Ants figure out details related to the size of their colonies by bumping into fellow ants while they randomly explore. But the ants don’t have to traverse the entire colony to know how many fellow ants they’re living with. The insects can figure it out through the number of nearby encounters they have.
Ad hoc wireless networks could use the same technique, say scientists from MIT. Just like ants learning about population densities help the creatures decide communally whether they need to build a new nest or not, the same could be true for sensors strewn around IoT environments.
+ Also on Network World:Using IoT-enabled microscopes to fight epidemic outbreaks +
The use of ad hoc networks is thought to be a viable future alternative to traditional networks, like the internet, for example. They’re networks created “on the fly,” explainsMIT on its website. Robots moving around, or sensors arbitrarily scattered around environments to measure pollution, are examples.
The most notable difference to traditional networks is that ad hoc networks don’t have central hand-offs, like routers, or phone switches. And that’s one of the appeals: Disaster workers, for example, can set up communications as they go about their work without a host device, and IoT can be built-out physically and geographically without returning to the base. In other words, configuring a router.
Experts say ad hoc is a viable way to organize some networks should billions of sensors come on stream as IoT becomes more important. But because of the lack of supervision—in other words the router or base station—the nodes have to figure out for themselves who to communicate with and perform other administrative tasks. That’s exacerbated by the fact that the sensor may unknowingly get moved.
“Suddenly some of the volcano sensors are farther away from their neighbors, with lower-bandwidth data connections than they had before,” MIT says on its website, referring to environmental changes.
Hence the need to devise ways of self-managing. And the ant model, though theoretical, is remarkably effective.
“It’s intuitive that if a bunch of people are randomly walking around an area, the number of times they bump into each other will be a surrogate of the population density,” says Cameron Musco, an MIT graduate student in electrical engineering and computer science and a paper co-author, in an MIT news article. “What we’re doing is giving a rigorous analysis behind that intuition.”
What’s more, they believe they’re on to something and that the ant population guesswork is actually very accurate.
The system, which is encapsulated in a grid, is called “random walk,” MIT explains.
The explorer ant starts off wandering from a cell in the grid “and with equal probability, moves to one of the adjacent cells. Then, with the same probability again, it moves over to one of the cells adjacent to that one, and so on,” writes Larry Hardest in the MIT News article
“The explorer counts the number of other ants inhabiting every cell it visits,” creating a random sampling, Hardest writes. The more cells covered, the better the random sampling, but as we know from marketing studies, sampling does capture pretty good estimates of things when you add algorithms for analysis.
The scientists don’t claim to have everything worked out and describe the work as theoretical, but the mathematical principals behind this could very well be how ad hoc networks ultimately get organized.