Dynamic Privacy-Aware Routing (DyPAR) for Wireless Sensor Networks

Abstract: The rapid expansion of the Internet of Things (IoT) has enabled pervasive sensing, automation, and data-driven decision-making. However, privacy and security challenges remain critical in Wireless Sensor Networks (WSNs), where limited computational and energy resources render traditional routing protocols vulnerable to traffic analysis, identity spoofing, and data manipulation. Existing routing schemes emphasize performance or energy efficiency but lack adaptive, privacy-aware mechanisms capable of responding to dynamic threats.

This paper presents Dynamic Privacy-Aware Routing (DyPAR), an adaptive probabilistic routing protocol that balances privacy preservation, energy efficiency, and computational feasibility for large-scale IoT networks. DyPAR incorporates entropy-based relay selection, dynamic adjustment of forwarding probabilities, and context-aware weighting to reduce adversarial traceability while maintaining efficient routing. The protocol integrates lightweight privacy-preserving components, including Efficient Key Management (EfKM), Privacy-Aware Data Aggregation (PrADA), and an Adaptive Privacy Parameter Change Mechanism (A2PCM) for real-time adjustment based on network conditions and data sensitivity.

Extensive simulations across heterogeneous network sizes and attack models show that DyPAR achieves high privacy compliance, strong resilience against Sybil, eavesdropping, and data-tampering attacks, and improved packet delivery performance relative to established privacy-aware routing baselines. While DyPAR maintains low energy consumption in benign scenarios, computational and energy overhead increase under multi-vector adversarial conditions, highlighting the need for further optimization in ultra–resource-constrained environments.

Future work will explore (i) lightweight cryptographic integration to reduce energy cost, (ii) federated learning–based adaptive routing to enhance real-time privacy decisions, (iii) real world and energy-efficient clustering for large-scale deployments, and (iv) blockchain-enabled distributed trust frameworks to mitigate identity spoofing and coordinated attacks.

Overall, DyPAR offers a scalable, adaptive, and privacy-preserving routing solution for next-generation IoT systems, providing a strong foundation for secure and resilient sensor network communication.