[Paper Review] An Efficient Fog-Assisted Unstable Sensor Detection Scheme with Privacy Preserved

Shuo Chen, Rongxing Lu+1 more|arXiv (Cornell University)|Nov 28, 2017

Security in Wireless Sensor Networks23 references3 citations

TL;DR

This paper proposes a fog-assisted, privacy-preserving scheme for detecting unstable sensors in IoT environments by leveraging fog computing to reduce communication and computation load on the control center. Using the Paillier cryptosystem for secure data aggregation and batch verification for efficient authentication, the scheme achieves high detection accuracy with low false positives while preserving data privacy and resisting common cyberattacks.

ABSTRACT

The Internet of Thing (IoT) has been a hot topic in both research community and industry. It is anticipated that in future IoT, an enormous number of sensors will collect the physical information every moment to enable the control center making better decisions to improve the quality of service (QoS). However, the sensors maybe faulty and thus generate inaccurate data which would compromise the decision making. To guarantee the QoS, the system should be able to detect faulty sensors so as to eliminate the damages of inaccurate data. Various faulty sensor detection mechanisms have been developed in the context of wireless sensor network (WSN). Some of them are only fit for WSN while the others would bring a communication burden to control center. To detect the faulty sensors for general IoT applications and save the communication resource at the same time, an efficient faulty sensor detection scheme is proposed in this paper. The proposed scheme takes advantage of fog computing to save the computation and communication resource of control center. To preserve the privacy of sensor data, the Paillier Cryptosystem is adopted in the fog computing. The batch verification technique is applied to achieve efficient authentication. The performance analyses are presented to demonstrate that the proposed detection scheme is able to conserve the communication resource of control center and achieve a high true positive ratio while maintaining an acceptable false positive ratio. The scheme could also withstand various security attacks and preserve data privacy.

Motivation & Objective

• To address the high communication and computational burden imposed on control centers by massive sensor data in IoT systems.
• To detect unstable sensors—particularly those with high noise variance—without relying on neighbor-based voting or centralized data analysis.
• To preserve sensor data privacy during aggregation and transmission using cryptographic techniques.
• To reduce network latency and bandwidth usage by performing fault detection locally at fog nodes.
• To ensure security against eavesdropping, man-in-the-middle, and replay attacks in the detection process.

Proposed method

• Fog computing architecture is employed with three tiers: environmental (sensors), edge (fog devices), and cloud (control center), enabling local processing.
• The Paillier cryptosystem is used to encrypt sensor data at the first-layer fog devices, ensuring privacy during aggregation.
• Fog devices perform local computation to compute dispersion measures (e.g., variance) on encrypted data to detect unstable sensors.
• Batch verification is applied to authenticate multiple sensor reports efficiently, reducing verification overhead.
• Conditional probability models are used to assess sensor correlation between parent and child fog nodes to detect anomalies.
• The scheme avoids transmitting raw plaintext data to the control center, minimizing network resource usage.

Experimental results

Research questions

• RQ1 How can faulty sensor detection be performed efficiently in large-scale IoT deployments without overloading the control center?
• RQ2 What techniques can preserve sensor data privacy during aggregation in fog-assisted systems?
• RQ3 Can a combination of homomorphic encryption and batch verification reduce communication and computation costs in sensor fault detection?
• RQ4 How effective is the scheme in detecting unstable sensors with high noise variance while maintaining low false positive rates?
• RQ5 What level of security can be achieved against common network attacks such as eavesdropping and replay attacks?

Key findings

• The proposed scheme significantly reduces communication and computation overhead at the control center by performing fault detection locally in the fog layer.
• The use of the Paillier cryptosystem ensures end-to-end privacy preservation during data aggregation and transmission.
• The scheme achieves a high true positive ratio for detecting unstable sensors with high noise variance, indicating strong detection effectiveness.
• The false positive ratio remains within an acceptable range, ensuring reliability in practical deployment.
• The scheme is resilient against eavesdropping, man-in-the-middle, and replay attacks due to cryptographic authentication and encryption.
• Performance evaluations confirm that the scheme outperforms traditional centralized approaches in terms of resource conservation and scalability.