Hidden side channels of quantum sources can impair secure communication

A team of researchers at the University of Toronto Engineering discovered a multidimensional side channel hidden in existing quantum communication protocols.

The new side channel occurs in quantum sources. This is a device that produces quantum particles (typically photons) that are used to send secure messages. This discovery can have important implications for quantum security.

“What makes Quantum Communication safer than classical communication is its exploiting the properties of quantum mechanics known as conjugated states,” says the doctorate. Student Amita Gunanapanditan, lead author of the paper published in the Physical Review Letter.

“For example, position and momentum are conjugated variables. If one is measured, it disturbs the other. If both variables are randomly selected for encoding, the person trying to listen to the message will be automatically detected by the party trying to communicate.

However, despite its inherent security, there are still ways to compromise quantum communication due to the flaws in the devices used in real implementations.

Between 2000 and 2012, researchers showed that the mechanisms of quantum detectors could lead to side channels. These side channels act as loopholes and allow someone to listen to the signal without introducing detectable interference.

To address this, in 2012 Professor Hoi-Kwong Lo and his collaborators developed a new protocol known as Measurement and Independent Quantum Key Distribution (MDI-QKD). The protocol effectively shorts all side channels associated with the quantum particle detector.

As the detectors were caring, Gnanapandithan, co-monitored by LO and Professor Li Qian, ended up looking for a source device, a potential side-channel associated with the other end of the communication.

“Let’s say you want to encode information based on how light comes from the source is polarized. This is what we call optical polarization,” says Gnanapandithan.

“We’ll use two conjugated polarization bases to perform the encoding. Ideally, we’d want to keep the encoding within the degrees of freedom of polarization, and we don’t want the polarization to correlate with other degrees of freedom.

The idea that the degree of freedom in encoding is uncorrelated with other degrees of freedom in optical quantum sources is known as the dimensional assumption. Violating this assumption means that the message may be unsafe.

In fact, today’s quantum sources often introduce such violations, for example, due to correlations between adjacent signals. This is called a pattern effect, and it provides information about the leakage of an earlier signal to a later signal.

However, in his latest work, Gnanapandithan used both theoretical models and physical quantum sources to demonstrate a new source of violations that had not been previously considered.

“We knew the modulation process could be distorted a bit, but what we found was that the modulation process could be time-varying, even within the same signal light pulse,” says Gnanapandithan.

“Specifically, we created a very subtle perception that this defect is in fact a violation of dimensional assumptions. Therefore, we call this type of defect “hidden multidimensional modulation.”

How big of a problem these side channels are depends on the type of equipment being used.

“If your equipment has high bandwidth, you should apply modulated signals to optical pulses to get you closer to what’s ideal,” she says.

“However, if your equipment is strictly limited to bandwidth, the modulation pulse can be severely distorted, which will exacerbate the problem.

“There is also a new type of quantum key distribution (QKD) source introduced in the literature called passive QKD sources. Passive QKD sources do not even use modulators, so these bandwidth issues do not apply.”

Lo says his future work from his team will focus on ways that could mitigate the newly discovered side channel.

“We can get creative and perhaps find ways to avoid these issues,” he says.

“But as we’ve learned in the past, our new methods can also cause problems in their own way. You may not know what layer to be, but I’m simply identifying the issues you have to deal with, and that’s what I did here.”