Open Collections

Blockchain for decentralized trusted communication networks

University of British Columbia

Future communication networks are expected to connect a large number of heterogeneous devices. For example, the fifth-generation infrastructure public-private partnership (5G PPP) targets to connect over 7 trillion devices. This makes trust management among devices challenging in trust-based applications due to their centralized architecture. Blockchain is promising to provide decentralized trust management. However, to use the blockchain in communication networks, we need to address delay and chain inconsistency challenges. In this thesis, we first focus on routing protocols as one of the main network functions. We propose a blockchain-based routing protocol (BCR) to mitigate the challenges raised by centralized network architecture. BCR uses blockchain smart contracts to transfer the routing protocol's control messages. The results show that BCR can reduce routing overhead by about $5$ times compared to the AODV routing protocol at the cost of a slightly lower packet delivery ratio. However, its performance depends on the time required to mine protocol transactions inside the blockchain. To study this issue, we propose a blockchain mining scheme, Wait-Min(D), in which miners wait for a certain number of transactions to arrive in the mining pool to minimize the average transaction waiting time. Numerical results show that the average waiting time can be reduced by about 10% compared to the existing mining schemes. This improvement is significant for time-critical network applications. To address the requirements of different network applications, we propose a blockchain mining framework, namely MCBS, that provides multi-class network services in a decentralized manner. We model the mining framework as an r-priority queue and derive the average blockchain service time. The results show that MCBS enables networks to categorize applications according to their priorities. Finally, we propose an adaptive double-spend attack (ADSA) which makes blockchains more inconsistent compared to traditional double-spend attacks (TDSA). Our results show that when attackers control 45% of the total mining processing power, to keep the successful attack probability less than 0.1%, network nodes should receive at least 547 confirmation blocks for each block compared to 340 confirmation blocks for a TDSA. Thus, to avoid inconsistency, network nodes should wait for more confirmation blocks.

Text

2021-04-21

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/77874

Electrical and Computer Engineering

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/