The US Department of Defense is looking for what could be considered the “holy grail of data encryption,” which would seal a loophole that allows hackers to access sensitive information while it is being processed.

If someone wanted to use the encrypted data to do something useful, they would first have to decrypt it back into so-called “plain text”, which makes it susceptible to snooping again. To help protect that now decrypted information, those who work with the plain text usually only trust soon computers . But, as can be seen from the usual data breach headlines at major organizations, it is difficult to know which devices are secure.

“Given all the news about these hacks, these malware attacks, we cannot fully trust all of our hardware or software systems,” Tom Rondeau, program director at the Defense Advanced Research Projects Agency ( DARPA).

Related: Flying saucers for mind control: 22 declassified military secrets

This is why DARPA is trying to push advances in something called fully homomorphic encryption (FHE). The technique makes it possible to analyze computer data while it is still in encrypted form. That could allow financial crime investigators to track confidential bank records without exposing customer details, for example, or allow health investigators to analyze private health data while preserving patient privacy, Rondeau said. The technique could also help the military keep battlefield data more secure and make it easier for allies to work with classified intelligence data.

The key to the approach is in its name, which is derived from the Latin words “homos”, which means “equal”, and “morphe”, which means “form.” Refers to the fact that certain Mathematical operations you can map data from one form to another without altering the underlying structure of the data. That means changes made to data on one form will be preserved when that data is converted back to the other. This principle can be applied to encryption, because computers represent all data, including text, as numbers.

Here’s a very simplified example of how it might work: Imagine an encryption scheme that encodes data by multiplying it by 3, so if you encrypt the number 8 you get 24. If you multiply your encrypted data by 2, you get 48. When you decrypt it again by dividing it by 3, you will get 16, which is the same result you would get if you just multiplied your unencrypted data by 2.

In this example, the encryption method is fairly easy to calculate from the result, so it is not secure. But FHE relies on something much more complicated called lattice cryptography, which encodes data as coordinates on a lattice. Lattices can be thought of as regularly spaced point grids, but unlike the 2D grids we are used to, FHE lattices are multidimensional.

Related: The 11 most beautiful mathematical equations

So instead of describing the position of each data point with simple X, Y coordinates, the number of axes can be huge, and each unique piece of data is described by thousands of coordinates. Data points can also be placed between points, so each coordinate can have many decimal places to indicate its precise location. This makes encryption essentially impossible to crack, even for quantum computers. That’s a promising feature, Rondeau said, because today’s leading encryption methods aren’t quantum-proof.

FHE is based on a multidimensional network. (Image credit: ROBERT BROOK / SCIENCE PHOTO LIBRARY via Getty Images)
The big problem is that the processing of this data is very slow on today’s computers, about a million times slower than the processing times of unencrypted data. That’s why DARPA has launched a research program called Data Protection in Virtual Environments (DPRIVE), which Rondeau runs, to speed things up. The program recently awarded contracts to crypto startup Duality Technologies, software company Galois, nonprofit SRI International, and a division of Intel, called Intel Federal to design new processors and software to increase speeds to only 10 times slower than normal, which is 100,000 times faster than current processing for fully homomorphic encryption.

The FHE is very slow due to the way the calculations are performed. To further complicate matters, those data points do not remain static. The researchers found that mathematical operations such as multiplication or addition can be performed by moving data points within the lattice. By combining many of these operations, researchers can perform all kinds of calculations without deciphering the data. When you decode the response, there is a possibility that someone could spy on it; but that answer still wouldn’t reveal anything about the data used to calculate it.

Related: The 9 most massive numbers that exist

The general problem with this process is that moving precisely placed data points in high-dimensional space is much more complicated than doing calculations with simple binary data – the typical ones and zeros of today’s computers.

“It’s this explosion of data,” Rondeau told Live Science. “Now, every calculation is not just manipulating one bit. It is manipulating all this information, all these representations of the dimensions.”

There are two main approaches that DARPA-funded companies can use to keep things simple, Rondeau said. One tactic is to improve the computer’s ability to handle high-precision numbers by changing the way numbers are represented in binary code and altering chip circuits to process them more efficiently. The other is to translate the data into a smaller-dimensional space where the calculations are simpler, which also requires new hardware and software approaches.

Each of the teams involved in the program is taking a slightly different approach, but Rondeau says he is confident that they will be able to achieve the target 100,000x improvement in processing speeds.

Originally posted on Live Science.