Back to projects

Radio Protocol Research

4 min read
STATUS: Active
Table of Contents

This is an ongoing program of research into radio protocols that borrow their behavior from living organisms. A protocol is a set of rules for when to speak, when to listen, how to find one another, and what to do when a message does not get through, and the protocols that run our networks today were designed by committees with throughput and reliability in mind. Organisms have been solving the same problems for far longer under different constraints, and the research asks what a radio protocol looks like when its rules are taken from how a slime mould routes nutrients, how an ant colony finds and abandons a path, how a bee swarm reaches a decision, how a pea plant grows toward a signal it can barely sense, or how a crow passes a warning through its social group.

The work builds on earlier projects in which the organism and the network were already the same thing. In Plant-to-Plant Protocols a garden carried infrared and 2.4 GHz traffic between plants and the weather glitched the data, and in Modeling Utopia with Slime Mould and Physarum Topologies the plasmodium was asked to lay out the topology of a network with its own body. Here the organism is not wired into the circuit but studied for its behavior, and that behavior is written down as rules a radio can follow.

The organisms

Each organism contributes a behavior that maps onto a problem a protocol has to solve.

  • Slime moulds. Physarum polycephalum grows a network of tubes toward food, thickens the tubes that carry flow and lets the others wither, and in doing so arrives at efficient and fault-tolerant routes without any central plan. Read as a protocol, this is route selection and pruning driven by use.
  • Ants. A colony finds food through pheromone trails that are reinforced each time an ant follows them and that evaporate when they are not, so that good paths strengthen and stale ones disappear on their own. Read as a protocol, this is path discovery and decay without a routing table.
  • Bees. Scouts report a candidate site through the waggle dance, which encodes direction and distance in the body, and the swarm commits once enough scouts agree. Read as a protocol, this is distributed consensus reached through repeated, cheap signaling.
  • Pea plants. Pea seedlings grow their roots toward the sound of running water and can learn to associate a cue with a resource, responding to signals at the edge of what they can sense and on a timescale of hours and days. Read as a protocol, this is slow, patient acquisition of a weak signal.
  • Crows. Crows recognize individuals, remember them, and pass warnings about them through the group, so that information about a threat spreads along social ties rather than being broadcast to everyone. Read as a protocol, this is selective relay through a trust network.

The software

Just as WSJT-X is the program that speaks FT8 and JS8Call the program that speaks JS8, these protocols need a program of their own, and that program is sig-x. It begins as a receive-only engine that turns live amateur radio into a stream of events, and it is designed so that new modes and new protocols can be added to the same stream as they are specified.

sig-x · Plant-to-Plant Protocols · sLLM · Modeling Utopia with Slime Mould · Radical Networks