دانلود ترجمه مقاله روش تکاملی در شبکه های حسگر بی سیم برای مشکل خوشه بندی متعادل بار – مجله الزویر

• فهرست مطالب:

 چکیده
۱ مقدمه
۲ کار مرتبط
۳ مدل انرژی
۴ مدل WSN و فرمولاسیون مشکل
۵ مروری بر الگوریتم ژنتیک
۶ الگوریتم پیشنهادی
۶ ۱ نمایش کروموزوم
۶ ۲ جمعیت اولیه
۶ ۳ تابع تناسب
۶ ۴ انتخاب
۶ ۵ Crossover
۶ ۶ جهش
۷ نتایج تجربی

• بخشی از ترجمه:

در این مقاله، ما یک  GA مبتنی بر الگوریتم خوشه بندی متعادل کننده بار برای WSN ارائه داده ایم.این الگوریتم شرح داده شده است با نمایش کروموزومی مناسب، تولید جمعیت اولیه ، فرایند انتخاب، به دنبال crossover وعملیات جهش شرح داده شده است. نتایج تجربی نشان داده است که عملکرد الگوریتم بهتر از GA مبتنی بر الگوریتم خوشه بندی، GA ساده، روش  تکاملی دیفرانسیلی، بار خوشه متوازن (LBC) و  الگوریتم کمترین فاصله خوشه (LDC) از نظر تعادل بار دروازه ها  برای بار برابر گره های سنسور و همچنین بار نابرابراست. مشاهده شده است که الگوریتم پیشنهادی عملکرد بهتری از نظر مصرف انرژی، تعداد گره سنسور فعال، نرخ همگرایی و زمان اجرا فراهم می کند.

• بخشی از مقاله انگلیسی:

. Introduction Wireless sensor networks (WSNs) have attracted many researchers for their potential uses in various fields including disaster warning systems, environment monitoring, health care, safety, surveillance, intruder detection and so on [1,2]. A WSN is composed of a large number of tiny sensor nodes, which are randomly or manually deployed in a target area. The sensor nodes consist of sensing, data processing, and communicating components along with a power unit. The sensor nodes sense the target area to collect local information, process them and send it to a remote base station called sink. The sink is connected to the Internet for the public notification of the phenomena. The main bottleneck of the WSNs is the limited and irreplaceable power sources of the sensor nodes as they are operated on small batteries. Moreover, in many applications, it is almost impossible to replace the sensor nodes when their energy is exhausted. Therefore, energy consumption for the sensor nodes is the most challenging issue for the long run operation of WSNs [3–۵]. Clustering is one of the most efficient techniques, which has been well researched for energy saving WSN. In a cluster based architecture (refer Fig. 1), the sensor nodes are grouped into distinct clusters with a leader, known as cluster head (CH) for each. Each sensor node belongs to only one cluster. The CHs collect and process the local data from their member sensor nodes and send it to the sink directly or via other CHs. A cluster based WSN has many advantages [6] as follows: (1) It can reduce energy consumption significantly as only one representative (i.e., CH) per cluster needs to be involved in data aggregation and routing process. (2) It can considerably conserve communication bandwidth as the sensor nodes need to communicate with their CHs only and can avoid exchange of redundant messages among them. (3) The clusters can be more easily managed as they can localize the route set up and require small routing tables for the sensor nodes. This in turn improves the scalability of the network significantly. However, in a cluster based WSN, CHs bear some extra work load contributed by their member sensor nodes as follows: (1) CHs communicate with all the sensor nodes within their cluster; (2) they perform data fusion to discard redundant and uncorrelated data sent by their member sensor nodes and finally (3) they send the processed data to the sink. Moreover, in many WSNs the CHs are usually selected amongst the normal sensor nodes which can die quickly owing to this extra work load. In this context, many researchers [7–۱۱] have proposed the use of some special nodes called gateways or relay nodes, which are provisioned with extra energy. These gateways are treated as the cluster heads and responsible for the same functionality of the CHs. But the gateways are also battery operated and hence power constrained. Life time of the gateways is very crucial for the long run operation of the network. Therefore, improper cluster formation may cause some CHs overloaded. Such overload may increase latency in communication, consumes high energy of the CH and degrade the overall performance of the WSN. Therefore, load balancing of the CHs is the most important issue for clustering sensor nodes. Particularly, this is a pressing issue when the sensor nodes are not distributed uniformly. It is noteworthy that for a WSN, with n sensor nodes and m gateways, the number of possible clusters is mn . This implies that the computational complexity of finding the optimal load balanced clustering for a large WSN seems to be very high by a brute force approach. In fact, load balanced clustering with unequal load of sensor nodes is a NP-hard problem [9]. Genetic Algorithm (GA) is one of the most suitable heuristics that can be applied for efficient load balanced clustering from such a large solution space. In this paper, we propose a new GA based clustering algorithm to solve the above load balancing problem. The algorithm forms clusters in such way that the maximum load of each gateway is minimized. The proposed algorithm differs from the traditional GA with the following respects:
In the phase of initial population generation, we restrict the generation of initial population by considering the connectivity between the sensor nodes and their CHs.