Which graph generation models are suitable as representation of large-scale (100-300 nodes) homogeneous wireless sensor nodes, in case I am only interested in the interconnection of those nodes and the existence/position of the base station is not relevant?

I consider unit disk/sphere graphs a good representation but I couldn't find a graph generator (e.g. in python) that provides features to generate those graphs as Model for WSNs. Furthermore I would like to ensure that the resulting graphs are connected and I would favour to have some control on the average node degree or at least can somehow ensure that the nodes are distributed well enough throughout a given space while maintaining the "connected" property.

Is there a good existing approach to generate those models?

I thought about algorithms to generate scale-free graphs and then extend the resulting graphs by adding connections of nodes in a certain areas (based on the unit disc/sphere concept) - which would also allow me to have some control over the average degree. The problem is that scale-free graphs have areas in which nodes are a lot closer together. Therefore they have a lot more neighbours. When applying the unit disk graph approach it is likely that the node degree in those areas significantly raises while in other areas the node degrees will not change. This could create inbalances that are atypical for many wireless sensor networks.

While I would like to know which approaches or even implemented graph generators are suitable to generate graphs as model for UDGs also feel free to correct me on further assumptions I've made.

  • 1
    try posting at softwarerecs.stackexchange.com – jsotola Mar 8 at 19:20
  • please provide an example of the type of graph that you want to generate ... someone here may have seen what you want, but cannot make the association just from your description – jsotola Mar 8 at 19:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.