外文文献-The basic of description of android system对Android系统的基本描诉

合集下载

英文参考文献标准字体.doc

英文参考文献标准字体参考文献是在学术研究过程中，对某一著作或论文的整体的参考或借鉴。

下文是英文参考文献标准字体，希望对你有所帮助。

英文参考文献标准格式【1】一、参考文献的类型参考文献(即引文出处)的类型以单字母方式标识，具体如下：--专著,著作--论文集(一般指会议发表的论文续集，及一些专题论文集，如《***大学研究生学术论文集》-- 报纸文章--期刊文章：发表在期刊上的论文，尽管有时我们看到的是从网上下载的(如知网)，但它也是发表在期刊上的，你看到的电子期刊仅是其电子版--学位论文：不区分硕士还是博士论文--报告：一般在标题中会有关于****的报告字样-- 标准--专利--文章：很少用，主要是不属于以上类型的文章--对于不属于上述的文献类型，可用字母z标识，但这种情况非常少见常用的电子文献及载体类型标识：--联机网上数据(database online)--磁带数据库(database on magnetic tape)--光盘图书(monograph on cdrom)--磁盘软件(computer program on disk)--网上期刊(serial online)--网上电子公告(electronic bulletin board online)很显然，标识的就是该资源的英文缩写，/前面表示类型，/后面表示资源的载体，如ol表示在线资源二、参考文献的格式及举例1.期刊类【格式】作者.篇名.刊名,出版年份,卷号(期号)起止页码.【举例】周融，任志国，杨尚雷，厉星星.对新形势下毕业设计管理工作的思考与实践.电气电子教学学报，2003(6):107-109.夏鲁惠.高等学校毕业设计(论文)教学情况调研报告.高等理科教育,2004(1):46-52.heider, e.r. d.c.oliver. the structure of color space in naming and memory of two languages . foreign language teaching and research, 1999, (3): 62 67.2.专著类【格式】作者.书名.出版地:出版社,出版年份:起止页码.【举例】刘国钧,王连成.图书馆史研究.北京:高等教育出版社,1979:15-18,31.gill, r. mastering english literature . london: macmillan, 1985: 42-45.3.报纸类【格式】作者.篇名.报纸名,出版日期(版次).【举例】李大伦.经济全球化的重要性. 光明日报，1998-12-27(3).french, w. between silences: a voice from china. atlantic weekly, 1987-8-15(33).4.论文集【格式】作者.篇名.出版地:出版者,出版年份:起始页码.【举例】伍蠡甫.西方文论选. 上海:上海译文出版社,1979:12-17.spivak,g. can the subaltern speak?. in c.nelson l. grossberg(eds.). victory in limbo: imigism . urbana: university of illinois press, 1988, pp.271-313.almarza, g.g. student foreign language teachers knowledge growth . in d.freeman and j.c.richards (eds.). teacher learning in language teaching . new york: cambridge university press. 1996. pp.50-78.5. 学位论文【格式】作者.篇名.出版地:保存者,出版年份:起始页码.【举例】张筑生.微分半动力系统的不变集.北京:北京大学数学系数学研究所, 1983:1-7.6.研究报告【格式】作者. 篇名.出版地:出版者,出版年份:起始页码.【举例】冯西桥.核反应堆压力管道与压力容器的lbb分析.北京:清华大学核能技术设计研究院, 1997:9-10.7.专利【格式】专利所有者.题名.国别:专利号,发布日期.【举例】姜锡洲.一种温热外敷药制备方案.中国专利:881056073, 1989 07 26.8.标准【格式】标准编号,标准名称.【举例】gb/t 16159-1996, 汉语拼音正词法基本规则.9.条例【格式】颁布单位.条例名称.发布日期【举例】中华人民共和国科学技术委员会.科学技术期刊管理办法.1991-06-0510.电子文献【格式】主要责任者.电子文献题名.电子文献出处.或可获得地址，发表或更新日期/引用日期.【举例】王明亮.关于中国学术期刊标准化数据库系统工程的进展.http: ///pub/wml.txt/980810 2.html, 1998 08 16/1998 10 04.万锦.中国大学学报论文文摘(1983 1993).英文版. 北京: 中国大百科全书出版社, 1996.11.各种未定义类型的文献【格式】主要责任者.文献题名.出版地:出版者, 出版年.特别说明：凡出现在参考文献项中的标点符号都失去了其原有意义，且其中所有标点必须是半角,如果你的输入法中有半角/全解转换,则换到半角状态就可以了,如果你的输入法中没有这一转换功能,直接关闭中文输入法,在英文输入状态下输入即可.其实，很多输入法(如目前比较流行的搜狐输入法)都提供了四种组合：(1)中文标点+ 全角：这时输入的标点是这样的，：【1】-(而这时，我没有找到哪个键可以输入/ 符号)也就是说，这些符号是一定不能出现在参考文献中的;(2) 中文标点+半角：这时输入的标点是这样的，：【1】-(这时，我还是没有找到哪个键可以输入/ 符号)也就是说，这些符号也不能出现在参考文献中的;上面列出的符号，中间没有任何的空格，你能看出它们有什么区别吗?我看只是-的宽度有一点点不同，其它都一样(3)英文标点+全角：这时输入的标点是这样的，.：-/(4)英文标点+半角：这时输入的标点是这样的,.:-/从这两项可以明显的看出，半角和全角其实最大的差别是所占的宽度不一样，这一点对于数字来说最为明显，而英文标点明显要比中文标点细小很多(也许因为英文中，标点的功能没有中文那么复杂，就是说英文中标点符号的能力没有中文那么强大)所以，很多人在写参考文献时，总是觉得用英文标点+半角很不清楚，间距也太小，其实这点完全不用担心如果你觉得真的太小不好看，就用英文标点+全角吧而在之后，一般也都有一个空格对于英文参考文献，还应注意以下两点：①作者姓名采用姓在前名在后原则，具体格式是：姓,名字的首字母. 如：malcolm richard cowley 应为：cowley, m.r.，如果有两位作者，第一位作者方式不变，之后第二位作者名字的首字母放在前面，姓放在后面，如：frank norris 与irving gordon应为：norris, f. i.gordon.②书名、报刊名使用斜体字，如：mastering english literature,english weekly.三、注释注释是对论文正文中某一特定内容的进一步解释或补充说明注释应置于本页页脚，前面用圈码①、②、③等标识英文论文参考文献格式写法【2】learning advice centreyour in the author/date system) and, for law students or in dissertations,the numeric system. different subject areas use slight variations of these systems (and other systems do exist) so you must consult your course/module handbook for clarification of the specific conventions that you are expected to use within your subject area(s).references need to appear in two places:1. within the body of your writing include: authors surname, year of publication and, if quoting, the exact page number from which the quote is taken;2. in the bibliography include at least: authors surname, initial(s), year of publication, title of text/chapter, edition, publisher place of publication; see page 3 for specific details that apply to referencing different sources. there are two ways in which you can reference, or cite, another persons work: a) by paraphrasing; this shows you have fully interpreted what you have read - see learning skills help sheet on how to paraphrase;1.1 references within the body of the text - harvard systemb) by quoting directly; follow with a comment to show relevance/understanding if the direct quotation is more than two lines, you should indent it as a separate paragraph e.g.:as cottrell (2003, p.148) points out“our views of what is intelligent can prevent us from developing our minds to their full potential. people who feel they are not very bright ornot very creative probably will fulfil that estimation of themselves. on the other hand, positive thinking and constructive mental activity develop the mind.”citing secondary sources:if you are referencing a work cited by the author of the text you are reading, you should cite the original work as being within a secondary source i.e. what what the writer of the book you are reading has read, using the following method:according to de bono (1994, cited in cottrell, 2003, p.148), “clever people are often hampered by their apparent intelligence in two ways:” they are good at arguing and defending their point of view; they indulge in negative criticism which is a quick, easy and dramatic way of proving someone wrong.contdyou can make an appointment with the learning advisor if you wish to receive more individual advice on your independent learning skills. email or phone 020 7911 5000 ext. 2364.learning advice centrereference appears as a footnote at the bottom of the page e.g.: there were many changes in the british diet in the period after 1870. fruit became more common, especially in the form of fruit jam. even the fish-and-chip shop dates from the same period.1(note: see bottom of this page for associated footnote.)to insert footnotes using word, click on the placein your document where you want to insert thefootnote.then click on the insert menu, select reference,then click on footnote.another window will then open that will allow you to format the footnotes manually, if you require.once the footnote has been inserted in the text and it appearsat the bottom of the page, click next to the number at the。

fundamentals of microelectronics英文原版

The fundamentals of microelectronics refer to the basic principles and concepts that form the foundation of the field. Microelectronics deals with the study and application of small-scale electronic components, such as integrated circuits and transistors. This field has played a crucial role in the development of various technologies, including computers, smartphones, and medical devices.One of the key concepts in microelectronics is the idea of miniaturization. Microelectronic components are designed to be small and compact, allowing for increased functionality in a limited space. This miniaturization is made possible by advancements in semiconductor technology, which enables the production of smaller and more efficient electronic devices.Another fundamental principle is the understanding of electronic circuits. Microelectronics relies on the design and analysis of circuits that control the flow of electric current. These circuits can be composed of different components, such as resistors, capacitors, and inductors, which work together to perform specific tasks.The behavior of microelectronic devices is guided by the laws of physics, particularly quantum mechanics. At the nanoscale level, where microelectronics operates, particles exhibit quantum effects that can significantly impact the performance of electronic devices. Understanding these effects is essential for designing and optimizing microelectronic components.In addition to the physical principles, microelectronics also encompasses the study of fabrication techniques. The process of manufacturing microelectronic devices involves multiple steps, including deposition, lithography, etching, and doping. Each of these steps contributes to the creation of complex integrated circuits and other microelectronic components.The field of microelectronics also includes the study of electronic materials. Different materials exhibit unique properties that can be leveraged in microelectronic devices. For example, semiconductors, such as silicon, are widely used in microelectronics due to their ability to control the flow of electric current.Overall, the fundamentals of microelectronics cover a wide range of topics, including circuit design, semiconductor physics, fabrication techniques, and electronic materials. Understanding these principles is crucial for the development of new and innovative microelectronic devices that drive technological advancements in various industries.。

科学文献

Arnaud DoucetEngineering Dept.Cambridge University ad2@Nando de FreitasKevin MurphyStuart RussellComputer Science Dept.UC Berkeleyjfgf,murphyk,russell @AbstractParticle ﬁlters (PFs)are powerful sampling-based inference/learning algorithms for dynamic Bayesian networks (DBNs).They allow us to treat,in a principled way,any type of probabil-ity distribution,nonlinearity and non-stationarity.They have appeared in several ﬁelds under such names as “condensation”,“sequential Monte Carlo”and “survival of the ﬁttest”.In this pa-per,we show how we can exploit the structure of the DBN to increase the efﬁciency of parti-cle ﬁltering,using a technique known as Rao-Blackwellisation.Essentially,this samples some of the variables,and marginalizes out the rest exactly,using the Kalman ﬁlter,HMM ﬁlter,junction tree algorithm,or any other ﬁnite di-mensional optimal ﬁlter.We show that Rao-Blackwellised particle ﬁlters (RBPFs)lead to more accurate estimates than standard PFs.We demonstrate RBPFs on two problems,namely non-stationary online regression with radial ba-sis function networks and robot localization and map building.We also discuss other potential ap-plication areas and provide references to some ﬁ-nite dimensional optimal ﬁlters.1INTRODUCTIONState estimation (online inference)in state-space models is widely used in a variety of computer science and engineer-ing applications.However,the two most famous algorithms for this problem,the Kalman ﬁlter and the HMM ﬁlter,are only applicable to linear-Gaussian models and models with ﬁnite state spaces,respectively.Even when the state space is ﬁnite,it can be so large that the HMM or junction tree algorithms become too computationally expensive.This is typically the case for large discrete dynamic Bayesian net-works (DBNs)(Dean and Kanazawa 1989):inference re-quires at each time space and time that is exponential in thenumber of hidden nodes.To handle these problems,sequential Monte Carlo meth-ods,also known as particle ﬁlters (PFs),have been in-troduced (Handschin and Mayne 1969,Akashi and Ku-mamoto 1977).In the mid 1990s,several PF algorithms were proposed independently under the names of Monte Carlo ﬁlters (Kitagawa 1996),sequential importance sam-pling (SIS)with resampling (SIR)(Doucet 1998),bootstrap ﬁlters (Gordon,Salmond and Smith 1993),condensation trackers (Isard and Blake 1996),dynamic mixture models (West 1993),survival of the ﬁttest (Kanazawa,Koller and Russell 1995),etc.One of the major innovations during the 1990s was the inclusion of a resampling step to avoid de-generacy problems inherent to the earlier algorithms (Gor-don et al.1993).In the late nineties,several statistical im-provements for PFs were proposed,and some important theoretical properties were established.In addition,these algorithms were applied and tested in many domains:see (Doucet,de Freitas and Gordon 2000)for an up-to-date sur-vey of the ﬁeld.One of the major drawbacks of PF is that sampling in high-dimensional spaces can be inefﬁcient.In some cases,however,the model has “tractable substructure”,which can be analytically marginalized out,conditional on cer-tain other nodes being imputed,c.f.,cutset conditioning in static Bayes nets (Pearl 1988).The analytical marginal-ization can be carried out using standard algorithms,such as the Kalman ﬁlter,the HMM ﬁlter,the junction tree al-gorithm for general DBNs (Cowell,Dawid,Lauritzen and Spiegelhalter 1999),or,any other ﬁnite-dimensional opti-mal ﬁlters.The advantage of this strategy is that it can drastically reduce the size of the space over which we need to sample.Marginalizing out some of the variables is an example of the technique called Rao-Blackwellisation ,because it is related to the Rao-Blackwell formula:see (Casella and Robert 1996)for a general discussion.Rao-Blackwellised particle ﬁlters (RBPF)have been applied in speciﬁc con-texts such as mixtures of Gaussians (Akashi and Ku-mamoto 1977,Doucet 1998,Doucet,Godsill and Andrieu2000),ﬁxed parameter estimation(Kong,Liu and Wong 1994),HMMs(Doucet1998,Doucet,Godsill and Andrieu 2000)and Dirichlet process models(MacEachern,Clyde and Liu1999).In this paper,we develop the general theory of RBPFs,and apply it to several novel types of DBNs.We omit the proofs of the theorems for lack of space:please refer to the technical report(Doucet,Gordon and Krishna-murthy1999).2PROBLEM FORMULATIONLet us consider the following general state space model/DBN with hidden variables and observed vari-ables.We assume that is a Markov process of ini-tial distribution and transition equation. The observations are assumedto be conditionally independent given the process of marginal distribution.Given these observations, the inference of any subset or property of the statesrelies on the joint posterior distribution .Our objective is,therefore,to estimate this distribution,or some of its characteristics such as theﬁlter-ing density or the minimum mean square error (MMSE)estimate.The posterior satisﬁes the following recursionThe problem of how to automatically identify which vari-ables should be sampled,and which can be handled analytically, is one we are currently working on.We anticipate that algorithms similar to cutset conditioning(Becker,Bar-Yehuda and Geiger 1999)might prove useful.the alternative recursionThis estimate is unbiased and,from the strong law of large numbers(SLLN),One way to estimate and con-sists of using the well-known importance sampling method (Bernardo and Smith1994).This method is based on the following observation.Let us introduce an arbitrary impor-tance distribution,from which it is easy to get samples,and such that implies.ThenGiven i.i.d.samples distributed accord-ing to,a Monte Carlo estimate ofis given bywhere the normalized importance weights are equal to Intuitively,to reach a given precision,will require a reduced number of samples over as we only need to sample from a lower-dimensional distribution.This is proven in the following propositions.Proposition1The variances of the importance weights, the numerators and the denominators satisfy for anyvar varvar varvar varA sufﬁcient condition for to satisfy a CLT is varand for any(Bernardo and Smith1994).This trivially implies that also satis-ﬁes a CLT.More precisely,we get the following result. Proposition2Underthe assumptions given above,and satisfy a CLTwhere,and being given byThe Rao-Blackwellised estimate is usually compu-tationally more extensive to compute than so it is of interest to know when,for aﬁxed computational com-plexity,one can expect to achieve variance reduction.Onehasso that,accordingly to the intuition,it will be worth gen-erally performing Rao-Blackwellisation when the average conditional variance of the variable is high.4RAO-BLACKWELLISED PARTICLE FILTERSGiven particles(samples)at time,approximately distributed according to the distribution,RBPFs allow us to compute particles approximately distributed according to the posterior,at time.This is ac-complished with the algorithm shown below,the details of which will now be explained.For,sample:and set:For,evaluate the importanceweights up to a normalizing constant:Multiply/suppress samples with high/lowimportance weights,respectively,to obtainrandom samples approximately distributedaccording to.3.MCMC step4.1IMPLEMENTATION ISSUES4.1.1Sequential importance samplingIf we restrict ourselves to importance functions of the fol-lowing form(3) we can obtain recursive formulas to evaluateand thus.The“incremental weight”is given byand the associated importance weight is Unfortunately,computing the optimal importance sampling distribution is often too expensive.Several deterministic approximations to the optimal distribution have been pro-posed,see for example(de Freitas1999,Doucet1998). Degeneracy of SISThe following proposition shows that,for importance func-tions of the form(3),the variance of can only in-crease(stochastically)over time.The proof of this propo-sition is an extension of a Kong-Liu-Wong theorem(Konget al.1994,p.285)to the case of an importance function of the form(3).Proposition4The unconditional variance(i.e.with the observations being interpreted as random variables) of the importance weights increases over time.In practice,the degeneracy caused by the variance increase can be observed by monitoring the importance weights. Typically,what we observe is that,after a few iterations, one of the normalized importance weights tends to1,while the remaining weights tend to zero.4.1.2Selection stepTo avoid the degeneracy of the sequential importance sam-pling simulation method,a selection(resampling)stage may be used to eliminate samples with low importance ra-tios and multiply samples with high importance ratios.Aselection scheme associates to each particle a num-ber of offsprings,say,such that. Several selection schemes have been proposed in the lit-erature.These schemes satisfy,but their performance varies in terms of the variance of the particles,var.Recent theoretical results in(Crisan, Del Moral and Lyons1999)indicate that the restrictionis unnecessary to obtain convergence re-sults(Doucet et al.1999).Examples of these selection schemes include multinomial sampling(Doucet1998,Gor-don et al.1993,Pitt and Shephard1999),residual resam-pling(Kitagawa1996,Liu and Chen1998)and stratiﬁed sampling(Kitagawa1996).Their computational complex-ity is.4.1.3MCMC stepAfter the selection scheme at time,we obtain par-ticles distributed marginally approximately according to .As discussed earlier,the discrete nature of the approximation can lead to a skewed importance weights distribution.That is,many particles have no offspring (),whereas others have a large number of off-spring,the extreme case being for a particular value.In this case,there is a severe reduction in the di-versity of the samples.A strategy for improving the re-sults involves introducing MCMC steps of invariant distri-bution on each particle(Andrieu,de Freitas and Doucet1999b,Gilks and Berzuini1998,MacEachern et al. 1999).The basic idea is that,by applying a Markov tran-sition kernel,the total variation of the current distribution with respect to the invariant distribution can only decrease. Note,however,that we do not require this kernel to be er-godic.4.2CONVERGENCE RESULTSLet be the space of bounded,Borel measurable functions on.We denote.The fol-lowing theorem is a straightforward consequence of Theo-rem1in(Crisan and Doucet2000)which is an extension of previous results in(Crisan et al.1999).Theorem5If the importance weights are upper bounded and if one uses one of the selection schemes de-scribed previously,then,for all,there exists inde-pendent of such that for anywhere the expectation is taken w.r.t.to the randomness in-troduced by the PF algorithm.This results shows that,un-der very lose assumptions,convergence of this general par-ticleﬁltering method is ensured and that the convergence rate of the method is independent of the dimension of the state-space.However,usually increases exponentially with time.If additional assumptions on the dynamic sys-tem under study are made(e.g.discrete state spaces),it is possible to get uniform convergence results(for any)for theﬁltering distribution.We do not pursue this here.5EXAMPLESWe now illustrate the theory by brieﬂy describing two ap-plications we have worked on.5.1ON-LINE REGRESSION AND MODELSELECTION WITH NEURAL NETWORKS Consider a function approximation scheme consisting of a mixture of radial basis functions(RBFs)and a linear regression term.The number of basis functions,,their centers,,the coefﬁcients(weights of the RBF centers plus regression terms),,and the variance of the Gaussian noise on the output,,can all vary with time,so we treat them as latent random variables:see Figure1.For details, see(Andrieu,de Freitas and Doucet1999a).In(Andrieu et al.1999a),we show that it is possible to simulate,and with a particleﬁlter and to com-pute the coefﬁcients analytically using Kalmanﬁlters. This is possible because the output of the neural network is linear in,and hence the system is a conditionally lin-ear Gaussian state-space model(CLGSSM),that is it is a linear Gaussian state-space model conditional upon the lo-cation of the bases and the hyper-parameters.This leads to an efﬁcient RBPF that can be combined with a reversible jump MCMC algorithm(Green1995)to select the numberFigure1:DBN representation of the RBF model.The hyper-parameters have been omitted for clarity.Figure2:The top plot shows the one-step-ahead output predictions[—]and the true outputs[]for the RBF model.The middle and bottom plotsshow the true val-ues and estimates of the model order and noise variance respectively.of basis functions online.For example,we generated some data from a mixture of2RBFs for,and then from a single RBF for;the method was able to track this change,as shown in Figure2.Further experiments on real data sets are described in(Andrieu et al.1999a).5.2ROBOT LOCALIZATION AND MAPBUILDINGConsider a robot that can move on a discrete,two-dimensional grid.Suppose the goal is to learn a map of the environment,which,for simplicity,we can think of as a matrix which stores the color of each grid cell,which can be either black or white.The difﬁculty is that the color Figure3:A Factorial HMM with3hidden chains. represents the color of grid cell at time,represents the robot’s location,and the current observation. sensors are not perfect(they may accidentallyﬂip bits),nor the motors(the robot may fail to move in the desired di-with some probability due e.g.,to wheel slippage).it is easy for the robot to get lost.And when robot is lost,it does not know what part of the matrix to So we are faced with a chicken-and-egg situation: robot needs to know where it is to learn the map,butto know the map toﬁgure out where it is.problem of concurrent localization and map learn-for mobile robots has been widely studied.In(Mur-2000),we adopt a Bayesian approach,in which wea belief state over both the location of the robot,,and the color of each grid cell,,,where is the number cells,and is the number of colors.The DBN we using is shown in Figure3.The state space has size .Note that we can easily handle changing envi-since the map is represented as a random vari-unlike the more common approach,which treats the map as aﬁxed parameter.The observation model is,where is a function thatﬂips its binary argument with someﬁxed probability.In other words,the robot gets to see the color of the cell it is currently at,corrupted by noise:is a noisy multiplexer with acting as a“gate”node.Note that this conditional independence is not obvious from the graph structure in Figure3(a),which suggests that all the nodes in each slice should be correlated by virtue of sharing a common observed child,as in a factorial HMM(Ghahra-mani and Jordan1997).The extra independence informa-tion is encoded in’s distribution,c.f.,(Boutilier,Fried-man,Goldszmidt and Koller1996).The basic idea of the algorithm is to sample with a PF, and marginalize out the nodes exactly,which can be done efﬁciently since they are conditionally independent given:Some results on a simple one-dimensional grid world aretime tg r i d c e l l i123456780.10.20.30.40.50.60.70.80.91g r i d c e l l iProb. location, i.e., P(L(t)=i | y(1:t)), 50 particles, seed 1246810121416123456780.10.20.30.40.50.60.70.80.91g r i d c e l l iBK Prob. location, i.e., P(L(t)=i | y(1:t))246810121416123456780.10.20.30.40.50.60.70.80.91a b cFigure 4:Estimated position as the robot moves from cell 1to 8and back.The robot “gets stuck”in cell 4for two steps in a row on the outgoing leg of the journey (hence the double diagonal),but the robot does not realize this until it reaches the end of the “corridor”at step 9,where it is able to relocalise.(a)Exact inference.(b)RBPF with 50particles.(c)Fully-factorised BK.shown in Figure 4.We compared exact Bayesian infer-ence with the RBPF method,and with the fully-factorised version of the Boyen-Koller (BK)algorithm (Boyen and Koller 1998),which represents the belief state as a product of marginals:We see that the RBPF results are very similar to the ex-act results,even with only 50particles,but that BK getsconfused because it ignores correlations between the map cells.We have obtained good results learning a map (so the state space has size )using only 100particles (the observation model in the 2D case is that therobot observes the colors of all the cells in aneighbor-hood centered on its current location).For a more detailed discussion of these results,please see (Murphy 2000).5.3CONCLUSIONS AND EXTENSIONSRBPFs have been applied to many problems,mostly in the framework of conditionally linear Gaussian state-space models and conditionally ﬁnite state-space HMMs.That is,they have been applied to models that,conditionally upon a set of variables (imputed by the PF algorithm),admit a closed-form ﬁltering distribution (Kalman ﬁlter in the con-tinuous case and HMM ﬁlter in the discrete case).One can also make use of the special structure of the dynamic model under study to perform the calculations efﬁciently using the junction tree algorithm.For example,if one had evolv-ing trees,one could sample the root nodes with the PF and compute the leaves using the junction tree algorithm.This would result in a substantial computational gain as one only has to sample the root nodes and apply the juction tree to lower dimensional sub-networks.Although the previoulsy mentioned models are the mostfamous ones,there exist numerous other dynamic systems admitting ﬁnite dimensional ﬁlters.That is,the ﬁltering distribution can be estimated in closed-form at any time using a ﬁxed number of sufﬁcient statistics.These includeDynamic models for counting observations (Smith and Miller 1986).Dynamic models with a time-varying unknow covari-ance matrix for the dynamic noise (West and Harrison 1996,Uhlig 1997).Classes of the exponential family state space models (Vidoni 1999).This list is by no means exhaustive.It,however,shows that RBPFs apply to very wide class of dynamic models.Con-sequently,they have a big role to play in computer vision (where mixtures of Gaussians arise commonly),robotics,speech and dynamic factor analysis.ReferencesAkashi,H.and Kumamoto,H.(1977).Random samplingapproach to state estimation in switching environ-ments,Automatica 13:429–434.Andrieu,C.,de Freitas,J.F.G.and Doucet,A.(1999a).Se-quential Bayesian estimation and model selection ap-plied to neural networks,Technical Report CUED/F-INFENG/TR 341,Cambridge University Engineering Department.Andrieu,C.,de Freitas,J.F.G.and Doucet,A.(1999b).Se-quential MCMC for Bayesian model selection,IEEE Higher Order Statistics Workshop ,Ceasarea,Israel,pp.130–134.Becker,A.,Bar-Yehuda,R.and Geiger,D.(1999).Randomalgorithms for the loop cutset problem.Bernardo,J.M.and Smith,A.F.M.(1994).Bayesian The-ory ,Wiley Series in Applied Probability and Statis-tics.Boutilier,C.,Friedman,N.,Goldszmidt,M.and Koller,D.(1996).Context-speciﬁc independence in bayesian networks,Proc.Conf.Uncertainty in AI .Boyen,X.and Koller,D.(1998).Tractable inferencefor complex stochastic processes,Proc.Conf.Uncer-tainty in AI .Casella,G.and Robert,C.P.(1996).Rao-Blackwellisationof sampling schemes,Biometrika 83(1):81–94.Cowell,R.G.,Dawid,A.P.,Lauritzen,S.L.and Spiegel-halter,D.J.(1999).Probabilistic Networks and Ex-pert Systems ,Springer-Verlag,New York.Crisan,D.and Doucet,A.(2000).Convergence of gen-eralized particleﬁlters,Technical Report CUED/F-INFENG/TR381,Cambridge University Engineering Department.Crisan,D.,Del Moral,P.and Lyons,T.(1999).Dis-creteﬁltering using branching and interacting parti-cle systems,Markov Processes and Related Fields 5(3):293–318.de Freitas,J.F.G.(1999).Bayesian Methods for Neu-ral Networks,PhD thesis,Department of Engineer-ing,Cambridge University,Cambridge,UK.Dean,T.and Kanazawa,K.(1989).A model for reason-ing about persistence and causation,Artiﬁcial Intelli-gence93(1–2):1–27.Doucet,A.(1998).On sequential simulation-based meth-ods for Bayesianﬁltering,Technical Report CUED/F-INFENG/TR310,Department of Engineering,Cam-bridge University.Doucet, A.,de Freitas,J. F.G.and Gordon,N.J.(2000).Sequential Monte Carlo Methods in Practice, Springer-Verlag.Doucet,A.,Godsill,S.and Andrieu,C.(2000).On se-quential Monte Carlo sampling methods for Bayesian ﬁltering,Statistics and Computing10(3):197–208. Doucet, A.,Gordon,N.J.and Krishnamurthy,V.(1999).Particleﬁlters for state estimation of jump Markov linear systems,Technical Report CUED/F-INFENG/TR359,Cambridge University Engineering Department.Ghahramani,Z.and Jordan,M.(1997).Factorial Hidden Markov Models,Machine Learning29:245–273. Gilks,W.R.and Berzuini,C.(1998).Monte Carlo in-ference for dynamic Bayesian models,Unpublished.Medical Research Council,Cambridge,UK.Gordon,N.J.,Salmond,D.J.and Smith,A.F.M.(1993).Novel approach to nonlinear/non-Gaussian Bayesian state estimation,IEE Proceedings-F140(2):107–113.Green,P.J.(1995).Reversible jump Markov chain Monte Carlo computation and Bayesian model determina-tion,Biometrika82:711–732.Handschin,J.E.and Mayne,D.Q.(1969).Monte Carlo techniques to estimate the conditional expectation in multi-stage non-linearﬁltering,International Journal of Control9(5):547–559.Isard,M.and Blake, A.(1996).Contour tracking by stochastic propagation of conditional density,Euro-pean Conference on Computer Vision,Cambridge, UK,pp.343–356.Kanazawa,K.,Koller,D.and Russell,S.(1995).Stochastic simulation algorithms for dynamic probabilistic net-works,Proceedings of the Eleventh Conference on Uncertainty in Artiﬁcial Intelligence,Morgan Kauf-mann,pp.346–351.Kitagawa,G.(1996).Monte Carloﬁlter and smoother for non-Gaussian nonlinear state space models,Journal of Computational and Graphical Statistics5:1–25.Kong, A.,Liu,J.S.and Wong,W.H.(1994).Se-quential imputations and Bayesian missing data prob-lems,Journal of the American Statistical Association 89(425):278–288.Liu,J.S.and Chen,R.(1998).Sequential Monte Carlo methods for dynamic systems,Journal of the Ameri-can Statistical Association93:1032–1044.MacEachern,S.N.,Clyde,M.and Liu,J.S.(1999).Sequential importance sampling for nonparametric Bayes models:the next generation,Canadian Jour-nal of Statistics27:251–267.Murphy,K.P.(2000).Bayesian map learning in dynamic environments,in S.Solla,T.Leen and K.-R.M¨u ller (eds),Advances in Neural Information Processing Systems12,MIT Press,pp.1015–1021.Pearl,J.(1988).Probabilistic Reasoning in Intelligent Sys-tems:Networks of Plausible Inference,Morgan Kauf-mann.Pitt,M.K.and Shephard,N.(1999).Filtering via simula-tion:Auxiliary particleﬁlters,Journal of the Ameri-can Statistical Association94(446):590–599.Smith,R.L.and Miller,J.E.(1986).Predictive records, Journal of the Royal Statistical Society B36:79–88.Uhlig,H.(1997).Bayesian vector-autoregressions with stochastic volatility,Econometrica.Vidoni,P.(1999).Exponential family state space models based on a conjugate latent process,Journal of the Royal Statistical Society B61:213–221.West,M.(1993).Mixture models,Monte Carlo,Bayesian updating and dynamic models,Computing Science and Statistics24:325–333.West,M.and Harrison,J.(1996).Bayesian Forecasting and Dynamic Linear Models,Springer-Verlag.。

journal of physics and chemistry of solids 分区

journal of physics and chemistry of solids 分区The Journal of Physics and Chemistry of Solids (JPCS) is a prominent scientific journal that focuses on the interdisciplinary study of materials science, solid-state physics, and physical chemistry. With its wide scope and rigorous peer-review process, JPCS has gained recognition as an influential publication in the field.1. IntroductionThe Journal of Physics and Chemistry of Solids (JPCS) serves as a platform for researchers, scientists, and scholars to disseminate their findings, theories, and innovations in the areas of materials science, solid-state physics, and physical chemistry. This article explores the journal's unique features, impact factor, and its categorization based on the Journal Citation Reports (JCR) journal rankings.2. Scope and FocusJPCS covers a broad range of subjects within the fields of materials science, solid-state physics, and physical chemistry. The journal publishes original research articles, review papers, and short communications on various topics, including but not limited to:2.1 Materials Science- Synthesis and characterization of novel materials- Functional materials and their applications- Nanomaterials and nanotechnology- Composite materials- Energy materials2.2 Solid-State Physics- Electronic and magnetic properties of solids- Superconductivity and magnetism- Optical properties and spectroscopy- Quantum phenomena in materials- Topological materials2.3 Physical Chemistry- Chemical reactions and kinetics- Surface science and catalysis- Theoretical and computational chemistry- Solid-state chemistry- Thermodynamics and phase transitions3. Impact and RankingsThe impact factor (IF) of a journal reflects its influence and importance within the scientific community. JPCS consistently maintains a high impact factor due to the quality and significance of the research published. The journal's impact factor is calculated annually by the JCR, which analyzes theaverage number of citations received per article in a given year. JPCS has demonstrated a strong impact on the scientific community, attracting citations and recognition from researchers worldwide.4. Journal CategorizationThe Journal of Physics and Chemistry of Solids (JPCS) is classified under the category of "Materials Science, Multidisciplinary" according to the Journal Citation Reports (JCR) journal rankings. This classification acknowledges the journal's interdisciplinary nature and its contributions to the advancement of materials science, solid-state physics, and physical chemistry. JPCS also falls within the larger field of materials science, which encompasses various specialized areas of research and applications.5. ConclusionIn conclusion, the Journal of Physics and Chemistry of Solids (JPCS) serves as a vital source of knowledge and research in the fields of materials science, solid-state physics, and physical chemistry. The journal's broad scope, rigorous peer-review process, and high impact factor make it an invaluable resource for researchers, scientists, and scholars worldwide. By publishing groundbreaking research and facilitating interdisciplinary collaboration, JPCS continues to contribute to the advancement of science and technology.。

软件工程阅读文献5000字

软件工程阅读文献5000字I have read a lot of literature on software engineering, and I must say, it is a fascinating field with so much to learn and explore. One of the key aspects that I found particularly interesting is the concept of agile software development. Agile methodologies emphasize flexibility and collaboration, allowing teams to quickly adapt to changing requirements and deliver high-quality software in a timely manner.For example, I read a case study on how a software development team used agile practices to successfullydeliver a complex project within a tight deadline. By breaking down the project into smaller, manageable tasksand regularly communicating with stakeholders, the team was able to stay on track and make necessary adjustments along the way. This not only improved the overall efficiency of the project but also ensured that the final product met the client's expectations.Another aspect of software engineering that caught my attention is the importance of software testing. Testing is crucial in identifying and fixing bugs or issues in the software before it is released to users. I learned about different testing techniques such as unit testing, integration testing, and acceptance testing, each serving a specific purpose in ensuring the quality and reliability of the software.In addition, I came across research on the impact of software architecture on the overall performance and scalability of a system. A well-designed architecture can significantly improve the maintainability and extensibility of a software application, making it easier to add new features or make changes in the future. I found it fascinating how the choice of architecture can have a profound effect on the success of a software project.Overall, my exploration of software engineering literature has broadened my understanding of the field and inspired me to delve deeper into topics such as software design patterns, code refactoring, and continuousintegration. I look forward to applying these concepts inmy own projects and continuing to learn and grow as a software engineer.---。

自动化专业-外文文献-英文文献-外文翻译-plc方面

1、外文原文（复印件)A: Fundamentals of Single-chip MicrocomputerTh e si ng le-ch i p mi cr oc om pu ter is t he c ul mi nat i on o f bo th t h e d ev el op me nt o f th e d ig it al com p ut er an d t he int e gr at ed ci rc ui ta r gu ab ly th e t ow m os t s i gn if ic ant i nv en ti on s o f t h e 20t h c en tu ry[1].Th es e to w t ype s o f a rc hi te ct ur e a re fo un d i n s i ng le—ch ip m i cr oc om pu te r。

S o me em pl oy th e s p li t p ro gr am/d at a me mo ry of t he H a rv ar d ar ch it ect u re, sh ow n in Fi g.3-5A—1，ot he r s fo ll ow t hep h il os op hy, wi del y a da pt ed f or ge n er al—pu rp os e c o mp ut er s an dm i cr op ro ce ss or s, of ma ki ng no lo gi c al di st in ct io n be tw ee n p ro gr am a n d da ta m em or y a s i n th e Pr in cet o n ar ch it ec tu re,sh ow n in F ig。

3-5A-2.In g en er al te r ms a s in gl e—ch i p mi cr oc om pu ter isc h ar ac te ri zed b y the i nc or po ra tio n of al l t he uni t s o f a co mp ut er i n to a s in gl e de v i ce，as s ho wn i n F ig3—5A—3。

plc英文文献

*Department of Manufacturing Engineering, City University of Hong Kong, 83 Tat Chee Avenue, Hong Kong and tMechatronics Research Group, De Montfort University, The Gateway, Leicester LE1 9BH, U.K.
1. CONVEYOR SYSTEMS Conveyor systems are often modular in nature and can be built up from basic units (or primitives) such as linear conveyor modules, either belt or roller type, and connecting devices such as lift stations and conveyor junction modules. Complex conveyor systems can easily be configured from combinations of these standard modules. Modules are available from a range of vendors in a wide portfolio of styles and varieties [1]. Conventional conveyor systems are typically installed as simple straight assembly lines and a number of workplaces are set on each side of the conveyor for manual and/or automated operations. For simple configurations of this type, the design and implementation is a relatively trivial task. Control programs are easily designed and coded using conventional Ladder Logic Diagrams (LLDs) which can be executed by Programmable Logic Controllers (PLCs) [2]. However, today's demands for multi-product mixes and flexibility for quick system reconfiguration can require more complex conveyor systems. Traditional types of sequentially controlled conveyor installations are often found to be too rigid for such demanding operational requirements. Conveyor systems which support multiple product mixes and variable product 799

6Differential Absorption Spectroscopy

6Diﬀerential Absorption SpectroscopyAbsorption spectroscopy is a well-established tool for the analysis of the chem-ical composition of gases.As such,it has played a prominent role in the dis-covery of the physical and chemical properties of the earth’s atmosphere. 6.1The History of Absorption SpectroscopySpectroscopic studies of the earth’s atmosphere date back more than100 years.Some milestones in the investigation of atmospheric composition through spectroscopy include:1879–Marie Alfred Cornu concludes from the change of the edge of the intensity decay in the UV that a trace species in the earth’s atmosphere must be causing the UV-absorption(Cornu,1879).1880–Sir Walter Noel Hartley discovers the absorption of UV-radiation (below300nm)by ozone.This led to the name Hartley-bands for ozone absorption below300nm(Hartley,1880,1881).1880–M.J.Chappuis discovers the absorption of visible light by ozone, which is today called the Chappuis-band.Chappuis also speculates that light absorption by ozone is the reason for the blue colour of the sky (Chappuis,1880).1890–Sir William Huggins discovers a new group of lines in the spectrum of Sirius,which are later explained by Fowler and Strutt as absorption of terrestrial ozone.The long wavelength UV-bands of ozone are,therefore, called Huggins-bands today.1904–Discovery of the infrared absorption of ozone near4.8,5.8,and9.1–10μm by Knut Johan˚A ngstr¨o m.1913–Balloon measurements of the UV absorption of ozone up to10km altitude by Albert Wigand showed essentially no change with altitude. 1918–John William Strutt(better known as Lord Rayleigh)concludes that atmospheric ozone must reside in a layer above10km altitude above the surface.U.Platt and J.Stutz,Diﬀerential Absorption Spectroscopy.In:U.Platt and J.Stutz,Diﬀerential Optical Absorption Spectroscopy,Physics of Earth and Space Environments,pp.135–174(2008)DOI10.1007/978-3-540-75776-46c Springer-Verlag Berlin Heidelberg20081366Diﬀerential Absorption Spectroscopy1920–First ozone column measurements were made by Charles Fabry and Henri Buisson,who determine a column of about3mm(at atmospheric pressure),with large variations.1925–First application of a dedicated ozone spectrometer by Gordon Miller Bourne Dobson(Dobson and Harrison,1926).1926–Paul G¨o tz conﬁrms the theory of an ozone layer by observing the so-called‘Umkehr’eﬀect,and determines its altitude to be about25km. 1934–Direct observation of the ozone layer by UV-spectroscopy by Erich Regener(TH Stuttgart).1948–Marcel Migeotte(Ohio State University)discovers methane and carbon monoxide in the earth atmosphere by near-infrared absorption spectroscopy(Migeotte,1948,1949).1950–Discovery of the emission bands of the hydroxyl radical(OH)in the nightglow(the Meinel bands of the OH-radical).As a consequence, HO x-chemistry is viewed in connection with ozone chemistry by David R.Bates and Marcel Nicolet(1950),and Bates and Witherspoon (1952).1975–First detection of OH in the atmosphere by Dieter Perner and col-leagues using diﬀerential optical absorption spectroscopy(Perner et al., 1976).This list illustrates the role that spectroscopy has played in the measure-ment of reactive trace gases in the atmosphere,most notably ozone.The reader may notice that the identiﬁcation and quantiﬁcation of gases was primarily accomplished by the analysis of atmospheric absorptions.This is still the case in most current applications of atmospheric spectroscopy.The use of emission bands is restricted to the thermal infrared wavelength region(see Chap.5) or to the excited gas molecules in the upper atmosphere,which emit light at higher energies,i.e.shorter wavelength.Both applications are in use today, but are not the topic of discussion in this book.The initial use of spectroscopy in the atmosphere concentrated on the identiﬁcation of various gases.Soon,however,this method was put in use to quantify the concentrations(or column densities)of these species.In particu-lar,the contributions of Dobson,who constructed theﬁrst instrument for the regular measurement of atmospheric ozone,should be singled out(Dobson and Harrison,1926).This chapter focuses on a modern method to quantitatively measure a large variety of trace gases in the atmosphere.DOAS is now one of the most commonly used spectroscopic methods to measure trace gases in the open atmosphere.At the beginning,we give a general introduction to absorption spectroscopy and DOAS.This is followed by an overview of diﬀerent exper-imental approaches of DOAS and a discussion of the precision and accuracy of this method.The last section of this chapter is dedicated to a rigorous mathematical description of the various DOAS applications.6.2Classical Absorption Spectroscopy 1376.2Classical Absorption SpectroscopyThe basis of the early spectroscopic measurements,and many present quanti-tative trace gas analytical methods in the atmosphere and the laboratory,is Lambert–Beer’s law,often also referred to as Bouguer–Lambert law.The law was presented in various forms by Pierre Bouguer in 1729,Johann Heinrich Lambert in 1760,and August Beer in 1852.Bouguer ﬁrst described that,‘In a medium of uniform transparency the light remaining in a collimated beam is an exponential function of the length of the path in the medium’.However,there was some confusion in the naming of this law,which may be either the name of individual discoverer or combinations of their names.In this book,we have referred to it as Lambert–Beer’s law.A variety of spectroscopic techniques make use of the absorption of elec-tromagnetic radiation by matter (Fig.6.1).In a formulation suitable for the analysis of gaseous (or liquid)absorbers,Lambert–Beer’s law can be writ-ten as:I(λ)=I0(λ)·exp (−σ(λ)·c ·L).(6.1)Here,I 0(λ)denotes the initial intensity of a light beam emitted by a suitable source of radiation,while I (λ)is the radiation intensity of the beam after pass-ing through a layer of thickness L ,where the absorber is present at a uniform concentration of c .The quantity σ(λ)denotes the absorption cross-section at wavelength λ.The absorption cross-section as a function of wavelength is a characteristic property of any species.The determination of the light path length,L ,is usually trivial for active DOAS applications (see Chap.4).Once those quantities are known,the average trace gas concentration,c ,can be calculated from the measured ratio I 0(λ)/I (λ):c =ln I 0(λ)I(λ) σ(λ)·L=D σ(λ)·L .(6.2)DetectorL IntensityI 0(λ)I(λ)LightSource Absorber with concentration cFig. 6.1.The basic principle of absorption spectroscopic trace gas detection.A beam of light passes through a volume of length L containing the absorber with concentration c .At the end of the light path the intensity is measured by a suitable detector1386Diﬀerential Absorption SpectroscopyThe expressionD=lnI0(λ)I(λ),(6.3)is called the optical density of a layer of a given absorber.Note that,in the literature,the decadal as well as the natural logarithm is used in the deﬁnition of optical density.In this book,we will exclusively use the natural logarithm.Equation(6.2)is the basis of most absorption spectroscopic applications in the laboratory,where the intensities I(λ)and I0(λ)are determined by measurements with and without the absorber in the light beam.However,the application of Lambert–Beer’s law is more challenging in the open atmosphere.Here,the true intensity I0(λ),as it would be received from the light source in the absence of any atmospheric absorber,is diﬃcult to determine.It would involve removing the air,or more precisely the absorbing gas,from the atmosphere.While this may seem to present a dilemma rendering atmospheric absorption spectroscopy useless in this case,the solution lies in measuring the so-called‘diﬀerential’absorption,i.e.the diﬀerence between the absorptions at two diﬀerent wavelengths.This principle was used by Dobson in the1930s to determine the total column of atmospheric ozone.In an ingenious experimental setup,the Dobson spectrometer compares the intensity of direct solar light of two wavelengths–λ1,λ2–with diﬀerent ozone absorption cross-section,σ1=σ(λ1),σ2=σ(λ2)(Dobson and Harrison,1926).6.3The DOAS PrincipleA schematic setup of an experiment to measure trace gas absorptions in the open atmosphere is shown in Fig.6.2.Similar to Fig.6.1,light emitted by a suitable spectral broadband source with an intensity I0(λ)passes through a volume with absorbers(here the open atmosphere),and is collected at the end of the light path.As the light travels through the atmosphere,its intensity is reduced through the absorption of a speciﬁc trace gas.However,it also undergoes extinction due to absorption by other trace gases,and scattering by air molecules and aerosol particles.The transmissivity of the instrument (mirrors,grating,retro-reﬂectors,etc.)will also decrease the light intensity, as will the light beam widening by turbulence.By expanding Lambert–Beer’s law,one can consider the various factors that inﬂuence the light intensity by an equation that includes the absorption of various trace gases with concentration c j and absorption cross-sectionsσj(λ),Rayleigh and Mie extinction,εR(λ) andεM(λ)(described byεR(λ)≈σR0(λ)·λ−4·c AIR andεM(λ)=σM0·λ−n·N A,respectively;see Chap.4),and instrumental eﬀects and turbulence, summarised in A(λ):I(λ)=I0(λ)·exp−L·(σj(λ)·c j)+εR(λ)+εM(λ)·A(λ).(6.4)6.3The DOAS Principle 139I 0(~λ–4~λ–(1...3)Fig.6.2.Sketch of an experiment to measure trace gas absorptions in the open atmosphereTo determine the concentration of a particular trace gas,it would,in principle,be necessary to quantify all other factors inﬂuencing the intensity.In the laboratory,this can be achieved by removing the absorber from the light path.In the atmosphere,however,where this is impossible,the multiple factors inﬂuencing the intensity pose a dilemma.Diﬀerential optical absorption spectroscopy overcomes this challenge by using the fact that aerosol extinction processes,the eﬀect of turbulence,and many trace gas absorptions show very broad or even smooth spectral charac-teristics.Certain trace gases,however,exhibit narrowband absorption struc-tures.The foundation of DOAS is thus to separate broad-and narrowband spectral structures in an absorption spectrum in order to isolate these narrow trace gas absorptions (Fig.6.3).The broad spectrum is then used as a new intensity spectrum I 0 (λ),and Lambert–Beer’s law can again be applied to the narrowband trace gas absorptions.Figure 6.3illustrates the separation of the narrow-and broadband struc-tures for one absorption band,both for the absorption cross-section and the intensity:σj (λ)=σj 0(λ)+σ j (λ)(6.5)σj 0in (6.5)varies ‘slowly’with the wavelength λ,for instance describing a gen-eral ‘slope’,such as that caused by Rayleigh and Mie scattering,while σ j (λ)shows rapid variations with λ,for instance due to an absorption band (see Fig.6.3).The meaning of ‘rapid’and ‘slow’variation of the absorption cross-section as a function of wavelength is,of course,a question of the observed wavelength interval and the width of the absorption bands to be detected.Inserting (6.5)into (6.4),we obtain:I(λ)=I 0(λ)·exp ⎡⎣−L ·⎛⎝ jσ j(λ)·c j ⎞⎠⎤⎦·exp ⎡⎣−L ·⎛⎝ j (σj0(λ)·c j )+εR (λ)+εM (λ)⎞⎠⎤⎦·A (λ),(6.6)1406Diﬀerential AbsorptionSpectroscopyFig.6.3.Principle of DOAS:I 0and σare separated by an adequate ﬁltering pro-cedure into a narrow (D ,and σ )and broad band part (I 0and σb )where the ﬁrst exponential function describes the eﬀect of the structured ‘diﬀerential’absorption of a trace species,while the second exponential con-stitutes the slowly varying absorptions as well as the inﬂuence of Rayleigh and Mie scattering.The attenuation factor A (λ)describes the broad wavelength-dependent transmission of the optical system used and turbulence.Thus,we can deﬁne a quantity I 0as the intensity in the absence of diﬀerential absorption:I 0(λ)=I 0(λ)·exp ⎡⎣−L ·⎛⎝ j(σj0(λ)·c j )+εR (λ)+εM (λ)⎞⎠⎤⎦·A (λ).(6.7)The corresponding diﬀerential absorption cross-section σ j (λ)is then substi-tuted for σj (λ)in (6.1)and (6.2).σ j (λ)is determined in the laboratory (i.e.taken from literature data),just like σj (λ).Likewise,a diﬀerential optical den-sity,D ,can be deﬁned in analogy to (6.3)as the logarithm of the quotient of the intensities I 0and I 0(as deﬁned in (6.7)and (6.6),respectively):D=ln I 0(λ)I(λ)=L · j σ j (λ)·c j .(6.8)Atmospheric trace gas concentrations can then be calculated according to (6.2),with diﬀerential quantities D and σ (λ)substituted for D and σ(λ),respectively.A separation of the diﬀerent absorptions in the sum of (6.8)is6.4Experimental Setups of DOAS Measurements141 possible because the structures of the trace gases are unique,like aﬁngerprint (see Sect.6.5).Both the separation of broad and narrow spectral structures and the sep-aration of the various absorbers in(6.8)require the measurement of the radi-ation intensity at multiple wavelengths.In fact,DOAS measurements usually observe the intensity at500–2000individual wavelengths to accurately de-termine the concentrations of the various absorbing trace gases.The use of multiple wavelengths is an expansion of the principles used,for example,by Dobson,which were based on two or four wavelengths.The use of diﬀerential absorptions over an extended wavelength range has a number of major advantages.Because the transmission of optical instru-ments typically shows broad spectral characteristics,no calibration of the optical properties or their change with time is necessary.This often makes the instrumentation much simpler and less expensive.The use of a multi-tude of wavelengths allows the unique identiﬁcation of trace gas absorptions.A further major advantage of this approach is the opportunity to observe and quantify extremely weak absorptions corresponding to optical densities around D =10−4.In particular,the ability to use very long light paths in the atmosphere,in active DOAS applications sometimes up to10–20km long (passive DOAS applications can reach1000km),increases the sensitivity of DOAS and,at the same time,provides spatially averaged values.Before giving a more rigorous mathematical description of the DOAS method,the basic experimental setups,the trace gases that are commonly measured,and the typical detection limits of DOAS will be reviewed in the following sections.6.4Experimental Setups of DOAS MeasurementsThe DOAS principle as outlined earlier can be applied in a wide variety of light path arrangements and observation modes(Fig.6.4).To provide a general overview of diﬀerent setups,we introduce a classiﬁcation system that will later be used in the description of the diﬀerent analysis methods(see Sect.6.7 and Chap.8)and the technical details(Chap.7).According to their light sources,we distinguish between active and pas-sive DOAS.In short,active DOAS uses artiﬁcial light,while passive DOAS relies on natural light sources,i.e.solar,lunar,or stars.An overview of the most common experimental setups illustrates the breadth of DOAS applica-tions that are in use today(Fig.6.4).6.4.1Active DOASActive DOAS applications have one thing in common–they rely on an arti-ﬁcial light source coupled to an optical setup that is used to send and receive1426Diﬀerential Absorption Spectroscopy1. Long-Path DOAS (LP-DOAS)2. Vertical Profiling LP-DOASReflectors Light sourceRetro-reflector3. Tomographic DOAS4. Folded-Path DOAS5. Direct Sunlight DOAS6. Balloon-borne (direct sunlight) DOASLPMA/DOAS Gondola + Balloon7. Satellite-borne DOAS - Occultation8. Zenith Scattered Light (ZSL-DOAS)9. Multi-Axis DOAS (MAX-DOAS)10. Airborne Multi-Axis DOAS (AMAX-DOAS)Fig.6.4.The DOAS principle can be applied in a wide variety of light path ar-rangements and observation modes using artiﬁcial (1–4)as well as natural direct (5–7)or scattered (8–14)light sources.Measurements can be done from the ground,balloons,aircrafts,and from space6.4Experimental Setups of DOAS Measurements14312. Satellite-borne DOAS -Nadir Geometry11. Imaging DOAS13. Satellite-borne DOAS - Scattered LightLimb Geometry14. Determination of the Photon Path length L (in Clouds) ‘inverse DOAS’Fig.6.4.Continuedlight in the atmosphere.Spectroscopic detection is achieved by a spectrom-eter at the end of the light path.In general,active DOAS is very similar to classical absorption spectroscopy,as employed in laboratory spectral pho-tometers.However,the low trace gas concentrations in the atmosphere require very long light paths (up to tens of kilometres in length,see above),making the implementation of these instruments challenging (see Chap.7for details).Active DOAS applications are typically employed to study tropospheric com-position and chemistry,with light paths that are often parallel to the ground.In addition,active DOAS systems are also used in smog and aerosol chamber experiments.The earliest applications of active DOAS,i.e.the measurement of OH radicals (Perner et al.,1976),used a laser as the light source along one single path (Fig.6.4,Plate 1).This long-path DOAS setup is today most com-monly used with broadband light sources,such as xenon-arc lamps,to mea-sure trace gases such as O 3,NO 2,SO 2,etc.(e.g.Stutz and Platt,1997a,b).Expansion of this method involves folding the light beam once by using retro-reﬂectors on one end of the light path (Axelsson et al.,1990).This setup simpliﬁes the ﬁeld deployment of long-path DOAS instruments.In ad-dition,applications that use multiple retro-reﬂector setups to probe on dif-ferent air masses are possible.Figure 6.4,Plate 2,shows the setup that is used to perform vertical proﬁling in the boundary layer with one DOAS sys-tem.An expansion that is currently under development is the use of mul-tiple crossing light paths to perform tomographic measurements (Fig.6.4,Plate 3).1446Diﬀerential Absorption SpectroscopyIn applications where detection in smaller air volumes with high sensitiv-ity is required,folded-path DOAS is often used(Fig.6.4,Plate4)(e.g. Ritz et al.,1992).Because the light can pass the multiple reﬂection cells in these systems up to144times,long light paths can be achieved in small air volumes.These systems are the most common DOAS setups in laboratory applications,where interference by aerosols makes the use of classical absorp-tion spectroscopy impossible(e.g.smog and aerosol chambers).Folded-path DOAS has also been used for the same applications as long-path DOAS(e.g. Alicke et al.,2003;Kurtenbach et al.,2002).In particular,the use of laser to measure OH has been successful(see Chap.10).Active DOAS measurements have contributed to the discovery and quan-tiﬁcation of a number of important atmospheric trace species,most notably the radicals OH and NO3(Perner et al.,1976;Platt et al.,1979).The ele-gance of active DOAS is that the expanded Lambert–Beer’s law(6.4)can be directly applied to the calculation of trace gas concentrations based only on the absorption cross-section,without the need for calibration of the instru-ment in theﬁeld.This gives active DOAS high accuracy and,with the long light paths,excellent sensitivity.6.4.2Passive DOASPassive DOAS utilises light from natural sources.The two most important sources are the sun and the moon.However,the use of light from other stars has also been reported.While the measurement of light directly from moon and stars is possible,sunlight oﬀers two alternatives:direct sunlight and sun-light scattered in the atmosphere by air molecules and particles.We will fur-ther subdivide passive DOAS applications into direct and scattered light measurements(see also Chap.11).Direct measurements use the sun,moon,or stars as light sources,and thus share the advantage of active DOAS of directly applying Lambert–Beer’s law. However,since the light crosses the entire vertical extent of the atmosphere, a direct conversion of absorptions to concentrations is not possible.Instead, the column density,i.e.the concentration integrated along the path,is the direct result of these measurements.Only by using geometric and radiative transfer calculations can these measurements be converted into vertically in-tegrated column densities(VCD)or vertical concentration proﬁles.The most common example for VCDs is the total ozone column,which is measured in Dobson units.Figure6.4gives several examples for direct passive DOAS setups.Besides direct measurements of sun,moon,and star light from the ground(Fig.6.4,Plate5),balloon-borne solar measurements have been very successful(Fig.6.4,Plate6).The measurements during the ascent provide ver-tical proﬁles of various trace gases.With the recent deployment of space-borne DOAS instruments,i.e.SCIAMACHY,occultation measurements(Fig.6.4, Plate7)have also become possible.6.4Experimental Setups of DOAS Measurements145Scattered sunlight measurements are more universally used in passive DOAS since they oﬀer the largest variety of applications.The measurement of scattered light from the zenith(Fig.6.4,Plate8)was one the earliest applications of passive DOAS(see Sect.11.2),and has contributed consider-ably to our understanding of stratospheric chemistry(e.g.Mount et al.,1987; Solomon et al.,1987,1988,1989).In addition,zenith scattered light has also been used to study the radiative transport in clouds(Fig.6.4,Plate14),which is an important topic in climate research(Pfeilsticker et al.,1998b,1999).A more recent development of passive scattered DOAS is the use of multiple viewing geometries(Fig.6.4,Plate9).This multi-axis DOAS(MAX-DOAS) uses the fact that,at low viewing elevations,the length of the light path in the lower troposphere is considerably elongated(e.g.H¨o nninger et al.,2004).It is thus possible to probe the lower troposphere sensitively.In addition,vertical proﬁles can be derived if enough elevation angles are measured.MAX-DOAS can also be employed from airborne platforms,allowing the measurements below and above theﬂight altitude(Fig.6.4,Plate10),as well as determina-tion of vertical concentration proﬁles.An expansion of MAX-DOAS,which is currently under development,is Imaging DOAS(Fig.6.4,Plate11),where a large number of viewing elevations are measured simultaneously to visualise pollution plumes.Over the past decade,DOAS has also been used for satellite-borne mea-surements(Fig.6.4,Plates12,13),which use sunlight scattered either by the atmosphere,the ground,or both(see Sect.11.5).Two viewing geometries of these measurements are possible(for details and examples,see Chap.11).In the nadir geometry,the DOAS system looks down towards the earth’s surface. Instruments such as GOME provide global concentrationﬁelds of trace gases, such as O3,NO2,and HCHO.The SCIAMACHY instrument also employs measurements in limb geometry,which allow the determination of vertical trace gas proﬁles with high resolution(Fig.6.4,Plate13).The advantage of passive DOAS applications is the relatively simple exper-imental setup.For example,scattered light measurements require only small telescopes.In addition,no artiﬁcial light source is needed.However,a number of additional challenges have to be addressed in passive DOAS applications. Because solar and lunar light is spectrally highly structured,special care needs to be taken.To detect very small trace gas absorptions,the strong Fraunhofer bands must be accurately measured.In addition,the fact that the light source structure contains narrow and deep absorptions also makes the application of the DOAS technique,which was outlined in Sect.6.3,more diﬃcult.This will be discussed in more detail below.The largest challenge in using passive DOAS is the conversion of the observed column densities to vertical column densities,concentrations,and vertical proﬁles.This is,in particular,the case for scattered light setups,where the length of the light path is diﬃcult to de-termine.The interpretation of these measurements,therefore,must be based on detailed radiative transfer calculations(see Chap.9).1466Diﬀerential Absorption Spectroscopy6.5Trace Gases Measured by DOASThe separation of broad and narrow spectral structures(Sect.6.3),while making absorption spectroscopy usable in the atmosphere,restricts DOAS measurements to trace gases that have narrow band absorption structures with widths narrower than∼10nm.In theory,all gases that have these narrow absorption bands in the UV,visible,or near IR can be measured.However, the concentrations of these compounds in the atmosphere,and the detection limits of today’s DOAS instruments,restrict the number of trace gases that can be detected.As DOAS instruments improve in the future,this list will most likely grow.Figures6.5and6.6show the absorption cross-sections of a number of trace gases that are regularly measured by DOAS.A number of features about these cross-sections should be pointed out here.First and most importantly,each trace gas spectrum has a unique shape.Most of the trace gases only absorb in certain wavelength intervals.However,many spectral regions can contain a large number of simultaneous absorbers.For example,between300and 400nm,the following trace gases will show absorption features if they are present at high enough concentrations:O3,SO2,NO2,HONO,HCHO,and BrO.Because of their unique spectral structure,a separation of the absorp-tions is possible.From(6.8),it is clear that spectral regions with higherσ will show the largest optical densities.These spectral intervals are thus preferred for DOAS measurements since the sensitivity improves in these wavelength regions.In principle,each trace gas has an optimal wavelength interval.In practice,however,one has to often compromise in the choice of the wave-length interval to measure more than one trace gas simultaneously.Because expanding the wavelength window reduces the spectral resolution of typical grating spectrometers,the sensitivity is also reduced.The choice of trace gases thus depends on the speciﬁc application.In Fig.6.7,we have attempted to aid in the choice of the best wavelength re-gion by visualising the detection limits of an extended set of trace gases for long-path applications in the troposphere.It should be added that a number of other trace gases,besides those shown in Figs.6.5and6.6,can be measured.Table6.1gives an overview of the var-ious trace gases measured by DOAS,including stratospheric trace gases.At shorter wavelengths,the usable spectral range of DOAS is limited by rapidly increasing Rayleigh scattering and O2absorption(Volkamer et al.,1998). Those eﬀects limit the maximum light path length to∼200m in the wave-length range from200–230nm,where,for instance,the sole usable absorption features of species such as NO(Tajime et al.,1978)and NH3are located(see Fig.6.7and Table6.1).6.5Trace Gases Measured by DOAS147200300400500600700NO 2NO 3SO 2A b s o r p t i o n c r o s s -s e c t i o n i n 10–19 c m 2 m o l e c u l e –1Wavelength in nmBrOHCHO CHOCHOIOOIOClOHONOO 4I 2OBrOOClO0100100.01.0 x 10–451000.00.802000602000100060050400100010001 × 102O 3 cross-section × 1000O 3Fig.6.5.Details of the absorption cross-section features of a number of species of atmospheric interest as a function of wavelength (in nm).Note the ‘ﬁngerprint’nature of the diﬀerent spectra1486Diﬀerential Absorption SpectroscopyFig.6.6.Details of the diﬀerential absorption cross-section features of a number of monocyclic aromatic species,O2,and O3as a function of wavelength(in nm).Note the‘ﬁngerprint’nature of the diﬀerent spectra。

1、下载文档前请自行甄别文档内容的完整性，平台不提供额外的编辑、内容补充、找答案等附加服务。
2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
3、如文档侵犯您的权益，请联系客服反馈,我们会尽快为您处理(人工客服工作时间：9:00-18:30)。

外文文献-The basic of description of android system对Android系统的基本描诉毕业设计外文文献原文及译文The basic of description of android system对Android系统的基本描诉学生姓名: 学号:电子与计算机科学技术系系别:计算机科学与技术专业:指导教师:2015年 5 月2015届毕业设计外文文献原文及译文The basic of description of android systemBy the Open Mobile Alliance (open Handset Alliance led by Google) developed the android system is a widely optimistic about an open source phone system, the system provides a basic operating system, a middle ware application layer, a java development tools and a system Application collector (collection of system applications).The android the SDK since 2007 on the release of the first android phone in October 2008 before the birth. Google opened since then on his own time, Taiwan's HTC, the manufacturer of the T-Mobile G1 estimate G1 shipments have more than one million at the end of 2008. According to industry insiders expect the G1 mobile phone sales in 2009 continue.Many other mobile phone suppliers in the near future plans to supportthis system.Around an android and a huge developer community has been established, while a lot of new products and applications on the android. Android's main selling point is that it enables developers to seamlessly expand online services to mobile phones. Calendar and Contacts Web applications through the system. Users only need to provide an android user name and password, the phone automatically sync with Google services. The other vendors are quickly adapt their existing instant messaging, social networking and gaming services. Android and many companies find new ways to integrate their existing business to the android.Traditional desktop and server operating system has been working for the integration of security features. These individuals and business applications on a single platform is very good, however a business phone platform like android is not very useful. It gives the hope of many researchers. Android is not parked in the body for other platform application support: the implementation of the application depends on a top-level JAVA middle ware, the middle ware running on the embeddedLinux kernel. Therefore, developers should deploy their applications to the Android must use a custom user interface environment.In addition, the android system applications limit the applicationto call each other API .Although these applications have certain safety features, some ofour experienced developers to create Android applications who revealed that the design of security2015届毕业设计外文文献原文及译文applications is not always straight forward. Android uses a simple permission label distribution mode to restrict access to resources, but the reasons for the necessity and convenience of other applications, the designers have increased the confusion on this system. This paper attempts to explain the complexity of the Android security, and pay attention to some of the possible development defects and application security. We try to draw some lessons learned, and hope that the safety of the future.Android application framework for developers is a mandatory framework. It does not have a main () function function or a single entry point for the implementation of the contrary, the developer must in the design of application components. We developed applications to help the API of the android SDK .The Android system defines four kinds of component type.Activity component that defines the application user interface. Usually, the application developer defines each activity screen.Activity can start, it may pass and return values. Can be handled at a time only a keyboard system Activity, all other Activity will be suspended at this time.Service components perform background processing. The need for some operations when an activity, after the disappearance of the user interface (such as downloading a file or playing music), it usually take such action specially designed services. Developers can also use a special daemon at system start up, the service is usually defined a remote procedure call (RPC), and other system components can be used to send the interface command and retrieve data, as well as to register a callback function.Content Provider component storage and share data with relational database interfaces. Each Content supplier has an associated "rights" to describe its contents contains. Other components when used as a handle to execute SQL queries (for example SELECT, INSERT,or DELETE content. Content suppliers are typically stored the values on the database records, data retrieval is a special case, the file is also shared by the content provider interface.The components of the broadcast receiver as to send a message from the mailbox to the application. Typically, the broadcast message, the application code implicit destination.2015届毕业设计外文文献原文及译文Therefore, the radio receiver subscribe to these destinationsreceive messages sent to it. The application code can also be solved explicitly broadcast receivers, including the name space allocation.The main mechanism of the interaction of the components of the Component Interaction, is an intent, which is a simple message object,which contains a destination address and data components. The Android API defines his approach into intent, and use that information to initiate an activity.such as start an activity (startActivity (An intent)) start services (the start Service (An intent)) and radio (sendBroadcast (An intent)). Android framework to inform the calls to these methods began to perform in the target application code. This process, the internal components of communication is called an action. Simply put, the Intent object defined in the "Intent to implement the" action ". One of the most powerful features of the Android is allowed a variety of intent addressing mechanism. The developer can solve the space of a target component using its applications, they can also specify an implicit name. In the latter case, the system determines the best components of an action by considering the installed applications and user choice.Android applications are written in the Java programminglanguage.The compiled Java code — along with any data and resource files required by the application — is bundled bythe apt tool into an Android package,an archive file marked by an .A suffix.This file is the vehicle for distributing the application and installing it on mobile devices;it's the file users download to their devices.All the code in a single.APK file is considered to be one application.In many ways,each Android application lives in its own world:(1) By default,every application runs in its own Linuxprocess.Android starts the process when any of the application's code needs to be executed,and shuts down the process when it's no longer needed and system resources are required by other applications.(2) Each process has its own virtual machine(VM),so application code runs in isolation from the code of all other applications.2015届毕业设计外文文献原文及译文(3) By default,each application is assigned a unique Linux userID.Permissions are set so that the application's files are visible only to that user and only to the application itself there are ways to export them to other applications as well.Application ComponentsA central feature of Android is that one application can make use of elements of other application (provided those application permit it).For example,if your application needs to display a scrolling list of images and another application has developed a suitable and made it available to others,you can call upon that to do the work,rather than develop your own.Your application doesn't incorporate the code of the other application or link to it.Rather,it simply starts up that piece of the other application when the need arises. For this to work,the system must be able to start an application process when any part of it isneeded,and instantiate the Java objects for that part.Therefore,unlike applications on most other systems,Android applications don't have a single entry point for everything in the application(nomain()function,for example).Rather,they have essential components that the system can instantiate and run as needed.There are four types of components:ActivitiesAn activity presents a visual user interface for one focused endeavor the user can undertake.For example,an activity might present a list of menu items users can choose from or it might display photographs along with their captions.A text messaging application might have one activity that shows a list of contacts to send messages to,a second activity to write the message to the chosen contact,and other activities to review old messages or change or change settings.Tough they work together to form a cohesive user interface,each activity is independent of the others.Each one is implemented as a subclass of the Activity base class.An application might consist of just one activity or,like the text messaging application just mentioned,it may contain several.What the activities are,and how many there are depends,of course,on the application and its design.Typically,one of the activities is marked as the first one that should be presented to the user when the application is launched.Moving2015届毕业设计外文文献原文及译文from one activity to another is accomplished by having the current activity start the next one.Each activity is given a default window to draw in.Typically,the window fills the screen,but it might be smaller than the screen andfloat on top of other windows.An activity can also make use ofadditional windows — for example,a pop-up dialog that calls for a user response in the midst of the activity,or a window that presentsusers with vital information when they select a particular item on-screen. The visual content of the window is provided by a hierarchy of views — objects derived from the base View class.Each view controls a particular rectangular space within the window.Parent views contain and organize the layout of their children.Leaf views(those at the bottom of the hierarchy)draw in the rectangles they control and respond to user actions directed at that space.Thus,views are where the activity's interaction with the user takes place. For example,a view might display a small image and initiate an action when the user taps thatimage.Android has a number of ready-made views that you can use —including buttons,text fields,scroll bars,menu items,check boxes,and more.ServicesA service doesn't have a visual user interface,but rather runs inthe background for an indefinite period of time.For example,a service might play background music as the user attends to other matters,or it might fetch data over the network or calculate something and provide the result to activities that need it.Each service extends the Service base class.A prime example is a media player songs from a play list.The player application would probably have one or more activities that allow theuser to choose songs and start playing them.However,the music playback itself would bot be handled by an activity because users will expect the music to keep the music going,the media player activity could start a service to run in the background.The system would then keep the music playback service running even after the activity that started it leaves the screen.It's possible to connect to (bind to)an ongoing service(and startthe service if it's not2015届毕业设计外文文献原文及译文already running).While connected,you can communicate with theservice through an interface that the service exposes.For the music service,this interface might allow users to pause,rewind,stop,andrestart the playback.Like activities and the other components,services run in the main thread of the application process.So that they won't block other components or the user interface,they often spawn another thread fortime-consuming tasks(like music playback).See Processes and Thread,later.Broadcast receiversA broadcast receiver is a component that does nothing but receiveand react to broadcast announcements.Many broadcasts originate in system code — for example,announcements thatthe timezone has changed,that the battery is low,that a picture has been taken,or that the user changed a language preference.Applications can also initiate broadcasts — for example,to letother applications know that some data has been downloaded to the device and is available for them to use.An application can have any number of broadcast receivers to respond to respond to respond to any announcements it considers important.All receivers extend the Broadcast Receiver base class.Broadcast receivers do not display a user interface.However,they may start an activity in response to the information they receive,or they may use the Notification Manager to alert the user.Notifications can get the user's attention in various ways — flashing the back light,vibrating the device,playing a sound,and so on,They typically place a persistent icon in the status bar,which users can open to get the message.Content providersA content provider makes a specific set of the application's data available to other applications.The data can be stored in the file system,in an database,or in any other manner that makes sense.The content provider extends the Content Provider base class to implement a standard set of methods that enable other applications to retrieve and store data of the type it2015届毕业设计外文文献原文及译文controls.However,applications do not call these methodsdirectly.Rather they use a Content Resolver object and call its methods instead.A Content Resolver can talk to any content provider;it cooperates with the provider to manage any inter process communication that's involved.See the separate Content Providers document for more information on using content providers.Whenever there's a request that should be handled by a particular component,Android makes sure that the application process of the component is running,starting it if necessary,and that an appropriate instance of the component is available,creating the instance if necessary.2015届毕业设计外文文献原文及译文The basic of description of android system对Android系统的基本描诉由开放手机联盟(Open Handset Alliance LED通过谷歌)开发的安卓系统是一个开源的手机系统被广泛看好，该系统提供了一个基本的操作系统，中间件的应用层，一个Java开发工具和应用系统集热器(系统应用收集)。