Imagine that there were thousands of tiny computerized sensors the size of dust-specks floating around in your body, constantly scanning for infection, viruses, or even cancer? Well, researchers have predicted that this sort of situation could be a possibility in the near future.
Currently, networks of matchbox-sized sensors are being used to analyze a wide variety of variables, from the movements of wildlife to the trajectory of bullets in military testing. Even more complex networking technology is currently in development, including cheap sensors the size of dust particles. However, the use of such networks has created an array of complications, namely errors in multiple sensors tallying changes in the same variable. This issue has led to two distinct solutions……
1) Shapes =)
Using topology, the qualitative study of shapes, sensor-network engineers will be able to develop sensors that can recreate the structures of what’s being monitored to differentiate between what they’ve already tallied and what they haven’t, or at least according to mathematician Robert Ghrist (University of Illinois).
2) A Compute Game!
Known as the “Minesweeper Approach”, this process involves analyzing how many “bombs” are adjacent to each corresponding sensor and, by griding the layout of the sensors, collecting an accurate physical count of the variable being tracked.
The technology is still in its premature stages, and researchers are giving no promises in regards to the extent of what this technology will be able to accomplish nor any timetables for the developments. However, according to Ghrist, topological sensor-networks are “right around the corner”.
