期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2018
卷号：6
期号：10
页码：8179-8187
DOI：10.15680/IJIRCCE.2018. 0610038
出版社：S&S Publications
摘要：Wireless sensor networks are a type of wireless network that contains wireless sensors and are used in
many fields. Sensors in this type of network are small, low cost, consume a small amount of energy and able to
communicate with each other over a short distance. These networks are composed of a large number of sensors, which
are usually deployed in random different places. As these networks may be fixed in a dangerous area or difficult to
reach, such as a battlefield or around a volcanic eruption.
The process of determining the real geographical location of these sensors is called localization and the location is very
important as in most applications in this field, the data received without knowing its location is useless. There are
several methods used to determine the geographical location of the sensor from which the data was received, including
the global positioning system (GPS). However, in this type of networks, sensors are deployed in large numbers, which
requires providing these large numbers of sensors with the positioning system for each sensor. Adding this technique to
the sensors increases their physical cost. In addition, the GPS system needs a lot of energy when used. This reduces the
life of the default sensor battery, thereby reducing the accuracy of the GPS system and it becomes difficult to obtain
information about the location in the closed area.
Because the sensors' location cannot be exactly determined, they are usually distributed randomly (also new sensors
may be added to the network or the energy of some sensors may depleted causing their exit from the network). This
paper focuses on the use of artificial neural networks (ANNs) to determine the geographical location of wireless sensor
networks depending only on three known location nodes (anchor nodes) located in different locations. The results are
compared in order to obtain the coordinates of all wireless sensors to be deployed or to be added in future, taking into
consideration the use of reduced power needed for the purpose of sending and receiving the information on the
location.
