Research

The technological advance in communication and the reduction of computational devices are providing the development of the Internet of Things (IoT). This network integrates from lamps, fridges, clothes to computational devices, promoting an interaction between objects and humans beings in environments such as home and industries environments, and health.The integration of communication between objects, computing devices, various services can be provided in real time such as the monitoring of vital functions, the location of objects, temperature measurement. However, IoT network to deal with factors such as resource constraints, heterogeneity of devices, link losses and the vulnerability of the wireless medium. This factors when exploited by malicious users make the dissemination insecure. The content dissemination service in these networks is subject to various malicious actions, among which stands out the personification of identities held by Sybil attack. While there are several approaches to Sybil attack detection, such as Lightweight Sybil Detection (LSD), they are costly, disregard heterogeneous devices, and do not take into account Sybil attackers with stolen identities.

This work presents a associations control mechanism, called SACI, which prevents attackers Sybil associations to spread content of IoT. This mechanism is based on elliptic curves cryptography (ECC), unclonable functions (PUF) and identities receipts. ECC provides the creation of a secure channel with low computational cost where the PUF device will be transported to the calculation and distribution of identity receipts, ensuring its legitimacy. The SACI was evaluated through simulations and compared with LSD through metrics of effectiveness and efficiency. The results will be present soon.

Accomplishments

  • Survey of weaknesses and vulnerabilities of IoT content dissemination
  • Definition of requirements for confidential content dissemination against Sybil attack
  • Evaluation of Sybil attack detection approaches in the IoT content dissemination
  • Specification of a mechanism that detects Sybil attacks taking into account heterogeneous devices, called SACI
  • Definition of metrics for evaluating SACI under Sybil attacks in IoT content dissemination
  • Evaluation of the SACI in smarthome and eHealth scenarios with Sybil attacks
  • Period

    08/2013 - 05/2016(finished)