Author: LEHÉRICY Gabriel 2020-02-12 10:11:38
Published on: 2020-02-12T10:11:38+00:00
The authors of a new paper on ant routing have responded to comments regarding intermediary nodes, spamming, payment failures, and scalability. They note that in large highly connected networks, triangulation is not feasible and the distance between most pairs of nodes is between 3 and 7. To obfuscate the counter, they want to avoid giving information to immediate neighbors about the origin or end of the transaction. Nodes are free to relay seeds and can choose to ignore seeds coming from nodes they suspect of acting maliciously. Ant routing solves the problem of high payment failure rates by forwarding pheromones to neighbors only if their channel balance allows the amount of the transaction to go through.The authors also address concerns about adding random numbers to the initial hop distance counter and point out that such additions do not obscure as much as might be thought. They note that such a technique may protect against single node operators but not medium-corporate-level surveillors who can run 3 nodes on the network and triangulate where the originator of the pheromone is. Finally, they note that Dijkstra algorithm is a simulation of pheromones in Ant Routing, which is why Dijkstra can discover shortest paths.Ant Routing is a proposed upgrade to the current Lightning Network source-pathfinding onion-routing scheme that adds privacy to the payee. However, having a single pheromone seed that is recognizable for the entire path prevents any kind of path decorrelation. Ant Routing also requires every payment to have a pheromone broadcasted, which negates the big-O scaling achieved by Lightning. Furthermore, surveillors can easily determine payments and the maximum distance to the destination and likely source.
Updated on: 2023-06-02T23:34:41.500410+00:00