Biomedical subjects

David Yevick

Publications and source records attributed to David Yevick.

8 recordsLinked to original sources

PubMed2006

Least-squares analysis of the Mueller matrix.

In a single-mode fiber excited by light with a fixed polarization state, the output polarizations obtained at two different optical frequencies are related by a Mueller matrix. We examine least-squares procedures for estimating this matrix from repeated measurements of the output Stokes vector for a random set of input polarization states. We then apply these methods to the determination of polarization mode dispersion and polarization-dependent loss in an optical fiber. We find that a relatively simple formalism leads to results that are comparable with those of far more involved techniques.

Journal Article↗

PubMed2006

Mueller matrix description of polarization mode dispersion and polarization-dependent loss.

We derive a differential equation that relates the Mueller matrices of an optical system at adjacent frequencies in the presence of polarization mode dispersion and polarization-dependent loss (PDL). We then demonstrate that a solution of this equation based on the Magnus expansion yields a description of the Mueller matrix in orders of the principal state vector that coincides with previously reported results for systems without PDL.

Journal Article↗

PubMed2006

Polarization mode dispersion in short fiber lengths.

We perform Monte Carlo simulations of the statistical properties of the differential group delay for fiber lengths less than and of the order of the birefringence correlation length. We find that the manner in which quantities related to the polarization mode dispersion evolve along the fiber depends significantly on the model assumed for the form of the birefringence.

Journal Article↗

PubMed2006

Improved multicanonical algorithms.

We introduce several easily programmed techniques that enhance the accuracy of multicanonical sampling. With minor modifications to the standard technique, our methods achieve equivalent or enhanced accuracy compared with existing, often far more complex, algorithmic refinements. Despite their simple formulation, these procedures have been previously overlooked because of the low cost of additional realizations in numerical calculations. When applied in the context of our recently introduced experimental multicanonical measurements, however, significant time savings can result.

Journal Article↗

PubMed2005

Spectral density matrix description of polarization mode dispersion.

We introduce a power spectral density matrix formalism that incorporates both the pulse shape and the field polarization and can therefore easily describe averages over random fluctuations of the local birefringence vector. We demonstrate that quantities such as the differential time delay, power diffusion, and decoherence effects can be obtained directly from the equations of motion for the power density matrix. This approach can be applied to pulses with arbitrary frequency-dependent polarization and intensity distributions and in particular makes possible the minimization of the eye-opening penalty through the proper choice of the initial pulse profile.

Journal Article↗

PubMed2005

Comparison of vector finite-difference techniques for modal analysis.

We compare a vector finite-difference method that correctly applies the boundary conditions at both horizontal and vertical dielectric interfaces (but not at corners or slanted interfaces) to algorithms that only approximately satisfy these boundary conditions. We find, rather unexpectedly, that for strongly guiding waveguides the boundary conditions imposed at the refractive-index discontinuities typically affect the calculated field distributions less than the procedure employed to assign the refractive index at a computational grid point. In fact, locally averaging the refractive index around each grid point transforms the precision of the most straightforward finite-difference models to that of far more sophisticated techniques. Further, H- and E-field formalisms exhibit identical accuracy.

Journal Article↗

PubMed2005

Chebyshev and Taylor approximations of polarization mode dispersion for improved compensation bandwidth.

We examine a series of experimentally realizable procedures for wide-bandwidth polarization mode dispersion compensation based on Taylor and Chebyshev approximations to the transfer matrix for light polarization in optical fibers. Our results demonstrate that a symmetric ordering of compensator elements in the Taylor procedure improves performance and that methods based on the Chebyshev approximation can significantly widen the compensation bandwidth.

Journal Article↗

PubMed2005

Multicanonical comparison of polarization-mode dispersion compensator performance.

We employ a modified version of the multicanonical algorithm to evaluate the system penalties and outage probabilities of different polarization-mode dispersion compensators. The procedure determines the optimal operating conditions for each compensator architecture far more efficiently than the standard Monte Carlo algorithm.

Journal Article↗