Appl. Phys. Lett. 104, 111101 (2014)

There are two main problems associated with conventional analog-to-digital converters — the maximum frequency that can be captured is limited to half the sampling rate (the so-called Nyquist rate) and a 'big data' storage problem in real-time instruments due to the long digital record length associated with Nyquist sampling. Now, Mohammad Asghari and Bahram Jalali from University of California at Los Angeles in the USA report that a self-adaptive stretch, which they term an anamorphic stretch transform, can help solve both problems. The approach enables digitizers to capture waveforms beyond their usual bandwidth limit, with the size of digital data size also being reduced at the same time. The approach relies on warping the signal's complex field in the analog domain before sampling and digitizing by reshaping the signal with a nonlinear transformation. The signal reshaping is then combined with complex field detection followed by digital reconstruction. In a proof-of-principle demonstration, the researchers compressed the modulation bandwidth of an optical signal by a factor of 500 and reduced its modulation time–bandwidth product by a factor of 2.73 while achieving a 16 dB improvement in the power efficiency compared to that obtained using a conventional dispersive Fourier transform.