Using an Average Consensus Algorithm to Securely Transmit Data Across Multiple Nodes

Author(s)

Jesse Han, GSSM

School Name

Governor's School for Science and Mathematics

Grade Level

12th Grade

Presentation Topic

Computer Science

Presentation Type

Mentored

Abstract

Average consensus is a concept essential to coordinating the flow of information between multiple computational machines. The process allows individual devices to act as a single system, working toward a shared goal. With the prevalence of decentralized computer systems growing, average consensus is the key to having each of the individual nodes, or machines, reach the same final conclusions and properly coordinate their actions. To accomplish this, the nodes must reveal confidential information about their personal states to the rest of the system, creating concerns about privacy. The proposed solution, based on previous research, utilizes homomorphic encryption and a random scaling multiplier in order to mask the node’s true state value. The method securely encrypts the node’s data while remaining time-efficient due to compression. To test the average consensus algorithm within a real-world setting, the project was concentrated on implementing the algorithm using a set of single-board computers. The produced code worked as intended, and indicated that the concept would function accurately in practice. With further development, the average consensus technology could be applied to systems such as swarm robotics, drones, and self-driving cars to control decentralized systems precisely and with complete privacy.

Location

Neville 206

Start Date

4-14-2018 9:15 AM

Presentation Format

Oral and Written

COinS
 
Apr 14th, 9:15 AM

Using an Average Consensus Algorithm to Securely Transmit Data Across Multiple Nodes

Neville 206

Average consensus is a concept essential to coordinating the flow of information between multiple computational machines. The process allows individual devices to act as a single system, working toward a shared goal. With the prevalence of decentralized computer systems growing, average consensus is the key to having each of the individual nodes, or machines, reach the same final conclusions and properly coordinate their actions. To accomplish this, the nodes must reveal confidential information about their personal states to the rest of the system, creating concerns about privacy. The proposed solution, based on previous research, utilizes homomorphic encryption and a random scaling multiplier in order to mask the node’s true state value. The method securely encrypts the node’s data while remaining time-efficient due to compression. To test the average consensus algorithm within a real-world setting, the project was concentrated on implementing the algorithm using a set of single-board computers. The produced code worked as intended, and indicated that the concept would function accurately in practice. With further development, the average consensus technology could be applied to systems such as swarm robotics, drones, and self-driving cars to control decentralized systems precisely and with complete privacy.