Systematic design of universal quantum processing capable all-photonic processors
Abstract
Photonic information processing constitutes applications where an electromagnetic wave encodes an information of interest, examples being optical communications, imaging, and various forms of optical sensors such as vibrometers, range-doppler radars, and hyper-spectral remote sensors.Any form of photonic information processing can be thoughts of as a special purpose optical computer, where the computation involves the following steps:(1) encoding the information of interest in light,(2) some form of unavoidable noisy propagation based on the application of interest, that causes degradation of the information stored in the light,(3) detection, or transduction of the information bearing light and associated conversion of the information bearing signal in the electronic domain, and finally(4) signal processing in the electronic domain to extract the information of interest.There is a lot of ongoing research on novel methods in domain (4), to extract information most efficiently with the least amount of computations. Compressive sensors, machine learning (ML) methods, and in general the artificial intelligence (AI) field concerns development of intelligent processing of information after it has been transduced into the electronic domain. Since light is a quantum mechanical object, optical domain signal processing allows for:(1) strictly more powerful ways to manipulate and extract information, and(2) improvements in the efficiency and size-weight-and-power (SWaP) of the computation.Imparting optical-domain transformations to the information-bearing light prior to detecting it can pre-dispose the light in a far more information-favorable fashion to the detection noise, and can have revolutionary performance improvements over conventional photonic information processing systems, such as optical sensor and communications systems. Quantum information and estimationtheoretic tools allow us to evaluate fundamental limits to the efficiency of photonic information processing, and in turn lets us evaluate~in terms of an abstract mathematical operator-theoretic sense~the optimal (often quantum-optical) transformations to be performed on the information-carryingoptical field to enable attaining information processing at the quantum performance limit. Most often, despite these powerful quantum information theoretic tools, the mathematical specifications of the optical domain quantum computations do not translate to obvious readily-realizable physical designs of optical quantum circuits. In this project, we will systematically address and solve the aforesaid bosonic circuit synthesis problem that lies at the heart of all applications of quantum-enhanced photonic information processing~by developing an efficient and systematic design of universal quantum processingcapable all-photonic processors into a set of readily realizable components that constitute a universal set (e.g., squeezing, linear optics and photon detection)~with applications to specific photonic quantum enhanced computations on optical-domain information in sensing and communications problems of interest to the Navy, and generally of interest to the US DoD.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Apr 25, 2019
- Source ID
- N000141912189
Entities
People
- Saikat Guha
Organizations
- Office of Naval Research
- United States Navy
- University of Arizona