期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2017
卷号：6
期号：11
页码：1670-1683
出版社：Shri Pannalal Research Institute of Technolgy
摘要：Wireless Sensor Network (WSN) is a collection of sensor nodes with self-arranged and organizational structure-less wireless networks. WSN performs better network communication to attain stable and consistent performance. Additionally, consistent wireless network communication is provided between the sensor nodes. But, it attains maximum data loss and delay during data transmission. In order to overcome above issues, K Nearest Neighbour-based Logistic Regression (KNN-LR) framework is developed. Stability-aware Logistic Regression model and Link Quality-based Consistent are two different phases for wireless network communication. In the beginning, Stability-aware Logistic Regression model is designed in WSN. Here, local and global pivotal values are measured with neighbors’ to make the decision regarding existence or non-existence of a sensor node. Additionally, machine learning logistic regression algorithm is utilized to achieve stable network with enhanced network lifetime and throughput. After that, consistent wireless network communication is attained by using the basis of link quality in WSN. The link quality enhances the consistency of network by selecting optimal route path during data transmission. As a result, data loss and end to end delay is minimized with efficient route path. The simulation is carried out to analyze the performance of proposed KNN-LR framework with the parameters such as throughput, network lifetime, data loss and end to end delay.
关键词：Wireless sensor networks; Logistic Regression model; Fusion Center; data Packet Acceptance Ratio
