Numerical optimization of large-scale fiber-optic splitting tree uniformity

In order to split the optical power of a signal into many fibered outputs, large-scale optical splitting trees (splitters) are used. In most applications, the optical power should remain constant from output to output. Large-scale splitting trees (typically 1 : 32 or 1 : 128) may be composed of a single-stage or of cascaded small splitters (typically 1 : 2). In the former case, power uniformity is determined by the design and manufacturing tolerances. In the latter case, considered here, uniformity may be optimized by a proper choice of the positions of the small splitters in the tree. Simulations show that less than 1-dB uniformity may be reached.     

Axis alternation for signal propagation over polarization-maintaining fibers

When a signal is split over both axes of polarization-maintaining fibers (PMFs), alternation of axes is well known to cancel the impact of the differential group delay, e.g., in filtering functions such as modulators based on Kerr effect and a Sagnac loop. This solution seems not of concern if the signal remains on one axis as is the case for polarization-maintained signal transportation. As opposed to the common belief, we show experimentally and numerically that alternating the slow and the fast axes is also important to avoid signal distortions in that case. As a matter of fact, strong distortions of polarized signals are observed when the signals are transported with PMF because of the finite extinction ratio of connectors.     

Provisioning Methods for Bit-Rate-Differentiated Services in Hybrid Optical Networks

This paper addresses the problem of routing and wavelength assignment of bit-rate-differentiated optical services in a hybrid network. Hybrid optical networks are composed of resources, such as fiber links and photonic/electronic switches, that vary in their capabilities and transmission qualities. These networks are also responsible for the realization of optical services with varying quality-of-service (QoS) guarantees. In such networks, it is required to have a cost-effective assignment of the optical and electronic resources to these services in order to maximize the revenue of the network operator. This paper deals with optical services that are defined according to their tolerance to transmission impairments. We first divide the provisioning problem into two phases: (1) routing and (2) wavelength assignment and regeneration reservation. In the routing phase, a set of k-routes are generated to select from in the second phase, where each route optimizes a specific aspect of the problem (e.g., number of hops, maximum accumulated noise, etc.). The second phase, using the information about the resources along each route, attempts at finding the best wavelength allocation on that route such that the signal quality meets the service-level agreement (SLA). The second phase also uses the minimum number of regenerator ports on intermediate nodes for the purpose of wavelength translation and signal clean-up. Comparisons of the above scheme with a probing-based method, reveal substantial enhancements to the blocking performance with a maximum running time increase of 60%. In addition, the use of multiple routes provides higher reduction in the blocking probability over single-routing schemes. Moreover, the proposed, non-pessimistic, provisioning approach has a major impact on reducing the regeneration budget of the network.     

Comparison of the propagation performance over standard dispersive fiber between InP-based /spl pi/-phase-shifted and symmetrical Mach-Zehnder modulators

We estimated and experimentally verified the output phase and intensity characteristics versus the applied voltage of an InP-based Mach-Zehnder modulator from the measured data of refractive index and loss variations specific to the phase section material (Fabry-Perot method). We have then simulated the propagation performance of a Mach-Zehnder modulator taking into account the possible phase-shift between both arm of the interferometer. In total agreement with experimental results, we show that the simple single-arm modulation scheme does not require high voltages and that the /spl pi/-phase-shifted Mach-Zehnder modulator is far more convenient for a 10-Gb/s propagation over standard dispersive fiber than a symmetrical modulator.     

Comparative study of classical, colorimetric and immunologic staining methods for the assessment of tumor cell viability

The trypan blue exclusion test, the MTT test and an immunostaining test for apoptosis were performed before and after incubation of SW620 human colonic carcinoma cells with different cytotoxic agents (CTAs) in order to assess tumor cell viability and CTA efficacy in vitro. A modified MTT test using light microscopy was also performed. A good correlation was found between the trypan blue assay and the MTT test, as determined by spectrophotometry. There was no 'overestimation' of cell viability as measured by the trypan blue test. The monitoring of formazan formation by light microscopy was feasible, but not very reliable since it did not show a good correlation with findings determined by spectrophotometry. The apoptosis test failed to show good correlation with other tests. Distilled water had no relevant cytotoxic effect, while chlorhexidine cetrimide (HAC 3.5%), chloramine 0.5% and polyvinylpyrrolidone iodine (PVP-I) > or = 0.05% damaged a large majority of cells. PVP-I at a concentration of > or = 5% was found to be the most effective CTA.     

PMD second-order effects on pulse propagation in single-modeoptical fibers

The time-space varying birefringence in single-mode optical fibers causes the polarization mode dispersion (PMD) to be a serious problem in high bit-rate transmissions. The PMD first- and second-order statistics are well known in the literature, but second-order PMD-induced pulse distortions have still to be clarified for sequences of pulses of arbitrary shape. We give, for the first time, the exact PMD time impulse response, up to second order. We show, both numerically and experimentally, that the effect of the rotation of the principal states of polarization (PSP) is to generate power overshoots on the “1” and “0” bit sequences     

Separate jasmonate-dependent and salicylate-dependent defense-response pathways in Arabidopsis are essential for resistance to distinct microbial pathogens

The endogenous plant hormones salicylic acid (SA) and jasmonic acid (JA), whose levels increase on pathogen infection, activate separate sets of genes encoding antimicrobial proteins in Arabidopsis thaliana. The pathogen-inducible genes PR-1, PR-2, and PR-5 require SA signaling for activation, whereas the plant defensin gene PDF1.2, along with a PR-3 and PR-4 gene, are induced by pathogens via an SA-independent and JA-dependent pathway. An Arabidopsis mutant, coi1, that is affected in the JA-response pathway shows enhanced susceptibility to infection by the fungal pathogens Alternaria brassicicola and Botrytis cinerea but not to Peronospora parasitica, and vice versa for two Arabidopsis genotypes (npr1 and NahG) with a defect in their SA response. Resistance to P. parasitica was boosted by external application of the SA-mimicking compound 2, 6-dichloroisonicotinic acid [Delaney, T., et al. (1994) Science 266, 1247-1250] but not by methyl jasmonate (MeJA), whereas treatment with MeJA but not 2,6-dichloroisonicotinic acid elevated resistance to Alternaria brassicicola. The protective effect of MeJA against A. brassicicola was the result of an endogenous defense response activated in planta and not a direct effect of MeJA on the pathogen, as no protection to A. brassicicola was observed in the coi1 mutant treated with MeJA. These data point to the existence of at least two separate hormone-dependent defense pathways in Arabidopsis that contribute to resistance against distinct microbial pathogens.     

Distortion cancellation of frequency converted pulses with simple linear signal processing and application to frequency modulation to amplitude modulation conversion in high power lasers

It is known that a linear filter may be easily compensated with its inverse transfer function. However, it was shown that this approach could also be valid even for such a complex nonlinear system as frequency conversion. As a matter of fact, it is possible to at least partly precompensate for distortions occurring within, or even downstream from, frequency conversion crystals with a simple linear optical filter set upstream. In this paper, we give the theoretical background and derive the optimum precompensation filter from simple analytical formulas even in the case of saturation. We first show the relevance of our approach for Gaussian pulses: the pulse may be short or not and chirped or not, and the same linear precompensation filter may be used as long as saturation is not reached. We then study the case of phase-modulated pulses, as can be found on high power lasers such as lasers for fusion. We show that previous experimental results are in perfect agreement with these calculations. Finally, justified by our simple analytical formulas, we present a rigorous parametrical study giving the distortion reduction for any second and third harmonic generation system in the case of phase-modulated pulses.     

FM-to-AM conversion in high-power lasers

FM-to-AM conversion is an important issue that could prevent fusion ignition with high-power lasers, such as the Laser MegaJoule (LMJ). We first overview the whole problem of FM-to-AM conversion in high-power lasers and we explain why AM spectral content of FM-to-AM conversion is important, although this information was not used in previous studies. We then propose simple analytical models to simulate FM-to-AM conversion in the LMJ frequency conversion system. We succeed in isolating every cause of spectrum distortion and give, for each of them, FM-to-AM predictions that are in very good agreement with simulations of a complex propagation code. Finally, we show how the last grating filters most of the FM-to-AM conversion. We conclude that the FM-to-AM conversion distortion criterion will be, on LMJ, below 40% in the last optics and 10% on the target.     

Optical duobinary modulation with alleviated phase constraints

Spectral compression is a necessary function for applications in optical communication such as noise reduction, use of bandwidth-limited devices, etc. Ideal duobinary modulation, which allows the reduction of the spectrum bandwidth, requires a π (rad) phase shift between the “-1” and the “+1” logical levels. We show that with a reasonable finite extinction ratio of 10 dB, a phase shift as low as 0.18 π can be used, with a resulting spectrum compression ratio of nearly 2