# 电子工程代写|嵌入式网络系统代写Embedded Networked Systems代考|EECS262

## 电子工程代写|嵌入式网络系统代写Embedded Networked Systems代考|Constraint Based Methods

Kogekar et al. [8] presented an approach for constraint-guided software reconfiguration in sensor networks. They have implemented a system which relies on monitoring the system requirements expressed as formal constraints. Those constraints drive the reconfiguration process that takes place in a base station that can communicate to all the sensor nodes. They subsequently demonstrate their approach using simulation results from a simple one-dimensional tracking problem.

In a related article from Eames [9], a method called DesertFD (Design Space ExploRation Tool using Finite Domain constraints) is demonstrated for design space exploration based on constraint satisfaction problems. The authors specifically mention that the tool was initially produced for design-time configuration, but was later embedded in a runtime reconfiguration framework in order to do on-the-fly optimization. They combine design-time derived metadata from profiles, benchmarks and expert knowledge, with runtime information from monitoring instruments. The process is then periodically called, and evaluates and prunes the set of system configurations with the goal of selecting the proper configuration to deploy.

An example of a reconfiguration model based on constraint satisfaction programming is given by Syrjänen [10] who examined of software configuration management. A declarative rule-based formal language is proposed for representing configuration knowledge. A case study is shown for managing Debian GNU/Linux packages. The author states that finding a stable in any program in normal logic is NP-hard. In order to find a solution for the configuration problem a method is used to find stable models semantics for logic programs developed by Simons [11]. This method called smodels is an algorithm for solving a constrained design problem, and as the authors denote, it does have some overhead and even possibly more than comparable methods. However this overhead provides the designer with a more powerful language. Consequently, problems that can be more compactly represented by logic programs can be more quickly solved with smodels than with a satisfiability checker.

## 电子工程代写|嵌入式网络系统代写Embedded Networked Systems代考|Logic Based Methods

A logic based reconfiguration method will rely on some type of logical reasoner to find a configuration that matches requirements in runtime. One example a first-order logic interpreter is Prolog [12]. Prolog uses backtracking methods and Selective Linear Definite clause resolution to find solutions $[13,14]$. This way of solving logic statements is proven to be both sound and complete [15]. Backtracking is the process of finding a solution by searching for partial solutions. For each candidate of a partial solution the validity is checked, if at one point no solutions exist for the last partial solution, a different one is searched for the previous one. This is analogous for finding solutions in a search tree, and looking depth-first for a solution that satisfies all sub statements.

Selective Linear Definite clause resolution is a method for checking statements by denoting them as disjunctive literals, and refuting all negated literals via the before mentioned backtracking method. SLD-resolution is very efficient both in terms of time and space. However, similar sub goals may be derived multiple times if the query contains recursive calls. Moreover, SLD-resolution is not guaranteed to always terminate [16].

In the SOSE project [17] a framework is described for reconfiguration at the heart of which lies a Prolog based reasoner. Using a relatively simple set of rules the authors can describe the reconfiguration strategies. The method is demonstrated for a greenhouse use case in which an optimal configuration needs to be employed for estimating the temperature distribution. Whereas in this case the goal of the system was purely state estimation, the framework was set up in a generic manner, so different applications are possible as well.

Isermann [18] already used a fuzzy logic controller for automatic supervision and fault diagnosis. He concluded that fuzzy logic provides a systematic framework to process vague variables and/or vague knowledge. As an example of this type of framework, Aubrun et al. [19] used fuzzy logic for monitoring and reconfiguring a power plant. They concluded that fuzzy logic provides better results for failure detection in terms of the robustness of the models comparing to classical methods. A freely available fuzzy logics reasoner called FuzzyDL [20] uses algorithms that comhine a tableaux algorithm and that of a MII .P-solver. It extends normal description logic with concepts of real values, integers, strings and definitions of fuzzy membership functions. It also allows for concepts in the knowledge base to be constrained using syntax which is comparable to constraint description. In a related article, the same authors propose a method for finding solutions using fuzzy logics from models described in OWL [21].

## 电子工程代写|嵌入式网络系统代写Embedded Networked Systems代考|Constraint Based Methods

Kogekar 等人。[8] 提出了一种在传感器网络中进行约束引导的软件重新配置的方法。他们实施了一个系统，该系统依赖于监控以形式约束表示的系统要求。这些约束推动了在可以与所有传感器节点通信的基站中进行的重新配置过程。随后，他们使用来自简单一维跟踪问题的模拟结果展示了他们的方法。

Syrjänen [10] 给出了一个基于约束满足编程的重新配置模型的例子，他研究了软件配置管理。提出了一种基于声明性规则的形式语言来表示配置知识。展示了管理 Debian GNU/Linux 软件包的案例研究。作者指出，在正常逻辑中的任何程序中找到一个稳定是 NP 难的。为了找到配置问题的解决方案，一种方法被用来为 Simons [11] 开发的逻辑程序找到稳定的模型语义。这种称为 smodels 的方法是一种用于解决受限设计问题的算法，正如作者所指出的，它确实有一些开销，甚至可能比同类方法更多。然而，这种开销为设计人员提供了更强大的语言。最后，

## 电子工程代写|嵌入式网络系统代写Embedded Networked Systems代考|Logic Based Methods

Isermann [18] 已经使用模糊逻辑控制器进行自动监督和故障诊断。他得出结论，模糊逻辑为处理模糊变量和/或模糊知识提供了一个系统框架。作为这类框架的一个例子，Aubrun 等人。[19] 使用模糊逻辑来监控和重新配置发电厂。他们得出结论，与经典方法相比，模糊逻辑在模型的鲁棒性方面为故障检测提供了更好的结果。一个免费提供的模糊逻辑推理器，称为 FuzzyDL [20]，使用结合了表格算法和 MII .P 求解器的算法。它用实数值、整数、字符串的概念和模糊隶属函数的定义扩展了正常的描述逻辑。它还允许使用类似于约束描述的语法来约束知识库中的概念。在一篇相关文章中，同一作者提出了一种使用 OWL [21] 中描述的模型中的模糊逻辑来寻找解决方案的方法。

myassignments-help数学代考价格说明

1、客户需提供物理代考的网址，相关账户，以及课程名称，Textbook等相关资料~客服会根据作业数量和持续时间给您定价~使收费透明，让您清楚的知道您的钱花在什么地方。

2、数学代写一般每篇报价约为600—1000rmb，费用根据持续时间、周作业量、成绩要求有所浮动(持续时间越长约便宜、周作业量越多约贵、成绩要求越高越贵)，报价后价格觉得合适，可以先付一周的款，我们帮你试做，满意后再继续，遇到Fail全额退款。

3、myassignments-help公司所有MATH作业代写服务支持付半款，全款，周付款，周付款一方面方便大家查阅自己的分数，一方面也方便大家资金周转，注意:每周固定周一时先预付下周的定金，不付定金不予继续做。物理代写一次性付清打9.5折。

Math作业代写、数学代写常见问题

myassignments-help擅长领域包含但不是全部: