Oct 9th: Dr. Monika Patel and Dr. Hari Krovi (BBN)
Dr. Monika Patel and Dr. Hari Krovi, BBN
“Enabling Optical Technologies for Quantum Information Processing”
THURSDAY, OCTOBER 9, 2014
12:00 PM – 1:00 PM
The Quantum Information Processing group at BBN Technologies focuses on exploiting quantum phenomena for sensing, computing and imaging. We will introduce a few of our ongoing projects in theoretical and experimental quantum optics.
Towards the Holevo limit: When sending classical information over quantum channels, it has been shown that there is a fundamental limit, called the Holevo capacity. Achieving this limit is possible theoretically, but seems to be a daunting task practically. There are three part of the communication system that contribute towards the capacity of the system. The states used for modulation, the communication codes used and finally, the type of receiver used to decode. The last part is related to the type of quantum mechanical measurement that one performs on the received quantum states. In this project, various scenarios were investigated to improve the capacity of the communication system and take it closer to the Holevo limit.
Imaging and ranging: Can one distinguish one star from two by collecting the received light? It turns out that using quantum mechanics, one can do better than using classical methods for this problem. In this case as well, there is a fundamental limit imposed by quantum mechanics on how well we can distinguish the two cases. In this project, we designed receivers that achieve this limit asymptotically. This is one example of passive imaging since we are not interrogating the target with an optical state.
Computation: Here we are interested in determining how incremental resources lead to increased computational power. For optical quantum computing, one has examples of resources are beamsplitters, phase-shifters, squeezers etc. In a related question, we are also interested in determining resources needed to perform a given computational task. In other words, one would like to know how many squeezers are needed in a particular model of computation for an algorithm of a given size. In addition, we would like to know the errors allowed in these resources that can still lead to reliable computation via error correction.
Imaging with a structured illumination source: For imaging unknown objects, a key objective is to be able to extract maximum information about the object while expending as few optical photons as possible. We demonstrate that in the presence of a strong background, image reconstruction employing a structured illumination light source with optimal processing of correlated measurements from a pair of detector arrays can outperform a laser light source of identical transmit power.
“The students in MIT’s new NSF training program will be encouraged to cross disciplines, and develop a common fellowship with their peers. We will also address training for post-academic jobs directly by connecting students to government and industrial members of the iQuISE Consortium.”
—Seth Lloyd, Co-Director, iQuISE, and Professor of Mechanical Engineering and Professor of Engineering Systems