A Novel Attention Control Modeling Method for Sensor Selection Based on Fuzzy Neural Network Learning
Student Cse, Nisha Phogat · 2014
A sensor network is defined with large number of energy nodes. As of the adhoc network, the nodes are distributed over the network randomly. Because of this, unequal distribution of nodes over the network is occurred. This unequal distribution also raises many communication problems over the network. To resolve this problem, it is required to monitor the critical nodes or area over the network so that effective timely decision can be taken. The identification of these critical nodes over the network is called attention control. In this present work, an intelligent parametric approach is defined to identify the critical nodes over the network so that effective node monitoring will be done. The presented work will be divided in two three main stages. In first stage, the identification of attention nodes will be done. The identification will be done based on multiple vectors such as energy, connectivity, load etc. To perform this identification, a at first fuzzy rules will be derived to obtain the parametric values in nominal form and later on the classification algorithm such as probabilistic neural network will be applied to identify the attention nodes. Once the attention nodes will be identified, the next work is about to identify the agent nodes that will monitor these critical nodes over the network. Here, an intelligent approach will be applied so that minimum nodes will be taken to monitor all the critical nodes. Once the agent nodes will be defined, the next work is to identify the alternate nodes to the attention nodes that can replace the nodes so that effective results will be drawn from the network. The work is about to improve the network again such problems and to improve the network life and communication. The presented work will be implemented in matlab environment.