Modeling and Analysis for Vertical Handoff Based on the Decision Tree in a Heterogeneous Vehicle Network

In this paper, we present an analytical framework to evaluate the accuracy of a vertical handoff algorithm based on a decision tree in a heterogeneous vehicle network. To quantify the effects of errors in the decisions made by this analytical framework, we consider two key design features: 1) the probability of a false alarm and 2) the probability of missing an alarm. Then, we propose using the Kalman filtering algorithm to obtain more accurate parameters. The probability threshold interval model is designed to characterize the errors in vertical handoff decisions made under imprecise information conditions. The accuracy of vertical handoff decisions based on a decision tree under Gaussian and linear models is analyzed. These analytical results are applied to evaluate the performance of the vertical handoff model based on the decision tree. Finally, we propose a robust vertical handoff algorithm based on the decision tree to improve the decision accuracy. A theoretical analysis shows that the proposed algorithm improves handoff decision accuracy. In addition, we conducted comprehensive simulations to validate the theoretical results. The simulation results demonstrate that our algorithm substantially improves the handoff performance in a heterogeneous vehicle network.