An energy-efficient random verification protocol for the detection of node clone attacks in wireless sensor networks

Yuping Zhou, Zhenjie Huang, Juan Wang, Rufeng Huang, Dongmei YuEURASIP 2014 · DOI

Abstract

Wireless Sensor Networks (WSNs) are highly susceptible to node replication attacks due to their often unattended deployment in hostile environments. While several detection schemes exist, many fail because witness nodes are exhausted or compromised before replicas are identified. This paper introduces the Randomized Multicell Linear Propagation (RMC) protocol, a distributed detection mechanism that maps node location information to randomly selected cells and employs linear-multicasting for verification. Simulation results demonstrate that RMC achieves a 100% detection rate while significantly extending the overall network lifetime compared to existing state-of-the-art protocols.

Introduction

In a node clone attack, an adversary captures a legitimate node, extracts its cryptographic credentials, and injects multiple replicas into the network. These clones can then launch internal attacks, such as data falsification or routing disruption. Centralized detection methods, which rely on a single base station, suffer from single points of failure and rapid energy depletion of nodes near the station. Distributed methods offer better resilience but often struggle with either low detection rates or excessive communication overhead.

The RMC protocol addresses these challenges by combining geographic grid partitioning with randomized mapping and localized linear verification.

The RMC Protocol

The protocol operates in three primary stages to ensure robust and energy-efficient detection:

  1. Geographic Grid Establishment: The deployment area is divided into a grid of exclusive clusters called "cells."
  2. Randomized Mapping: When a node claims a location, that claim is mapped to $q$ randomly selected cells. Based on the Birthday Paradox, selecting a small number of random cells ($q \approx \sqrt{m}$, where $m$ is the total number of cells) makes the probability of conflicting claims (from a clone and an original node) landing in the same cell nearly certain.
  3. Linear Multicast Verification: Once a claim reaches a designated cell, it is propagated along horizontal and vertical lines within that cell.
Intersection of linear propagation paths for conflict detection.

As shown in the figure above, the intersection of these propagation paths allows nodes at the intersection point to detect conflicting location claims for the same node ID. Unlike previous methods that only checked nodes exactly on a line, RMC utilizes a "verification surface" including one-hop neighbors, ensuring that at least one node will detect the conflict and alert the base station.

Visualization of the verification surface and node distribution within a cell.

Key Insights and Design Principles

  • Randomization as a Shield: By randomly selecting cells for verification in every cycle, the protocol prevents "smart" adversaries from predicting and compromising witness nodes in advance.
  • Localized Efficiency: By confining the linear multicast to specific cells rather than the entire network, RMC avoids the massive communication overhead associated with global flooding protocols.
  • Load Balancing: Because witness nodes are selected randomly across the grid, the energy burden is distributed evenly, preventing the formation of energy "hotspots" that lead to premature node failure.

Performance Evaluation

The protocol was evaluated using OMNeT++ simulations in a 500m x 500m area with 1,000 to 10,000 nodes. Key results include:

  • Detection Success: RMC achieved a 100% detection rate, outperforming the Line-Selected Multicast (LSM) protocol by 24.4%.
  • Network Lifetime: After 300 rounds, 54% of nodes survived in the RMC-enabled network, compared to 46% for LSM and only 22% for the P-MPC protocol.
  • Communication Overhead: RMC maintains a superior balance between security and energy consumption, proving more efficient than global Randomized Multicast (RM) protocols.

Conclusion

The RMC protocol provides a high-security, energy-efficient solution for detecting node clones in WSNs. While currently optimized for uniform topologies, future work will focus on adapting the protocol for non-uniform distributions and mobile sensor networks where node coordinates change dynamically.

Summary updated

Abstract

It is easy for adversaries to mount node replication attacks due to the unattended nature of wireless sensor networks. In several replica node detection schemes, witness nodes fail to work before replicas are detected due to the lack of effective random verification. This paper presents a novel distributed detection protocol to counteract node replication attacks. Our scheme distributes node location information to multiple randomly selected cells and then linear-multicasts the information for verification from the localized cells. Simulation results show that the proposed protocol improves detection efficiency compared with various existing protocols and prolongs the lifetime of the overall network.
BibTeX
@inproceedings{zhou2014energy,
  author = {Yuping Zhou and Zhenjie Huang and Juan Wang and Rufeng Huang and Dongmei Yu},
  title = {{An energy-efficient random verification protocol for the detection of node clone attacks in wireless sensor networks}},
  booktitle = {EURASIP},
  year = {2014},
  doi = {10.1186/1687-1499-2014-163},
}
Figure