Mohammad Ehdaie - Parsa Sharif Research Center
Nikolaos Alexiou - KTH
Panos Papadimitratos - KTH

Sybil attacks pose a serious threat for Wireless Sensor Networks (WSN) security. They can create problems in routing, voting schemes, decision making, distributed storage and sensor re-programming. In a Sybil attack, the attacker masquerades as multiple sensor identities that are actually controlled by one or a few existing attacker nodes. Sybil identities are fabricated out of stolen keys, obtained by captured benign nodes. Existing Sybil defensive mechanisms suffer from the restricted sensor network size, or cause excessive resource consumption for the sensor network. In this work we propose a Sybil node detection mechanism, based on Random Key Distribution (RKD) schemes that can cope with large network sizes and minimize the waste of resources. We explain the techniques each node can use in a network running q-composite RKD to detect Sybil identities and restrict their number. Our method requires no trust to other sensors, which is important to defend against the attack.

Random Key Predistribution, Sybil Attack, Wireless Sensor Networks