security, we ensure:

Cryptanalysis Result: Even if an attacker intercepts all transmitted values (C′₁, C′₂, R′₁, R′₃, H∧), XOR combinations reveal no information about original challenge-response pairs or the nonce. The one-way hash function further protects authentication tokens.

Attack Mitigation

Attack Type Traditional WiFi Our PUF Protocol
MAC Spoofing ✗ Vulnerable ✓ Protected (PUF uniqueness)
Replay Attacks ✗ Vulnerable ✓ Protected (fresh nonce each session)
Physical Tampering ✗ Vulnerable ✓ Protected (PUF response changes)
Brute Force ⚠️ Possible ✓ Impractical (2ⁿ nonce space)
Credential Storage ✗ Stored in memory ✓ Never stored (computed dynamically)

Performance Evaluation

One of the primary design goals was minimizing computational and storage overhead for resource-constrained IoT devices.

Computational Cost

8P + 4H

8 PUF operations + 4 hash function calls per authentication session

Storage Requirement

6n bits

Only 3 challenge-response pairs stored (C₁,C₂,C₃,R₁,R₂,R₃)

Message Size

≤6n bits

Maximum transmitted data per authentication exchange

Nonce Space

2ⁿ

For n=64 or 128 bits, brute-force becomes computationally infeasible

Key Features & Innovations

1. Hardware-Based Uniqueness

Each device's PUF is physically unclonable—even the manufacturer cannot replicate it. This eliminates the need for secure key storage.

2. Dynamic Session Keys

Every authentication generates a new nonce, effectively creating a unique "password" for each connection attempt.

3. Mutual Authentication

Both device and router authenticate each other, preventing rogue access point attacks.

4. Tamper Evidence

Physical tampering alters the PUF's response pattern, immediately invalidating authentication capability.

Challenges & Limitations

No security solution is perfect. During this research, we identified several practical challenges:

⚠️ Environmental Variations

PUF behavior can be affected by temperature and voltage fluctuations. This requires careful calibration or error-correction mechanisms for reliable operation.

⚠️ Aging Effects

Over time, physical characteristics of the PUF may drift, potentially affecting response consistency. Long-term stability testing is essential.

⚠️ Scalability Concerns

Routers must store CRPs for each enrolled device. In large IoT deployments, this creates database management challenges.

Comparison with Existing Work

Our protocol builds upon and improves existing PUF-based authentication approaches:

Protocol CRPs Required Computational Cost WiFi-Specific
Chatterjee et al. (2017) High (scalability issue) ECC + Identity-based encryption No
Aman et al. (2016) Multiple Multiple hash functions + MAC No
Mahalat et al. (2018) 3 sets XOR + Hash Yes
Our Protocol 3 CRPs (optimized) 8P + 4H (lightweight) Yes (WiFi-focused)

Lessons Learned

1. Security Must Match Constraints

Designing for IoT means accepting trade-offs. We prioritized lightweight operations without sacrificing core security guarantees.

2. Hardware Security Complements Software

PUFs demonstrate that physical-layer security can address vulnerabilities that software-only solutions cannot touch.

3. Research Requires Iteration

Multiple protocol versions were developed before arriving at the final design. Each iteration addressed weaknesses discovered during security analysis.

4. Documentation is Critical

Writing the research paper forced clarity in thinking. Every security claim needed cryptographic justification.

Real-World Applications

🏭 Industrial IoT (IIoT)

Secure sensor networks in manufacturing environments where physical tampering is a concern

🏥 Healthcare Devices

Medical IoT devices requiring HIPAA-compliant authentication without storing credentials

🏠 Smart Home

Consumer IoT devices that need secure WiFi connectivity without complex user configuration

🚗 Automotive IoT

Vehicle-to-infrastructure communication requiring tamper-resistant authentication

Physically Unclonable Functions
IoT Security
WiFi Authentication
Lightweight Cryptography
Challenge-Response Protocol
Hardware Security

📚 Academic Context

This research was completed as my undergraduate minor project and published as a research paper. The work demonstrates foundational skills in cryptographic protocol design, security analysis, and academic writing—skills that continue to inform my approach to cloud security and detection engineering.