• فهرست مطالب:

چکیده
مقدمه
بهینه سازی ازدحام ذرات دودویی اصلاح شده، MBPSO
به روزرسانی سرعت و موقعیت برای MBPSO
عملیات جهش برای MBPSO
فرمول نویسی مسئله
الگوریتم انتخاب گره ها مبتنی بر MBPSO
آزمایش شبیه سازی و آنالیز نتیجه حاصله
خلاصه

• بخشی از ترجمه:

برای به حداکثر رساندن اطلاعات اندازه گیری برای تابع هدف، الگوریتم انتخاب گره ها مبتنی بر MBPSO پیشنهاد گردید. برطبق مشخصات ذرات و متغیرهای تصمیم، روش کدگذاری جدیدی برای ذرات طراحی و از تابع انتقال V شکل برای بهبود سرعت همگرایی استفاده گردید. به علاوه، از عملیات جهش برای اجتناب از بهینه محلی، و از وزن اینرسی کاهشی غیر خطی برای تعادل بین توانایی جستجوی کلی (سراسری) و محلی استفاده گردید. نتایج شبیه سازی نشان می دهد که مکانیزم انتخاب گره پیشنهادی در مدت زمان نسبتاً کوتاهی قادر به یافتن راه حلی با کیفیت بالا می باشد. 

• بخشی از مقاله انگلیسی:

Introduce Wireless sensor networks is a network system that integrated monitoring, control and wireless communication. It has broad application, such as military reconnaissance, environmental monitoring, target tracking, space exploration and disaster emergency. Especially, the network characteristics of rapid expansion, distributed information processing and anti-destroying ability, for maneuvering target tracking has created one of the most representative applications of WSN[1]. In recent years, many methods and strategies are proposed for the nodes selection problem. A mixed integer programming model established in [2], which using the branch and bound algorithm to solve the nodes selection problem. [3]took the diversity into account, partitioning the region based on the target prediction location, and selecting closest sensor to the target in each sector, then calculate the object function for each set to find the best active set. A global node selection algorithm was proposed in [4], which derives the boundaries of the target tracking performance, using a greedy strategy that “add one node at a time” and “simplify” for node selection. All the above are only for a single target, [5] considered the distributed node selection for multi-target tracking problem. The initial 0-1 integer programming problem is relaxed into a convex optimization problem, using gradient projection search as the solver. In this paper, we consider the problem of nodes selection in multi-target tracking to maximize the measurement information. A modified discrete binary particle swarm optimization (MBPSO) algorithm with V-shaped transfer function is proposed as the solver. Finally, through multi-set of simulation experiment proves the feasibility and effectiveness of the algorithm. Modified binary particle swarm optimization, MBPSO In basic BPSO, along with the iterations, the random walk slows down the convergence speed of the algorithm. In addition, it is easy to fall into local optimum [6]. Therefore, a modified version of BPSO is proposed.