1AP / MCA, EBET Group of Institutions, Kangayam, 638108, India
2Dept. of CSE, Bannariamman Institute of Technology, Sathy, Tamilnadu,India,
.Abstract: WLAN infrastructure planning for maintaining service quality gains importance due to numerous wireless devices getting connected to the internet. To maintain desired service quality users movement pattern should be known. Mobility prediction involves locating mobile device's next access point when it moves through a wireless network. Hidden Markov models and Bayesian approach were suggested to predict next hop This study proposes a new method for feature extraction and suggests a hidden Genetic Algorithm layer-GA-SOFM based new neural network classifier. The hypothesis is evaluated through the use of a month long syslog data of Dartmouth college mobility traces available online. This extracts mobility features and uses them to find the proposed model’s classification accuracy.
[VELMURUGAN, L, THANGARAJ, P. Mobility Prediction using Hidden Genetic Layer Based Neural Network. Life Sci J 2013;10(4s):549-553] (ISSN: 1097-8135). http://www.lifesciencesite.com. 83
Keywords: Mobility prediction, Neural Network, Naïve Bayes, Data Mining, Wireless Infrastructure, Quality Of Service, Location Based Services, User Mobility Pattern (UMP).