This course will develop the mathematical theory of noise in optical detection from first principles, with the goal of understanding the fundamental limits of efficiency with which one can extract information encoded in light. We will explore how optical-domain interferometric manipulations of the information bearing light, i.e., prior to the actual detection, and the use of detection-induced electro-optic feedback during the detection process can alter the post-detection noise statistics in a favorable manner, thereby facilitating improved efficiency in information extraction. Throughout the course, we will evaluate applications of such novel optical detection methods in optical communications and sensing, and compare their performance with those with conventional ways of detecting light. We will also compare the performance of these novel detection methods to the best performance achievable---in the given problem context---as governed by the laws of (quantum) physics, without showing explicit derivations of those fundamental quantum limits. The primary goal behind this course is to equip students (as well as interested postdocs and faculty) coming from a broad background who are considering taking on theoretical or experimental research in quantum enhanced photonic information processing, with intuitions on a deeper way to think of optical detection, and to develop an appreciation of: (1) the value of a full quantum treatment of light to find fundamental limits of encoding information in the photon, and (2) how pre-detection manipulation of the information-bearing light can help dispose it information favorably with respect to the inevitable detection noise.
This course will not assume any background in optics, stochastic processes, quantum mechanics, information theory or estimation theory. However, an undergraduate mathematical background and proficiency in complex numbers, probability theory, and linear algebra (vectors and matrices) will be assumed.
Instructor(s)
- saikat