دانلود مقاله ترجمه شده استقرار حسگر براساس PSO باینری گسسته اصلاح شده در شبکه های حسگر بی سیم – مجله IEEE

مشخصات مقاله انگلیسی (PDF)
سال انتشار	۲۰۱۰
تعداد صفحات مقاله انگلیسی	۵ صفحه با فرمت pdf
رشته های مرتبط با این مقاله	مهندسی کامپیوتر و مهندسی فناوری اطلاعات
گرایش های مرتبط با این مفاله	مهندسی الگوریتم ها و محاسبات، نرم افزار، معماری سیستم های کامپیوتری، هوش مصنوعی و شبکه های کامپیوتری
مجله	کنفرانس بین المللی شبکه های ارتباطی و هوش محاسباتی(International Conference on Computational Intelligence and Communication Networks)
دانشگاه	دانشکده مهندسی کامپیوتر و فناوری اطلاعات، دانشگاه آزاد اسلامی قزوین
کلمات کلیدی	PSO دودویی اصلاح شده، شبکه حسگر توزیع شده، استقرار حسگر
رفرنس	دارد
لینک مقاله در سایت مرجع	لینک این مقاله در سایت IEEE
نشریه آی تریپل ای

• فهرست مطالب:

 چکیده
۱ مقدمه
۲ تعریف مسئله
۳ الگوریتم پیشنهادی
۴ نتایج شبیه سازی
۵ نتیجه گیری

• بخشی از ترجمه:

مقاله حاضر مسئله استقرار حسگر برای مکان یابی اهداف تحت محدودیت ها را بررسی می نماید (پوشش کامل شبکه حسگر با حداقل هزینه). اولاً این مسئله NP-Complete را به صورت مدل بهینه سازی ترکیبی تعریف کرده، سپس الگوریتم PSO دودویی اصلاح شده برای حل مسئله توسعه یافت. نتایج بدست آمده نشان داد در مقایسه با الگوریتم تبرید شبیه سازی شده، الگوریتم پیشنهادی توانایی تشخیص موثرتر راه حل بهینه سازی در زمان و هزینه های محدود را دارد که در این وضعیت استقرار حسگرها برای افزایش پوشش روی فیلد حسگر، شانس فرار از حد بهینه محلی را نیز افزایش می دهد. به علاوه، الگوریتم پیشنهادی مفیدتر، مقیاس پذیر و بادوام می باشد. 

• بخشی از مقاله انگلیسی:

Introduction In the distributed sensor networks, the issue of sensor placement is of paramount importance in researches. A sensor network can arrange in two ways, one as a random placement and the second as a grid-based placement. Once the surrounding is unknown the random placement is the only option and the sensors may disintegrated everywhere but when the features of the network were known before, then the sensor placement could be done with great scrutiny and we could guarantee the quality of providing services along with satisfying the limitations. The strategy of sensor placement depends on the application of the distributed sensor network (DNS). In this article the focus is on the gird-based placement. And we applied the modified binary PSO algorithm for solving these NP-complete problems [6]. Considering the existence of many networks with high velocity and computational capabilities of these sensor networks, we can say that they have different applications for example in aviation, military, medical, robot, air forecasting, security and anti terrorism applications and also we can use them in very important infrastructures like power plants ,environmental and natural resource monitoring, and military applications like communication systems, commanding, reconnaissance patrols, looking –out etc [7][8]. In [9] and [10], they present a resource-bounded optimization framework for sensor resource management under the constraints of sufficient grid coverage of the sensor field. In [11], they formulate the sensor placement problem in terms of cost minimization under coverage constraints. In [12] Node placement in heterogeneous WSN is formulated using a generalized node placement optimization problem to minimize the network cost with lifetime constraint, and connectivity. In [13] they formulate and solve the sensor placement problem for efficient target localization in a sensor network, they develop a mathematical framework for the localization of the missile using multiple sensors based on Cramer-Rao Lower Bound (CRLB) analysis. In [14] they present the practical problem of optimally placing the multiple PTZ cameras to ensure maximum coverage of user defined priority areas with optimum values of parameters like pan, tilt, zoom and the locations of the cameras. Moreover in [15] a heuristic algorithm is proposed based on Simulation Annealing Algorithm to solve this problem considering the coverage and cost limitations. The rest of the paper is organized as follows: in section 2, we state the sensor placement problem and then present the mathematical model in section 3. Section 4, proposes an algorithm. The performance evaluations are in Section 5. Section 6 concludes the paper. 2- Defining the problem The sensor network based on gird-based could be considered as a two or three dimensional network [9]. A set of sensors are settled on different points of gird points in order to monitor the sensor field. In this section we defined a power vector for each point of the field to show whether these sensors could cover that point on the field or not, for which the number of components are as many as the number of sensors available. Now if the Euclidean distance between each grid point and the corresponding sensor is less than the coverage radius of the sensor (d<r), so the coverage is assumed to be full (1), and it becomes a parallel component of that sensor on the power vector, Otherwise, the coverage is ineffective and the parallel components equaled to (0). If each point on the gird point in a sensor field can be covered by at least one sensor so that the sum of the vector components of that field equals to one, the field is called completely covered. In Fig.1, a complete and discriminated sensor field of 4*4 with radius =1 is illustrated, that a target can be detected at any place in the field. In figure 1, for example the power vector for point 7 equals to (0, 1, 0, 0) which is calculated based on the sensors of 2, 8, 9 and 15. When a target appears at the grid point 7, the backend will receive reports from sensor 8 [15]. In a completely covered sensor field, when each grid point is identified by a unique power vector, the sensor field is said to be completely discriminated, as shown in Fig.1.