No abstract

In most applications of wireless sensor networks, the location of every node within the network is needed for network function. Hence, the plan of node location systems for sensor networks ought to be based on sensor network utilization scenarios so these components make node’s localization in wireless sensor networks a challenging problem. Node’s localization in wireless sensor networks can be partitioned into two common categories of Range-free and Range-based strategies. Distance or Range-based positioning strategies precisely measure distance, while Range-free algorithms utilize communication information between nodes. Range-based procedures generally have two major downsides. First, distance information is easily affected by fading, multipath fading, noise, and environmental changes. Secondly, additional range-finding equipment is usually needed, which increases energy consumption and increases overall cost. One of the advantages of Range-free methods is that they can be implemented at a low cost and its other advantage is its strength because the interconnection information between nodes is not easily affected by environmental conditions, but its major disadvantage is the low accuracy of these methods as compared with Range-based methods. This study can be used to design and implement a WSN localization algorithm to detect abnormalities in a harsh environment such as the Nuclear Power Plant. Given that GPS performs well in outdoor environments and the equipped-GPS sensors would not be working properly in indoor environments such as inside a reactor, the WSN localization algorithm used in such applications must have at least some dependence on GPS technology. We proposed a two-stage localizing algorithm with acceptable accuracy and taking advantage of both Range-free and Range-based methods. In this way, first using the Range-free localization algorithms, the nodes with unknown locations (normal nodes) and with at least three neighboring anchors (which have a known position) will estimate their location. Then, these nodes are considered as anchor nodes for their neighbors. This approach can increase the accuracy of localizing the normal nodes and significantly reduce the number of anchor nodes (equipped-GPS sensors) needed for localization. The results show that the proposed localization algorithm's error as a tool for fault diagnostics is remarkably decreased.