Blinded Outcome Diversification Cycles for Blockchain Address Privacy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing blockchain transactions face security vulnerabilities due to publicly available data allowing monitoring nodes to trace associations between input and output addresses, even with coin mixing operations, leading to potential exposure of user identities through data mining and machine-learning analyses.
Innovation Solution
Implement Blinded Outcome Diversification (BOD) methods involving a dealer node and multiple participating nodes, where each node generates ciphertext, a selector node randomly selects transaction options, and propagator nodes shuffle and encrypt outputs, increasing entropy and obscuring address associations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If coin mixing operations are used to obscure token movement between addresses, then traceability is reduced, but monitoring nodes can still perform data mining and machine-learning operations on publicly available data to determine associations between inputs and outputs
Solution Approach 1:
The patent segments the coin mixing process into multiple independent cycles, each involving a different selector node. This segmentation prevents monitoring nodes from easily tracing token movement across the entire mixing operation, as each cycle appears as an independent transaction pattern. The segmentation of mixing operations into discrete cycles with different selectors breaks the continuity of traceable patterns that machine learning algorithms would otherwise exploit.
Solution Approach 2:
The patent introduces dynamic selection of selector nodes for each mixing cycle, where the selector is chosen randomly or pseudo-randomly from the network. This dynamic approach changes the mixing parameters and participant roles for each cycle, making it difficult for monitoring nodes to establish static patterns or correlations. The dynamic reconfiguration of mixing participants across cycles increases entropy and obscures the relationship between input and output addresses.
2Ease of operation
If third party coordinators are used to facilitate coin mixing, then mixing operations can be performed, but the third party coordinator may be compromised resulting in exposure of more data than would otherwise be publicly available
Solution Approach 1:
The patent extracts the coordinator role from the mixing process and replaces it with distributed selector nodes that are randomly selected from the network for each cycle. This extraction eliminates the single point of failure represented by a third-party coordinator. Each selector node operates independently for its designated cycle and has no access to information from other cycles, removing the risk of centralized compromise while maintaining the facilitation of mixing operations.
Solution Approach 2:
The patent introduces randomly selected selector nodes as temporary intermediaries for each mixing cycle. These selectors act as mediators that facilitate the mixing operation for a single cycle without having access to or control over other cycles. This intermediary approach distributes trust across multiple temporary actors rather than relying on a single persistent coordinator, reducing the security risk while maintaining operational ease.
3Measurement precision
If publicly available blockchain data is analyzed using machine learning algorithms, then patterns attributable to specific sources can be detected, but this enables monitoring nodes to determine associations between addresses and identify owners
Solution Approach 1:
The patent changes the parameters of mixing operations for each cycle by selecting different selector nodes and potentially different mixing configurations. This parameter variation prevents the formation of consistent patterns that machine learning algorithms could detect and exploit. By randomly varying the selector and other operational parameters across cycles, the system increases entropy and eliminates the stable patterns that would otherwise enable precise measurement of address associations through data mining.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
There is provided a computer-implemented method to increase security of a blockchainimplemented transaction, in a network comprising a dealer node, and a plurality of participating nodes each having its own public key and private key. For each of a plurality of cycles, a set of output addresses are obtained by the following steps. The dealer node, orders the participating nodes such that there exists a cycle node order. It designates a first node in the node order as a message originator and the last node as a propagator, and designates one of the remaining participating nodes as a selector, wherein each node is not the message originator in more than one cycle node order, and each node is not the propagator in more than one cycle order. The message originator generates ciphertext comprising a nested encrypted first message and a nested encrypted second message, each message comprising one or more transaction details, and transmits the ciphertext to a next participating node in the cycle node order. The ciphertext is transmitted from node to node in the cycle node order, each node decrypting the ciphertext before transmitting it, wherein the selector additionally removes one of the nested encrypted messages from the ciphertext before transmitting it. The propagator decrypts the remaining single message to obtain the one or more transaction details contained in the message. When all the cycles are concluded, all the nodes designated as propagator generate a shuffled order of the transaction details obtained from all the cycles. A blockchain transaction is generated wherein each output of the transaction is based on one of the transaction details, and the outputs are ordered according to the shuffled order.