May 29, 2023: Privacy-Preserving Data Aggregation in IoT Networks using Secure Multi-Party Computation

MAster assignment

Privacy-Preserving Data Aggregation in IoT Networks using Secure Multi-Party Computation

TYPE : MASTER CS

Period: Start date: as soon as possible

Student: Unassigned

If you are interested please contact:

dr.ing. M. Elhajj (Mohammed) Lecturer

+31534893475

 m.elhajj@utwente.nl

Building: Zilverling 2031

Personal page

Abstract:

Data aggregation is a fundamental operation in IoT networks, but it raises concerns about privacy and data confidentiality. This thesis aims to explore the application of secure multi-party computation (MPC) techniques for privacy-preserving data aggregation in IoT networks. The proposal involves designing a secure data aggregation framework using MPC protocols, evaluating its privacy guarantees, and assessing the performance overhead in terms of computational complexity and communication overhead.

Literature Review:

1. Conduct a comprehensive review of existing literature on secure multi-party computation (MPC) protocols for privacy-preserving data aggregation in IoT networks.
2. Explore different MPC techniques, such as garbled circuits, secret sharing, and homomorphic encryption, and their suitability for IoT data aggregation.
3. Identify the security and privacy properties, scalability, and efficiency considerations of MPC protocols.

System Design:

1. Design a secure data aggregation framework based on MPC protocols for IoT networks.
2. Specify the data aggregation operations, such as sum, average, or maximum, that will be performed using MPC.
3. Define the roles of IoT devices and the central aggregator in the secure data aggregation process.

MPC Protocol Implementation:

1. Implement the selected MPC protocols for secure data aggregation.
2. Consider the specific requirements and constraints of IoT devices, such as limited computational power and memory resources.
3. Optimize the implementation to minimize the computational and communication overhead.

Privacy Analysis:

1. Evaluate the privacy guarantees provided by the implemented MPC-based data aggregation framework.
2. Assess the level of data privacy achieved by preventing individual device data from being exposed during the aggregation process.
3. Analyze the resilience of the framework against various privacy attacks, such as inference attacks or information leakage attacks.

Performance Evaluation:

1. Measure the computational complexity and communication overhead of the MPC-based data aggregation framework.
2. Evaluate the impact of various factors, such as the number of participating devices, data volume, or network conditions, on the performance.
3. Compare the performance of the framework with traditional data aggregation techniques in terms of efficiency and resource consumption.

Practical Evaluation:

1. Deploy the implemented MPC-based data aggregation framework in a real-world IoT network or a simulation environment.
2. Perform experiments to assess the practicality and scalability of the framework in large-scale IoT deployments.
3. Measure the overall performance, resource utilization, and communication efficiency of the framework under realistic scenarios.

Discussion and Future Directions:

1. Analyze the findings from the privacy analysis, performance evaluation, and practical experiments.
2. Discuss the strengths, limitations, and potential use cases of the MPC-based data aggregation framework.
3. Identify areas for further research and improvement, such as optimization techniques or integration with other privacy-preserving mechanisms.

Expected Outcome:

The expected outcome of this research is a secure multi-party computation (MPC)-based framework for privacy-preserving data aggregation in IoT networks. The thesis will provide insights into the privacy and performance aspects of using MPC protocols for IoT data aggregation. The findings will contribute to the development of practical and efficient privacy-preserving solutions for data aggregation in IoT applications.

References:

1. Clifton, C., Kantarcioglu, M., Malin, B., & Wang, W. (2012). Privacy-preserving Data Integration and Sharing. IEEE Security & Privacy, 10(2), 45-52. DOI: 10.1109/MSP.2012.26
2. Gascón, H., Gentry, C., Halevi, S., & Raykova, M. (2018). Secure Outsourcing of Large-Scale Computation. Proceedings of the IEEE, 106(11), 1836-1848. DOI: 10.1109/JPROC.2018.2860844
3. Abadi, M., et al. (2016). Deep Learning with Differential Privacy. Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security, 308-318. DOI: 10.1145/2976749.2978318