Generate a simulated Monero network — gen.network • xmrpeers

Generates a simulated Monero network based on user-provided IP addresses of reachable nodes. The assumed number of unreachable nodes can be provided. A list of malicious node IP addresses can specify nodes that do not establish outbound connections. The user can specify whether the peer selection algorithm should perform subnet deduplication, and at what subnet mask level. Network summary statistics are optionally computed.

Usage

gen.network(
  outbound.ips,
  malicious.ips = NULL,
  n.unreachable = 0,
  already.connected.subnet.level = 16,
  deduplication.subnet.level = 24,
  do.deduplication = TRUE,
  default.outbound.connections = 12,
  dropped.connection.churns = 12,
  compute.network.stats = TRUE
)

Arguments

outbound.ips: A character vector of IP addresses that host reachable Monero nodes. Required.
malicious.ips: Optional character vector of IP addresses that are suspected to be malicious. Can contain IP address ranges with subnet notation.
n.unreachable: Number of unreachable nodes assumed to be in the network.
already.connected.subnet.level: The subnet mask level of the "already-connected-subnet" disqualifying condition. Set to 32 to disable this disqualifying condition.
deduplication.subnet.level: The subnet mask level at which to perform subnet deduplication.
do.deduplication: If TRUE, perform subnet deduplication. If FALSE, do not.
default.outbound.connections: The number of outbound connections of each node.
dropped.connection.churns: After default.outbound.connections has been reached, the number of times to drop one connection and add another one. The "already-connected-subnet" behavior makes some peer churning necessary to get the correct probability distribution.
compute.network.stats: If TRUE, compute network summary statistics of the simulated network.

Value

A list with three elements:

nodes: A data.table with seven columns: index, an index of the node. ip, the IP address of the node, if it is reachable (NA if it is not). already.connected.subnet, the subnet that the node belongs to, based on the already.connected.subnet.level argument specified by the user. deduplication.subnet, the subnet that the node belongs to, based on the deduplication.subnet.level argument specified by the user. reachable, TRUE if reachable and FALSE if not. malicious, TRUE if node is on the malicious.ips list supplied by the user. n.inbound, number of inbound connections of the node in the simulated network. n.malicious.outbound, the number of outbound connections to nodes on the malicious.ips list.
edgelist: A data.table. The network edge list of the directed graph. The origin column is the node establishing the connection. The destination column is the node accepting the connection. The index column of the nodes data.table is used as the identifier.
network.stats: A list of four network summary statistics, computed by the igraph package: centr_betw, centr_clo, centr_degree, and centr_eigen. See their documentation in the igraph package for interpretation.

Examples

if (FALSE) { # \dontrun{
data(good_peers)

good_peers <- stringr::str_extract(good_peers,
  "[0-9]{1,3}[.][0-9]{1,3}[.][0-9]{1,3}[.][0-9]{1,3}")
good_peers <- unique(good_peers)
good_peers <- na.omit(good_peers)
# Clean IP addresses

suspected.malicious.ips <- readLines(
  "https://raw.githubusercontent.com/Boog900/monero-ban-list/refs/heads/main/ban_list.txt")

future::plan(future::multisession,
  workers = max(c(1, floor(parallelly::availableCores()/6))))
# Multi-threaded is recommended

share.reachable <- 0.2
# Set share of nodes that are reachable to 20 percent

n.assumed.unreachable <- floor(length(good_peers) *
    ((1 - share.reachable) / share.reachable))

set.seed(314)
# This is a random simulation

generated.network <- gen.network(outbound.ips = good_peers,
  malicious.ips = suspected.malicious.ips,
  n.unreachable = n.assumed.unreachable)

hist(generated.network$nodes[
  reachable == TRUE & malicious == FALSE, n.inbound], breaks = 50)

# Network stats
sapply(names(generated.network$network.stats), function(x) {
  generated.network$network.stats[[x]]$centralization
})
} # }