Early steps have been taken toward the development of analog-to-digital converters (ADCs) that would incorporate photonic quantizers based on the technology of InP semiconductors. These photonic ADCs are intended to overcome the sampling speed and temporal resolution limitations of state-of-the-art all-electronic ADCs, so that outputs of radar and other sensor systems at frequencies as high as tens of gigahertz could be sampled directly, without need for analog signal processing to effect down-conversion in frequency prior to sampling.
The figure depicts the architecture of a basic photonic ADC as envisioned. The primary photonic subsystems would be the following:
- A mode-locked laser, which would generate an optical pulse train to be used as a sampling source and clock signal;
- An electro-optical modulator, which would amplitude-modulate a portion the laser pulse train power with the analog radio-frequency (RF) signal to be digitized; and
- A quantizer.
Because the specifications of these components differ substantially from those of photonic components now commercially available, further development effort will be necessary to enable realization of a practical photonic ADC.
There are many potential alternative approaches to development of the quantizer. In the primary approach considered thus far, the quantizer would include a module containing passive semiconductor (e.g., InGaAs) saturable absorbers having saturation levels corresponding to the various quantization levels. The saturable-absorber module would be followed by an optoelectronic module, which would include photodiodes, photonic integrate-and-reset circuits, and electronic comparators. The outputs of the electronic comparators would be the desired electronic digital samples of the incoming analog RF signal, in a format suitable for further digital signal processing.
There are also potential alternative approaches to development of the electro- optical modulator: (1) one based on LiNbO3 Mach-Zehnder modulators (which are commercially available but may not satisfy performance requirements for original intended military applications), (2) one based on electroabsorption in such semiconductor material systems as InAlAs/InGaAlAs, and (3) one based on polymer modulators (the state of development of which is not mature as are the states of development of LiNbO3 and semiconductor devices). Part of the early development effort was focused on electroabsorption modulators (EAMs) for two main reasons: (1) they can be made smaller and lighter in weight (relative to LiNbO3 Mach-Zehnder modulators); and (2) because the EAM semiconductor materials are compatible with the semiconductor materials to be used for other components, it may become possible to integrate the EAMs the saturable absorbers, photodiodes, and associated electronic circuitry.
Another part of the early development effort was focused on a laser that is required to be environmentally stable and compact and to emit pulses characterized by timing jitter of less than 10 fs over a repetition-frequency range from 10 Hz to 5 MHz. One laser considered was based on a semiconductor laser diode placed within an external optical cavity. Another, taking advantage of experience with fiber lasers, was a coupled optoelectronic oscillator, the basic nature of which eliminated the need for an expensive, large, heavy RF source to provide mode-locking. Additional investigations addressed the use of high-concentration-Er-doped fibers and an Er-doped waveguide as gain media.
This work was done by Rebecca Bussjager, Michael Hayduk, Steven Johns, Michael Fanto, Reinhard Erdmann, Joseph Osman, John Malowicki, and David Winter of the Air Force Research Laboratory. AFRL-0026