Network Coding for Anonymous P2P Communication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anonymous networking systems, particularly those relying on public key infrastructure (PKI), face challenges in providing strong information-theoretic security and are vulnerable to adversarial nodes that can collude to identify senders and receivers in peer-to-peer (P2P) overlay networks, especially when a reliable PKI is absent or compromised.
Innovation Solution
A novel network coding scheme is introduced that constructs an overlay subgraph with a formal information-theoretic security characterization, using a randomized strategy to maximize adversary uncertainty and improve anonymity, where each node decodes its intended information while maintaining a constant packet size and using linear algebra operations instead of computationally expensive cryptographic transformations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If network coding is used at intermediate nodes to improve networking performance and reduce complexity, then computational complexity is reduced by replacing cryptographic operations with linear algebra operations, but information-theoretic security against colluding adversarial nodes becomes more challenging to guarantee
Solution Approach 1:
The communication path is segmented into multiple layers with multiple nodes per layer. Messages are split into multiple shares and distributed across different paths. This segmentation ensures that no single node or small group of colluding nodes can reconstruct the original message, thereby maintaining information-theoretic security while using computationally simpler network coding operations at each intermediate node.
Solution Approach 2:
The patent implements a nested structure where messages are encoded with multiple layers of secrecy shares. Each layer contains encoded information that requires specific combinations of shares from different paths to decode. This nested encoding provides formal information-theoretic security guarantees while allowing intermediate nodes to perform simpler linear algebra operations rather than expensive cryptographic transformations.
2Reliability
If a randomized strategy is used to maximize adversary uncertainty and improve anonymity, then anonymity is improved by making it harder for adversaries to identify senders and receivers, but protocol complexity and resource usage increase
Solution Approach 1:
The patent employs dynamic randomized strategies where the network path, number of layers, and distribution of secrecy shares are randomly determined for each communication session. This dynamic randomization maximizes adversary uncertainty and improves anonymity. The protocol complexity is managed by using efficient randomization algorithms and pre-computing certain parameters, balancing the trade-off between anonymity and protocol complexity.
3Reliability
If multiple layers with width larger than or equal to two are constructed to prevent identification of communicating nodes, then anonymity is improved, but resource usage and network overhead increase
Solution Approach 1:
The patent designs a multi-layer overlay subgraph structure where each layer serves multiple functions: it provides anonymity by hiding the communication path, enables network coding operations for efficient data transmission, and distributes secrecy shares for information-theoretic security. This multi-functionality allows the system to achieve strong anonymity guarantees with controlled network overhead, as the same structural elements serve multiple protective and functional purposes simultaneously.
Data Source
AI summary
A computer-based peer-to-peer anonymous networking using a coding scheme with a formal information theoretic security characterization to construct a corresponding overlay subgraph is presented. A code construction based on linear coding which can be used for data transfer in an anonymous network is also presented. Various performance and simulation of the presented codes are provided in the Annex A1 which makes part of the present disclosure.