Quasi-3D Light Con?nement in Double Photonic Crystal Re?ectors VCSELs for CMOS-Compatible Integration Corrado Sciancalepore,Badhise Ben Bakir,Xavier Letartre,Jean-Marc Fedeli,Nicolas Olivier,Damien Bordel, Christian Seassal,Pedro Rojo-Romeo,Member,IEEE,Philippe Regreny,and Pierre Viktorovitch Abstract—A novel architecture of one-dimensional photonic crystal membrane(PCM)re?ectors embodying a heterostructure is proposed as a robust design aimed at a3-D ef?cient con?ne-ment of light in single-mode polarization-controlled 1.55-m vertical-cavity surface-emitting laser(VCSEL)microsources for heterogeneous integration on complementary metal-oxide-semi-conductor(CMOS).On the basis of a theoretical approach,the paper focuses on the deep interweaving between the kinetics of light transport in the mirrors and the physical nature of the ex-ploited Fano resonances.An example of VCSEL design for optical pumping employing heterostructure-con?ned photonic crystal mirrors is presented.The predicted photons kinetics along with the considerable improvement in cavity modal features owing to the enhanced mirror architecture have been con?rmed by per-forming rigorous three-dimensional?nite-difference time-domain (3-D FDTD)calculations.Finally,experimental observations of photoluminescence(PL)emission performed on?rst-ever fabri-cated devices for optical pumping show striking agreement with theoretical considerations and ab initio modelling. Index Terms—Photonic crystals,photons,semiconductor lasers, slow Bloch mode,vertical-cavity surface-emitting lasers(VC-SELs). I.I NTRODUCTION A CHIEVING a full control of photons in the real—reciprocal space as well as in the frequency—time domain is decisive for the design of innovative optical components aimed at further breakthroughs in the?eld of micro-nano-photonics.An ef?cient harnessing of light is made possible by con?ning photons within the tiniest spatial domain (in comparison to the wavelength)for the longest time possible (as compared to the oscillation period),while allowing to be ef?ciently collected(from)or addressed(to)the photonic structures where are meant to be con?ned. Manuscript received December22,2010;revised April19,2011,May05, 2011;accepted May10,2011.Date of publication May23,2011;date of current version June15,2011.This work was supported by the European Commission in the framework of the project HELIOS. C.Sciancalepore is with the Institut des Nanotechnologies,Ecole Centrale de Lyon,F-69134Ecully,France(e-mail:corrado.sciancalepore@ec-lyon.fr) and with the Commissariatàl’énergie Atomique et auxénergies Alternatives, Département Optronique,(CEA-LETI Minatec),F-38054Grenoble,France (e-mail:corrado.sciancalepore@cea.fr). X.Letartre,C.Seassal,P.Rojo-Romeo,P.Regreny,and P.Viktorovitch are with the Institut des Nanotechnologies,Ecole Centrale de Lyon,F-69134Ecully, France. B.Ben Bakir,J.-M.Fedeli,N.Olivier,and D.Bordel are with the Commis-sariatàl’énergie Atomique et auxénergies Alternatives,Département Optron-ique,(CEA-LETI Minatec),F-38054Grenoble,France. Digital Object Identi?er10.1109/JLT.2011.2157303 An effective control of light is highly desirable in the case of laser microcavities and devices for non-linear applications where the need for a stronger light-matter coupling is even more binding.This concept is particularly true in vertical-cavity sur-face-emitting lasers(VCSELs),where the coupling of the op-tical mode with the active material is crucial for low-threshold emitters and modal control constitutes an additional require-ment to be addressed especially for telecommunication-oriented applications.As widely proposed in the literature,while ver-tical con?nement is achieved through diffractive phenomena provided by distributed Bragg re?ectors(DBRs),the lateral op-tical waveguiding(or antiguiding)and modal behaviour in VC-SELs generally relies on a complex interplay of index-[1]and gain-guiding mechanisms[2]. Speci?cally,in arsenide lasers,the enhancement of the pho-tons-matter coupling along with modal selection was accom-plished by introducing an optimized transverse optical and elec-trical con?nement via oxide windows[3]–[7].However,the main drawback of such solutions lies with the ineludible de-sign trade-offs between single-mode operation,low threshold and optical output power[6].Using shallow surface reliefs[5] has only partly addressed the issue given the still considerable drop in emitted power;on the other hand,VCSELs’designs based on external cavity con?gurations and index antiguiding [7]for the suppression of higher order modes are characterized by rather complicated and possibly unstable processing.Con-cerning InP-based VCSELs,while ef?cient carrier funnelling has been obtained by means of structured[8]–[10]or proton-im-planted[11]tunnel junctions,this device class is still waiting for an ef?cient optical con?nement owing to the lack of index guiding[11].Solutions based on the incorporation of2-D pho-tonic crystals(PhCs)in proton-implanted devices[11]are af-fected by signi?cant drawbacks due to light leaking through the photonic crystal holes disrupting laser operation,even calling into question the actual feasibility of the fabrication process. In electro-thermally tunable micro-electro-mechanical sys-tems(MEMS)-VCSELs,curved micro-machined DBR mem-branes have been successfully employed in order to improve the transverse con?nement of the fundamental mode while main-taining a good modal selection[8].Nonetheless,these mirrors are affected by several important disadvantages such as the lack of lateral(typical diameters100m)and longitudinal(cavity physical length15m)compactness,tight epitaxial and fab-rication constraints. Regarding polarization control,different strategies have been so far adopted in VCSELs:elasto-[12]and electro-optic[13], [14]induced birefringence,asymmetric cavity geometries[15] 0733-8724/$26.00?2011IEEE
Quasi-Contract 准合同 1 The term ‘quasi-contract’, once used to describe the area of law now called ‘restitution’ or ‘unjust enrichment’, is now out of favour. ‘Quasi-contract’ says only that the matter is not contract. So far as it suggests that there is a sort of contract, it deceives, unintelligibly. Quasi-contractual liability should be understood not as part of unjust enrichment, but as a different basis of liability that can help us see what liability for unjust enrichment might be: liability grounded in notions of fairness. 1“准合同”这一术语,以前用于描述现在法律领域的“恢复原状”或“不当得利”，现在不合时宜。从字面意义，准合同并非合同。目前为止，它还是一种合同，具有欺骗性，难以理解。准合同责任不应该作为“不当得利”来理解，而是作为责任的基础，这有助于我们明白不当得利的责任基础，即基于公平正义观念的责任。 2 The notion of quasi-contract can help us understand what is at stake. whether to impose liability in certain circumstances in which no contract has been made between the parties but when we have good reason to believe that such a contract would have been made if the parties had had the opportunity to do so. This analysis is more fitting for these cases because by trying to find what the parties would have contracted for, it adopts an ex ante perspective. Interestingly, once again we see that commentators who reject the quasi-contractual analysis end up explaining the situation by invoking contractual concepts. For example, in explaining why liability should be imposed only on successful attempts, Burrows writes: ‘A reasonable man would surely pay for someone to try to rescue his drowning daughter or to try to save his burning house’. Burrows comes close to stating the quasi-contractual rationale for imposing liability: the reason why liability should be imposed in such cases is because people would have been willing to pay for the service (even without the guarantee of success), if they had had the opportunity to do so. 2准合同观念能帮助我们理解什么才是利害攸关所在。在某些情形下，双方当事人之间没有订立合同，但是当我们有充足的理由相信如果双方有条件订立合同，他们之间的合同已经订立，以此可以强加双方相应的责任。从事前的角度看，这种解释更适合发现当事人双方最初的合同目的。有趣的是，我们再次看到拒绝准合同分析的评论家通过调用合同概念的做法终结对准合同的解释。例如，在解释为什么强制责任应该只在成功的事例中实施，Burrows写道：“一个理性的人肯定愿意为成功救助他溺水的女儿或成功挽救他着火的房子支付对价”。Burrows进一步说明准合同强制责任的理论基础：如果在条件允许的前提下，人们愿意为服务支付对价（尽管没有成功保证），所以在这些情形下必须有强制责任。 3 Within a quasi-contractual analysis it is not difficult to explain why liability need not be limited to successful attempts. In many contracts for service, the service provider does not promise a certain result, only a certain degree of effort. If the promisor fulfils her contractual liability by performing to that level, she does not breach her contractual obligation even if the service she provides does not match a certain desired outcome. By contrast, in principle, if the promisor fails to perform to the same degree required by the contract, she breaches the contract even if the non-contracted yet desired outcome is achieved. 3在准合同分析的范围内，不难解释为什么强制责任不必要限于成功的事例。在很多多服务合同,服务提供方不承诺明确的合同结果,只有某种合同行为。如果义务人履行合同责任达到了约定的承担责任的水平,那么义务人就没有违反合同义务，即使义务方提供的服务不符合某种预期的结果。相比之下，原则上，如果义务人没有履行达到合同预期的行为，即使非合同预期的目的达到了，义务方也是违反了合同的。
Fearon J D1991Counterfactuals and hypothesis testing in political science.World Politics43:169–95 Goldberger A1983Abnormal selection bias.In:Karlin S, Amemyia T,Goodman L A(eds.)Studies in Econometrics, Time Series,and Multi ariate Statistics.Academic Press,New York Goldstein H1987Multile el Models in Educational and Social Research.Oxford University Press,New York Heckman J J1988The microeconomic evaluation of social programs and economic institutions.In:Chung-Hua Series of Lectures by In ited Eminent Economists no.14.The Institute of Economics,Academia Sinica,Teipei,China Hsiao C1986Analysis of Panel Data.Cambridge University Press,Cambridge,UK Isaac L W,Gri?n L J1989A historicism in time-series analyses of historical process.American Sociological Re iew54: 873–90 Jackman R1986The politics of economic growth in industrial democracies,1974–1980:Leftist strength or North Sea oil? Journal of Politics49:242–56 Janoski T,Hicks A(eds.)1994The Comparati e Political Economy of the Welfare State.Cambridge University Press, Cambridge,UK Lange P,Jarrett G1985The politics of growth.Journal of Politics47:792–827 Maddala G S1983Limited-dependent and Qualitati e Variables in Econometrics.Cambridge University Press,Cambridge, UK Nickell S1997Unemployment and labor market rigidities: Europe versus North America.Journal of Economic Per-specti es11(3):55–74 Nieuwbeerta P,Ultee W1999Class voting in Western industrial-ized countries,1945–1990:systematizing and testing explan-ations.European Journal of Political Research35:123–60 Przeworski A,Alvarez M R,Cheibub J A,Limongi F in press Democracy and De elopment:Political Regimes and Material Welfare in the World,1950–1990.Cambridge Uni-versity Press,New York Ragin C1987The Comparati e Method.University of California Press,Berkeley,CA Ragin C1994A qualitative comparative analysis of pension systems.In:Janoski T,Hicks A M(eds.)The Comparati e Political Economy of the Welfare State.Cambridge University Press,Cambridge,UK,pp.320–45 Stimson J A1985Regression in time and space:A statistical essay.American Journal of Political Science29:914–47 Western B1998a Causal heterogeneity in comparative research: A Bayesian hierarchical modeling approach.American Journal of Political Science42:1233–59 Western B1998b Between Class and Market:Postwar Unioniza-tion in the Capitalist Democracies.Princeton University Press, Princeton,NJ G.Esping-Andersen and A.Przeworski Quasi-Experimental Designs The theory of quasi-experimentation addresses one of the great unanswered questions in the social and behavioral sciences:how can researchers make a valid causal inference when units cannot be randomly assigned to conditions?Like all experiments,quasi-experiments deliberately manipulate presumed causes to discover their e?ects.However,quasi-experiments are also de?ned by what they do not do:the researcher does not assign units to conditions randomly.Instead, quasi-experiments use a combination of design features,practical logic,and statistical analysis to show that the presumed cause is likely to be responsible for the observed e?ect,and other causes are not.The term nonrandomized experiment is synonymous with quasi-experiment;and the terms observational study and nonexperimental design often include quasi-experiments as a subset.This article discusses the need for quasi-experimentation,describes the kinds of designs that fall into this class of methods,reviews the intellectual and practical history of these designs,and notes important current developments. 1.The Need for Quasi-experimentation Given the desirable properties of randomized experi-ments,one might question why quasi-experiments are needed.When properly implemented,randomized experiments yield unbiased estimates of treatment e?ects,accompanied by known probabilities of error in identifying e?ect size.Quasi-experimental designs do not have these properties.Yet quasi-experiments are necessary in the arsenal of science because it is not always possible to randomize.Ethical constraints may preclude withholding treatment from needy people based on chance,those who administer treatment may refuse to honor randomization,or questions about program e?ects may arise after a treatment was already implemented so that randomization is im-possible.Consequently,the use of quasi-experimental designs is frequent and inevitable in practice. 2.Kinds of Quasi-experimental Designs The range of quasi-experimental designs is large, including but not limited to:(a)Nonequivalent control group designs in which the outcomes of two or more treatment or comparison conditions are studied but the experimenter does not control assignment to conditions;(b)Interrupted time series designs in which many consecutive observations over time(proto-typically100)are available on an outcome,and treatment is introduced in the midst of those observa-tions to demonstrate its impact on the outcome through a discontinuity in the time series after treat-ment;(c)Regression discontinuity designs in which the experimenter uses a cuto?score on a measured variable to determine eligibility for treatment,and 12655 Quasi-Experimental Designs
用R语言做数据分析——泊松回归 当通过一系列的连续型或类别型预测变量来预测计数 型结果变量时，泊松回归是一个非常有用的工具。为阐述泊松回归模型的拟合过程，并探讨一些可能出现的问题，我们将使用robust包中Breslow癫痫数据。特别地，我们将讨论在治疗初期的八周内，抗癫痫药物对癫痫发病数的影响。我们就遭受轻微或严重间歇性癫痫的病人的年龄和癫痫发病 数收集了数据，包含病人被随机分配到药物组或者安慰剂组前八周和随机分配后八周两种情况。响应变量sumY（随机化后八周癫痫发病数），预测变量为治疗条件（Trt）、年龄（Age）和前八周内的基础癫痫发病数（Base）。之所以包含基础癫痫发病数和年龄，是因为它们对响应变量有潜在影响。在解释这些协变量后，我们感兴趣的是药物治疗是否能减少癫痫发病数。首先，看看数据集的统计汇总信息：注意，虽然数据集有12个变量，但是我们只关注之前描述的四个变量。基础和随机化后的癫痫发病数都有很高的偏度。现在，我们更详细的考察响应变量，如下代码可生成的图形如下图所示：从上图可以清楚地看到因变量的偏移特性及可能的离群点。初看图形，药物治疗下癫痫发病数似乎变小了，且方差也变小了（泊松分布中，较小的方差伴随着较小的均值）。与标准最小二乘回归不同，泊松分布并不关注方差异质性。
接下来拟合泊松回归：输出结果列出了偏差、回归参数、标准误差和参数为0的检验。除以，这里的预测变量在p解释模型参数使用coef()函数可获得模型系数，或者调用summary()函数的输出结果中的Coefficients表格：Φ在泊 松回归中，因变量以条件均值的堆属性是ln(λ)来建模。年龄的回归参数为0.0227，表明保持其他预测变量不变，年龄增加一岁，癫痫发病数的对数平均值将相应增加0.03。截距项即当预测变量都为0时，癫痫发病数的对数平均值。由于不可能为0岁，且调查对象的基础癫痫发病数均不为0，因此截距项没有任何意义。通常在因变量的初始尺度（癫痫发病数、而非发病数的对数）上解释回归系数比较容易。为此，指数化系数：现在可以看到，保持其他变量不变，年龄增加一岁，期望的癫痫发病数将乘以1.023。这意味着年龄的增 加与较高的癫痫发病数相关联。更重要的是，一单位Trt的 变化（即从安慰剂到治疗组），期望的癫痫发病数将乘以0.86，也就是说，保持基础癫痫发病数和年龄不变，服药组相对于安慰剂组发病数降低了20%。另外需要牢记的是，与Logistic 回归中的指数化参数相似，泊松模型中的指数化参数对响应变量的影响都是成倍增加的，而不是线性相加。同样，我们还需要评价波形模型的过度离势。过度离势泊松分布的方差与均值相等。当响应变量观测的方差比一句泊松分布预测的方差大时，泊松回归可能发生过度离势。由于处理计数型数
Quasistationarity in populations that are subject to large-scale mortality or emigration P.K.Pollett Department of Mathematics The University of Queensland Queensland4072 Australia Phone:(+617)33653459 Fax:(+617)33651477 Email:pkp@https://www.doczj.com/doc/667878930.html,.au Abstract We shall examine a model,?rst studied by Brockwell et al.(1982),which can be used to describe the long-term behaviour of populations that are subject to catastrophic mor-tality or emigration events.Populations can suffer dramatic declines when disease,such as an introduced virus,affects the population,or when food shortages occur,due to overgrazing or?uctuations in rainfall.However,perhaps surprisingly,such populations can survive for long periods and,although they may eventually become extinct,they can exhibit an appar-ently stationary regime.It is useful to be able to model this behaviour.This is particularly true of the ecological examples which motivated the present study,since,in order to properly manage these populations,it is necessary to be able to predict persistence times and to esti-mate the conditional probability distribution of population size.We shall see that although our model predicts eventual extinction,the time till extinction can be long and the stationar-ity exhibited by these populations over any reasonable time scale can be explained using a quasistationary distribution.
October 2012 Quasi-Resonant Controller Features Burst mode operation at light load for low standby power consumption Control the number of valley skip by detecting on-off width of MOSFET (FA564x series) Overload protection compensation by detecting input voltage (FA564x series) Linear frequency reduction with decreasing load (FA5572/73/74) Built-in soft-start with fixed time Leading-Edge Blanking Auto restart mode for VCC Undervoltage protection Latch-off mode for Overvoltage protection Latch-off mode for external circuit Auto restart mode for Overload protection (FA564x series) Built-in External Latch-off mode Latch-off mode for Short-winding protection (FA5571/72/73/74 and FA5643)mode controller. The active burst mode operation enables low standby power consumption due to reducing switching loss of Power MOSFET. FA564x series provides low audible noise switching power supply by the fixed mode valley skip operation detecting on-off pulse width of MOSFET. Because adjust overcurrent protection to two levels automatically by detecting input voltage, the FA564x series is suitable for wide-range power supply. Fuji's Quasi-Resonant controller is pin compatibility regardless of the generation, therefore replacement from previous generation is easy. Description Applications
C H A P T E R6 Quasi-Newton Methods In the mid1950s,W.C.Davidon,a physicist working at Argonne National Laboratory, was using the coordinate descent method(see Section9.3)to perform a long optimization calculation.At that time computers were not very stable,and to Davidon’s frustration, the computer system would always crash before the calculation was?nished.So Davidon decided to?nd a way of accelerating the iteration.The algorithm he developed—the?rst quasi-Newton algorithm—turned out to be one of the most creative ideas in nonlinear optimization.It was soon demonstrated by Fletcher and Powell that the new algorithm was much faster and more reliable than the other existing methods,and this dramatic