Speckle imaging and hidden phase.

The Knox-Thompson, or cross-spectrum, method provides two two-dimensional difference equations for the phase of the object spectrum. We demonstrate that, in general, the object spectrum phase can be decomposed into a regular, single-valued function determined by the divergence of the phase gradient, as well as a multivalued function determined by the circulation of the phase gradient; this second function has been called the hidden phase. The standard least-squares solution to the two-dimensional difference equations will always miss this hidden phase. We present a solution method that gives both the regular and the hidden parts of the object spectrum phase. Finally, we illustrate several examples of imaging through turbulence and postprocessing with the Knox-Thompson method, including the hidden phase.

Analysis of grating-coupled surface-emitting lasers.

Grating-coupled surface-emitting diodes are modeled and analyzed in terms of radiation power, threshold gain, and saturation effects. We find that the dominant (i.e., the lowest-loss) mode radiates poorly, while a weaker mode has a much higher radiation power and appears dominant if only the radiated power is measured. The weakly radiating mode effectively lowers the device efficiency and broadens the far-field radiation pattern. We also show that an increase in the length of the distributed Bragg reflector sections at the end can double the radiated power and reduce the threshold gain.

Low-frequency self-pulsations in asymmetric external-cavity semiconductor lasers due to multiple-feedback effects.

We investigate the dynamical stability of external-cavity semiconductor lasers with deliberate tilt asymmetries. Experimentally observed low-frequency self-pulsations are explained by small-signal analysis based on a new rateequation model that uses a time-dependent effective reflectivity to include multiple-feedback effects. We observe and explain the dependence of the self-pulsation frequency on the strength and degree of asymmetry of the feedback, on the injection current, and on the external-cavity length. These results are important in interpreting instabilities, chaos, and other complex dynamical phenomena in external-cavity semiconductor lasers.

Experimental evaluation of an unstable ring resonator with 90 degrees beam rotation: HiQ experimental results.

The first experimental results on a confocal unstable resonator with 90 degrees beam rotation (UR90 or HiQ) are presented. A beam quality of 1.06 was observed, and it varied by a maximum of 0.08 over a resonator misalignment range of 1 mrad. The outcoupled beam's polarization was circular. Reverse mode suppression was achieved on UR90 with a suppressed vs unsuppressed ratio of 115:1. Measurable reverse mode suppression was obtained with as little as 2% of the full reverse mode strength. Most of these results are compared with those for a conventional ring resonator.

Saturation effects in the carrier-induced refractive index in a semiconductor gain medium.

We derive expressions for the loaded gain and carrier-induced refractive-index change for a semiconductor gain medium. Our main conclusion is that the loaded gain and the loaded carrier-induced refractive-index change saturate differently, with the index being a much weaker function of the laser flux. This has important implications for the laser mode structure above threshold and is particularly important for filamentation suppression and stability in numerical modeling studies of gain-guided structures.

Unstable resonator with canceling edge waves.

Diffractive feedback from the edge of the output mirror of an unstable resonator can cause intensity spikes and make the resonator sensitive to tilts. A design for a stepped feedback mirror is proposed that substantially reduces the effect of edge diffraction on the resonator modes. The predicted results are a more uniform intensity profile and greatly reduced tilt sensitivity.

Matrix methods for bare resonator eigenvalue analysis.

Bare resonator eigenvalues have traditionally been calculated using Fox and Li iterative techniques or the Prony method presented by Siegman and Miller. A theoretical framework for bare resonator eigenvalue analysis is presented. Several new methods are given and compared with the Prony method.