Quantum mechanics imposes a limit on what we can know about subatomic particles. If physicists measure a particle’s position, they cannot also measure its momentum, so the theory goes. But a new experiment has managed to circumvent this rule—the so-called uncertainty principle—by ascertaining just a little bit about a particle’s position, thus retaining the ability to measure its momentum, too.
 
The uncertainty principle, formulated by Werner Heisenberg in 1927, is a consequence of the fuzziness of the universe at microscopic scales. Quantum mechanics revealed that particles are not just tiny marbles that act like ordinary objects we can see and touch. Instead of being in a particular place at a particular time, particles actually exist in a haze of probability. Their chances of being in any given state are described by an equation called the quantum wavefunction. Any measurement of a particle “collapses” its wavefunction, in effect forcing it to choose a value for the measured characteristic and eliminating the possibility of knowing anything about its related properties.

Recently, physicists decided to see if they could overcome this limitation by using a new engineering technique called compressive sensing. This tool for making efficient measurements has already been applied successfully in digital photographs, MRI scans and many other technologies. Normally, measuring devices would take a detailed reading and afterward compress it for ease of use. For example, cameras take large raw files and then convert them to compressed jpegs. In compressive sensing, however, engineers aim to compress a signal while measuring it, allowing them to take many fewer measurements—the equivalent of capturing images as jpegs rather than raw files.
 
This same technique of acquiring the minimum amount of information needed for a measurement seemed to offer a way around the uncertainty principle. To test compressive sensing in the quantum world, physicist John C. Howell and his team at the University of Rochester set out to measure the position and momentum of a photon—a particle of light. They shone a laser through a box equipped with an array of mirrors that could either point toward or away from a detector at its end. These mirrors formed a filter, allowing photons through in some places and blocking them in others. If a photon made it to the detector, the physicists knew it had been in one of the locations where the mirrors offered a throughway. The filter provided a way of measuring a particle’s position without knowing exactly where it was—without collapsing its wavefunction. “All we know is either the photon can get through that pattern, or it can’t,” says Gregory A.  Howland, first author of a paper reporting the research published June 26 in Physical Review Letters. “It turns out that because of that we’re still able to figure out the momentum—where it’s going. The penalty that we pay is that our measurement of where it’s going gets a little bit of noise on it.” A less precise momentum measurement, however, is better than no momentum measurement at all.
 
The physicists stress that they have not broken any laws of physics. “We do not violate the uncertainty principle,” Howland says. “We just use it in a clever way.” The technique could prove powerful for developing technologies such as quantum cryptography and quantum computers, which aim to harness the fuzzy quantum properties of particles for technological applications. The more information quantum measurements can acquire, the better such technologies could work. Howland’s experiment offers a more efficient quantum measurement than has traditionally been possible, says Aephraim M. Steinberg, a physicist at the University of Toronto who was not involved in the research. “This is one of a number of novel techniques which seem poised to prove indispensible for economically characterizing large systems.” In other words, the physicists seem to have found a way to get more data with less measurement—or more bang for their buck.