Proximity mobile services: Neural network based connection admission controller and other issues

摘要

Proximity aware mobile service is a wireless application for mobile users to interact with services offered by service providers present in the user's environment. Many issues are still prevalent in this area which includes context modeling, sensors integration, proximity resolution and service management. One important aspect in service management is congestion control which plays a key part in providing proximity aware services in a wireless networking environment. This paper investigates the feasibility of neural network based connection admission controllers by comparing simulated Back-Propagation and Learning Vector Quantization Neural Networks. It is found that the Back-propagation Neural Network has higher performance.