APPLICATION OF NEURAL NETWORK TECHNIQUS FOR LOCATION PREDICTION IN MOBILE NETWORKING

APPLICATION OF NEURAL NETWORK TECHNIQUS FOR LOCATION PREDICTION IN MOBILE NETWORKING

Abstract:

Wireless network resource use depends in large part on the mobility of network users. The ability to predict this mobility at least in part enables the network to anticipate resource use in the future and take precautionary measures if necessary. This work presents a neural network prediction system that is able to capture some of the patterns exhibited by users moving in a wireless environment and can then predict the future behavior of these users. These predictions can then be used in a multitude of ways to ensure proper and predictable resource use. In this paper, for location prediction, the performance of a multi-layer perceptron (MLP) network has been studied first. A paging technique has been proposed based on this predicted location. Next, a hybrid network composed of a self-organizing feature map (SOFM) network followed by a number of MLP networks has been employed for prediction. Simulation studies show that the latter performs better for location management. This approach is free from all unrealistic assumptions about the movement of the users. It is applicable to any arbitrary cell architecture. It attempts to reduce the total location management cost and paging delay.


for more info visit.
http://www.enjineer.com/forum