Abstract: A grey-relational based algorithm for wireless channel is put forward in Vehicle Ad-hoc Networks (VANETs). First, by examining the relationship between channel load and its influence factors of time series, a grey correlation analysis method is proposed to select the main factors that affect the channel load. Then, according to the multiple relationship model, the channel load prediction algorithm based on Kalman filter recursion is developed. To verify the effectiveness of the channel load prediction algorithm, a traffic survey method is employed to conduct a short-term channel load forecasting experiment. Traffic data on a segment of Changchun urban road is collected using a floating car. The experimental results show that the proposed algorithm is effective with higher prediction accuracy and good performance in adaptation and robustness.

Key words: engineering of communications and transportation system, vehicular ad-hoc network(VANET), channel load, Kalman filter, grey relation