(Purdue.edu) A new quantum random walk technique developed by engineers at Purdue University could eventually allow computers to search through data at speeds beyond that of conventional computers.
In a random walk, an agent randomly moves to the right or left at each time interval, much like a ball in a Galton board hopping to adjacent slots at each step. Such an agent can sweep a database space to search for a special value, for example finding its minimum.
In the quantum realm, the laws of random walks are fundamentally different. A quantum agent can move to right and left simultaneously at each step. In other words, a quantum particle can travel in several trajectories at the same time, which can lead to searches of large-scale databases dramatically faster.
Andrew Weiner, Purdue University’s Scifres Family Distinguished Professor of Electrical and Computer Engineering, said their technique uses photons at specific colors, or frequencies. “We like to refer to this as the quantum walk of the rainbow,” Weiner said. “One major advantage is that, unlike other types of quantum computing, this technique can be done at room temperature.”
Another advantage is that experiments can be done with integrated photonics and other components common in lightwave or optical communication, which could reduce cost and bring compatibility with fiber optics communications infrastructure.