Computational photonics

A major ongoing theme of our research is to use computation to solve key problems in photonics and to use photonics to solve problems in computation.

For example, computational techniques can be used to probe hidden physics. A long-standing problem in frequency combs was how to measure the temporal profile of low-intensity combs. We showed how with off-the-shelf electronics and a little bit of math, the temporal profile of low-intensity combs could be directly measured.

We also use them to improve sensing. Dual comb spectroscopy traditionally had very challenging mutual coherence requirements, requirements that would effectively preclude its widespread adoption. However, we were able to show that the dual comb signal alone contained enough information to be corrected entirely computationally. This approach circumvents the usual requirements for temperature stability, optical isolation, and bias stability usually associated with these sorts of systems. (The updated Github repository where example code can be found is here.)

Computationally-corrected dual comb spectroscopy. The signal is broadened by phase noise, but statistical inference allows for the signal to be corrected. [2]

Shifted Wave Interference Fourier Transform Spectroscopy (SWIFTS), a technique developed by the PI for computationally determining the profile of low-intensity combs. [3]


  1. D. Burghoff, N. Han, and J. H. Shin, “Generalized method for the computational phase correction of arbitrary dual comb signals,” Opt. Lett., vol. 44, no. 12, pp. 2966–2969, Jun. 2019. (pdf, Github repository)
  2. D. Burghoff, Y. Yang, and Q. Hu, “Computational multiheterodyne spectroscopy,” Sci. Adv., vol. 2, no. 11, p. e1601227, Nov. 2016. (pdf, supplementary, demo code)
  3. D. Burghoff, Y. Yang, D. J. Hayton, J.-R. Gao, J. L. Reno, and Q. Hu, “Evaluating the coherence and time-domain profile of quantum cascade laser frequency combs,” Opt. Express, vol. 23, no. 2, p. 1190, Jan. 2015. (pdf, notes on SWIFTS)