Following the steps of the ants for knowing a networking logic? Does it sound crazy to the ears? Maybe yes, but those little creatures making their movements always in disciplined queues are evoking interest among scientific communities for their exceptional ability of networking. We always looked at ants with amused feelings. They are so systematic, disciplined and even far-sighted.
Yes, they follow a rule to gather in a place, communicate with each other and they are even far-sighted to save food for the rainy seasons. So, ants are really good in an array of things. It seems that they really made their movement and living modeled after a scientific idea. They are also very much democratic in participating and following their leaders.
We always exclaimed on this particular phenomenon but recently scientists took a serious interest in them. Scientists now recognize that the ants really want to determine the density of their own on the basis of their frequent collision in moving around or exploring the environments. Recently some scientists at Massachusetts Institute of Technology (MIT) observed this and elaborated their findings in a research paper.
The research paper argued that the so-called exploration of ants through “random walk” can actually offer us the very foundation for algorithms of network communication function. The repeating of encounters is another scenario in their movements that allows deeper insights in the natural networking algorithm they follow in their daily life and explorations. Let us have a quick look at each of these aspects in their daily movements.
The researchers at MIT characterize each of the ant’s environment as a grid while other ants are scattered randomly over that area. The ant which is being examined may be termed as the explorer begins its movement at any of the cells in his grid has a probability to move to surrounding cells. The ant of another grid has equal probability to move to the nearby cell. Each ant as the explorer counts the number of ants in the adjacent or surrounding cells he visits.
In many ways the result found by the researchers in studying the random walk of ants is extraordinary. The explorer ant very likely has to return to a previously visited cell. This causes a likelihood of oversampling of data. According to researchers such repeat encounters may cause oversampling in some cases but it balances by the sheer fact that all the ants in far away cells are not going to be met by the explorer.
In the research, the grid being used for modeling the environment of the ant is a typical data structure referred as a graph. Now this graph includes nodes that are represented by circles and their respective edges. These edges are again represented as linked segments bordering nodes. So, in a grid, each of the cells is a node sharing edges with the adjacent cells.
The researchers of MIT followed an analytic technique that best applies to any graph, similar to the one used to describe networked members on a social networking site. The new research paper beside focusing on one explorer ants movement took consideration of such movements of various ants to make the estimate more accurate.