Ants in the system aid to networks

Electronics Times, July 17, 2000

Biologists studying the complex social structures of insects such as ants may have developed knowledge which computer scientists can use to develop optimisation and control algorithms.

Eric Bonabeau, of the Santa Fe Institute in New Mexico, and colleagues have reported in Nature on the `swarm' method of food gathering used by ants.

Ant colonies send out foragers which explore the nest's surroundings virtually at random. If a forager finds food, it returns to the colony, laying a pheromone trail which other ants follow. Through the laying of pheromones, other ants can find and establish shorter routes to the food source. By simulating the ants' technique, researchers have developed the `ant colony optimisation algorithm' (ACOA); an approach to computational problem-solving called swarm intelligence.

Engineers are using the ACOA in areas such as identifying the best way of routing through a comms network. The searching and feedback processes offers great flexibility and the ability to respond to changes such as channels being added or removed to the network.

British Telecom is exploring the possibilities of ACOAs.